Category: 8. Health

This year may prove to be an important year for advances in gene editing and gene therapies.

But what does that really mean, and how have we got to this point?

In what is claimed as a world first, doctors in the United States recently treated a baby with a rare genetic disease using a highly specific and personalised gene-editing technique.

What did this involve and how is the baby doing?

Baby boy KJ was born with a rare metabolic disease known as CPSI deficiency. This prevented his body from getting rid of ammonia during the metabolism of protein. And while he was put on a highly restricted protein diet and given drugs to remove protein from his blood, he remained at high risk of brain damage or even death.

Following the early diagnosis of KJ’s condition, gene-editing researchers at the University of Philadelphia worked with colleagues throughout the US to quickly develop a personalised gene-editing infusion to correct a genetic variant that led to the disorder.

KJ responded well to the three-dose regimen and is now reaching his developmental milestones. “Ultimately, we hope this has set a precedent where we have firmly entered a world of genetic cures on demand,” said Fyodor Urnov, scientific director of the Innovative Genomic Institute in the University of California, Berkeley, a member of the treatment team.

Have similar gene-editing techniques been used on patients in Ireland too and how do they differ from gene therapy?

The gene-editing platform known as Crispr is widely used by researchers in scientific laboratories in their search for new therapies for cancer and other diseases. Essentially, Crispr – whose inventors won the Nobel Prize in chemistry in 2020 – is a cheap and efficient way of finding and altering specific pieces of DNA within cells.

Gene therapy is slightly different in that it introduces new DNA materials into cells to replace or correct a gene or inactivate a target gene in the treatment of a specific disease.

Earlier this year, ophthalmologic surgeon Dr Max Treacy treated 20-year-old Maros Tomko, who had severe visual loss since birth, with gene therapy at the Royal Victoria Eye and Ear Hospital in Dublin. The once-off treatment replaced the faulty gene RPE65 with a healthy copy. This was done via a viral vector which carried the healthy copy of the gene into the cells of the eye.

“Within four weeks, the vision in his eyes had doubled. From not being able to see any letters, he could read the first and second lines. His visual field [the total area seen] also doubled in size,” explains ophthalmologist Dr Emma Duignan. Tomko adds that he is very grateful to the doctors for this opportunity to have surgery. “I can see people’s faces now and I can read the numbers on my bank card for the first time. It’s only three months since the surgery so it will take longer to get better,” he says.

A 31-year-old Sligo man with a similar congenital blindness got his sight back after being treated with ocular gene therapy at the Mater hospital in Dublin in 2024.

Gene therapies, however, remain very expensive and the Royal Victoria Eye and Ear Hospital treatment, which was paid for by the Health Service Executive, cost close to €800,000.

Dr Duignan says it took years for the team to get the funding in place, but such treatments now mean that Dublin will have two potential centres for gene therapy clinical trials. “Clinical trials of gene therapies for inherited eye conditions are desperately needed for the treatment of an estimated 5,000 patients on the island of Ireland affected by these diseases,” says Dr Duignan. Drugs used in clinical trials are made available free of charge.

Manipulating human DNA – the genetic material responsible for life – is like the stuff of science fiction. How is it even possible?

Research into gene therapy goes back to the 1940s and 1950s. The first studies began to recognise that DNA was a transforming substance capable of changing the living characteristics of individuals through a biochemical process. Later, studies identified that human stem cells (mainly found in the bone marrow) could be genetically modified to carry therapeutic DNA which could then differentiate into various cell types to correct genetic defects.

In the 1980s, researchers at Boston Children’s Hospital published a paper to show a virus could be used to insert genes into blood-forming stem cells. These so-called viral vectors later became an established way to deliver specific genetic material to human cells. In the meantime, the Human Genome Project was completed in 2003, paving the way for gene therapy to become a reality for multiple diseases, especially those caused by mutations in a single gene.

What specific diseases are we talking about?

The painful and life-threatening sickle cell disease, a rare genetic immune deficiency disorder called chronic granulomatous disease, and the rare childhood disease adrenoleukodystrophy have all been treated with gene therapy. Genetic diseases including cystic fibrosis, muscular dystrophy, sickle-cell anaemia and haemophilia could potentially be treated with gene therapy.

But are there risks to making such precise changes to the basic building blocks of life?

Yes, there are huge risks, which is why it has taken so long for gene therapy to reach clinical settings. In 1990, four-year-old Ashanthi de Silva became the first patient to be successfully treated with gene therapy. She was given a healthy adenosine deaminase (ADA) enzyme to cure the severe immunodeficiency disorder caused by the absence of ADA. Although she continues to take a drug to keep her condition under control, she leads an active life to this day.

[ Delving into the power of DNA for patientsOpens in new window ]

Throughout the 1990s, European researchers focused on other forms of severe immunodeficiency disorders, reporting the first cures in 2000. However, some years later, five of the 20 treated children developed cancer. The viral vector which had delivered the gene to their T cells (immune cells in the body) had also activated an oncogene, triggering leukaemia. Also in the US, an 18-year-old boy died after receiving gene therapy for a rare metabolic disorder.

These incidents delayed research into gene therapy for almost 10 years. But in the early 2010s, scientists developed better viral vectors that could more precisely target expressions of genes in specific cell types which don’t go astray in the body and don’t trigger an immune response. This spurred further developments in genetic therapies and several gene therapy drugs were approved for use.

The use of messenger RNAs (mRNAs) – as was used in Covid-19 vaccines – also represents a form of gene therapy. In this case, the mRNA vaccine introduces information that cells then use to make the coronavirus spike protein, which then stimulates the person’s immune system to develop antibodies to the virus.

What are the next steps for gene editing and gene therapy?

There are now hundreds of active gene therapy studies around the world and more than a dozen gene therapy drugs on the market, according to researchers at Boston Children’s Hospital.

A more finely tuned approach called base-editing, which uses Crispr technology to chemically change one “letter” of a gene’s code at a time, is deemed to be the next technological advance in genetic therapies. The small changes of base-editing can correct a “spelling error” mutation, silence a disease-causing gene or help activate a specific gene. However, while this approach hasn’t yet been tested in clinical trials, places such as Boston Children’s Hospital have several base-editing projects under way.

Dr Duignan adds that new so-called gene agnostic therapies, which will treat diseases caused by different gene mutations without having to develop a specific infusion for each mutation, represent the next frontier in gene therapy.

Picture this: you’re in the middle of a Zoom marathon or drowning in tied-up emails, your brain begging for five minutes of calm. You open the fridge, pull out a chilled Diet Coke, crack it open, feel the fizz on your tongue – and just like that, you’ve had a “fridge cigarette.”In the midst of Zoom fatigue, endless deadlines, and life in fast-forward, a new ritual has emerged: the “fridge cigarette.” Not a cigarette at all, but that cold crack-and-sip moment when you crack open a Diet Coke straight from the fridge. No smoke, no guilt, just that crisp ritual of pop you didn’t know you needed.TikTok’s Gen Z has turned this fizzy moment into a cultural wink – a mini smoke break with zero nicotine and full aesthetic energy. ‘Tis Gen Z’s latest obsession – a tongue-in-cheek break that’s half nostalgia, half self-care, and fully TikTok-viral. It’s not just fizzy drink culture – it’s a symbolic pause in a fast-paced digital life.

But just the part of omitting nicotine – does it help health-wise at all?Let’s unpack.

What’s a Fridge Cigarette?

The term “fridge cigarette” originated on TikTok, popularized by creator @reallyrachelreno, whose June video comparing cracking open a Diet Coke to lighting a cigarette went viral with near about 4M views. What she captured was a moment: the hiss, the fizz, the pause – and that psychological hit. It’s not about substances; it’s emotional.

On TikTok, fridge-cig posts feature ASMR-worthy fizz, minimalist setups, moody lighting, and captions like “time for a crispy ciggy.” One user described it as “main‑character energy” – a daily starring moment in everyday life. The ritual resonates because it blends irony, self-care, and aesthetic performance in a few crisp seconds.Gen Z reveres this ritual as a purposeful break from digital overload, recasting a smoke break into something healthier.. well, healthier-ish.

Why Diet Coke?

Diet Coke provides the quintessential chill experience: fizzy, caffeinated, calorie-free, and effortlessly cool. Users on TikTok have even playfully mapped Coke variants to cig brands – Diet Coke as Parliaments, Coke Zero as American Spirits, regular Coke as Marlboro Reds, full-fat glass-bottle Coke as a cigar.Interestingly, Coca‑Cola didn’t engineer this trend—it happened organically. Their marketing simply delivered delicious taste, and TikTok reshaped it into a cultural phenomenon.

Health trade‑offs: Less harm, but not harmless

No nicotine? Does that imply no health hazards at all?Well, that’s not true.Acidic erosion: Diet Coke has a pH of around 2.7–3.0, thanks to phosphoric and citric acids – enough to weaken enamel below the critical 4.0 level. Studies show that frequent soda exposure roughens enamel and accelerates erosion – punctual exposure may be manageable, but repeated sipping overwhelms saliva’s restorative ability.Artificial sweeteners: Though sugar-free, Diet Coke relies on aspartame and acesulfame K. WHO now labels aspartame “possibly carcinogenic,” and studies link diet sodas with heart disease, stroke, diabetes, and gut disruption.Dental and chemical concerns: Yes, it’s a vast improvement from tobacco, but Diet Coke isn’t innocent. Experts warn that its acidity can erode enamel and artificial sweeteners like aspartame may be linked to long-term health issues such as stroke, heart disease, and metabolic disruptions.Caffeine and bloating: Caffeine can cause jitters, disrupted sleep, and mild dehydration, while carbonation leads to bloating and digestive discomfort.Most professionals advise treating fridge cigarettes as occasional reminders, not all-day habits. Sip with water, use a straw, and maybe rotate in non-sweet beverages and mindful breaks.

The final sip: A balanced act

Yes – “fridge cigarettes” offer Gen Z a clever, safer stand-in for old-school smoke breaks. But they come with real risks: enamel erosion, enamel weakening, artificial sweetener exposure, and caffeine side effects. It’s one thing to enjoy the fizz and take a moment; it’s another to sip mindlessly all day.So, what’s the bottom line? Enjoy the crack and fizz, but make it intentional. A mindful fridge cigarette – sipped fast, chased with water, spaced out through the day – is a fun ritual. But replace it regularly with water, tea, or gum to safeguard your health. After all, you can pause without damaging your smile – or your wellness.

Introduction

For the past few decades, the incidence of thyroid cancer is increasing rapidly worldwide with papillary thyroid carcinoma (PTC) being the most common type of thyroid cancer.¹ Except for the actual increase in tumor occurrence, the increased prevalence of PTC is due to the increasing use of high-resolution ultrasound (US) imaging and US-guided fine-needle aspiration biopsies (FNAB).² The increased accuracy of pathological thyroid examinations also contributes to the increased diagnosis of PTC, especially the overdiagnosis of papillary thyroid microcarcinoma (PTMC),³ which is defined as a PTC with a diameter of ≤10 mm.⁴ Studies reported that most of PTMCs (over 70%) are diagnosed incidentally in autopsies and thyroidectomy specimens, and most PTMCs are benign with reported mortality from 0% to 1%.⁵ Therefore, the risks of overdiagnosis and overtreatment are highly possible for PTMCs, which causes potential complications, distress, and economic burdens for patients with PTMCs.⁶ Preoperative differentiation of PTMC from PTC is of great clinical significance to avoid overtreatment and to determine the appropriate treatment options for patients with PTMCs.⁷

US imaging of the thyroid and neck is usually the initial workup for a patient with a thyroid nodule, but the visual interpretation of US images in the diagnosis of PTC and PTMC is limited not only because the diagnosis highly depends on the radiologists’ experience but also the interobserver variation.⁸ Although the study of Ma et al demonstrated that combining conventional US, contrast-enhanced ultrasound (CEUS) and real-time elastography (RTE) is able to improve the diagnostic accuracy of PTMC, it is not a routine clinical practice for a patient with all three types of US exams.⁹ With the emergence of radiomics, studies reported that US-based radiomics is able to differentiate benign and malignant thyroid nodules.^10,11 Deep learning models with convolutional neural network (CNN) have also been investigated for the differentiation of benign and malignant thyroid nodules with US images and demonstrated excellent performance compared with radiologists.^12–14 However, few studies have addressed the application of radiomics and deep learning models in the differentiation of PTC and PTMC.

In this study, the feasibility and accuracy of US-based radiomics, deep learning, and combined deep learning radiomics models were investigated in the differentiation of PTMC and PTC to decrease the risk of overtreatment with patients’ data from two hospitals, one hospital for training and testing, and second hospital for external validation.

Materials and Methods

Patients

Patients diagnosed with papillary thyroid neoplasm in Hospital One from January 2018 to September 2020 were retrospectively reviewed according to the electronic medical records. Enrolled patients were randomly divided into training, validation, and independent testing sets. Fifty PTC cases acquired from Hospital Two were used as an independent external testing dataset. The inclusion criteria were as follows: 1) pathologically confirmed PTMCs and PTCs; 2) diagnosed by US images with detailed sonographic features described. The exclusion criteria included the followings: 1) with preoperative therapy (resection biopsy, neoadjuvant radiotherapy, or chemotherapy); 2) benign thyroid lesions; 3) missing important histopathological results (immunohistochemical results or lymph-nodule results); 4) incomplete information or images. Routine clinical tests and patients’ characteristics were also extracted from the records. Figure 1 shows the flowchart for patients’ enrollment. This study was approved by the ethics committee in Clinical Research (ECCR) of the First Affiliated Hospital of Wenzhou Medical University and conducted following the Declaration of Helsinki (ECCR no. 2019059) with confirmed patient confidentiality. The requirement of informed consent was waived by the ECCR according to the retrospective nature of this study.

Figure 1 The patient enrollment process for the training set and the two independent testing sets. The training cohort and the independent testing set 1 patient enrollment process are shown on the left, and the independent testing set 2 is shown on the right. These two datasets are from two different hospital patients.

US Examinations and Clinical Factors

US examinations of thyroid nodules were performed with high-frequency linear probes (5 MHz to 14 MHz) with a variety of US systems: Philips EPIQ7C (Philips Medical Systems, the Netherlands), GE Volume E8 (GE Medical Systems, USA), Siemens ACUSON OXANA 2 (Siemens Medical Solutions, USA), Esaote MyLab Class C (Esaote, Italy), Hitachi HI VISION Preirus (Hitachi-Aloka Medical, Japan) and Mindray Resona 7T (Mindray Medical International, China). The US images included both transverse and longitudinal sections of nodules. Clinical characteristics included basic information (age and sex) and ultrasound findings, which consist of composition, echogenicity, shape, margin, and echogenic foci, as well as stages scored according to the thyroid imaging reporting and data system (TI-RADS) criteria of the American College of Radiology.¹⁵

Radiomics Features Extraction and Modeling

Target volumes were contoured manually by one junior radiologist on the US images and confirmed by a senior radiologist with over 15 years of experience. Supplementary Figure 1 shows the typical US with contoured target volumes. Python (v. 3.7.0; https://www.python.org/) and package pyradiomics 2.2.0 (version 2.2) were used to extract radiomics features from the manually segmented target volumes. Based on different matrices that capture the spatial intensity distribution and wavelet filtering, a total of 1566 radiomics features were extracted, which includes 88 exponential features, 88 gradient features, 88 logarithm features, 70 square features, 352 wavelet features, and 880 log features, respectively. Radiomics features with a p <0.05 in Mann–Whitney U-tests were selected as potentially informative features, then the least absolute shrinkage selection operator (LASSO) was applied to identify optimal features for PTMC and PTC classification.¹⁶ A ten-fold cross validation was applied to tune the elastic net parameters to reduce the redundant information and to avoid over-fitting. A minimum standard deviation and maximum area under curves (AUC) were achieved by tuning coefficient λ. The linear combination of selected radiomics features with respective weights makes the final radiomics signature.

Deep Learning Models

In the preprocessing, a rectangular region of interest (ROI) was cropped from raw US images according to the tumor segmentation mask and resized to 224×224 pixels for normalization. Five deep learning networks, visual geometry group 13 (VGG13),¹⁷ VGG16, VGG19, AlexNet, and EfficientNet were pre-trained on Imagenet and then adopted as the deep learning networks in this study using python 3.7 programming language Tensorflow 2.4, and Keras 2.2.4 open-source programming packages.¹⁸ As shown in Supplementary Figure 2 the structures of different deep learning networks, VGG uses a pooling layer as a demarcation and has six block structures with each having the same number of channels. Because both the convolutional and fully connected layers have weight coefficients, they are also referred to as weight layers. The pooling layer does not involve weights. For the VGG CNN, its convolutional layers and pooling layers are responsible for feature extraction, and the final 3 fully connected layers are responsible for the classification task (×2 means that the module in brackets is repeated twice). The AlexNet network consists of 8 layers with the first 5 layers convolutional and the last 3 layers of fully connected. Each backbone of the EfficientNet contains 7 blocks with each having different numbers of sub-blocks.

In the training stage, rectangular ROIs were fed into the networks to update model parameters with classification results as the output. The loss function was calculated based on the cross-entropy of the outputs and labels. A learning rate of 1e-4, dropout parameter of 0.6, training Epoch of 100, the activation function ReLu, the classification function softmax, and the Adam optimizer were applied to update the model parameters with a batch size of 32 and a maximum iteration step of 300. Compiled prediction codes were generated based on the data weights obtained from training to obtain the predicted value of each test picture. Identical parameters were applied for the training of these deep learning networks. The computer environment configuration for deep learning modeling was Linux 64-bit operating system, GPU RTX2080, and video memory 6GB with all other applications shut down while the program is running.

Fused Models and Evaluation

Firstly, radiomics and deep learning models were developed to differentiate PTC and PTMC independently. In order to improve the classification performance, the prediction scores of radiomics and deep learning models were fused by applying an information fusion method.¹⁹ As shown in Figure 2, the prediction results were obtained independently by training each model in the early stage, then logistic regression was applied to fuse the outputs of each model to make a decision. The minimum loss value, positive predictive value (PPV), negative predictive value (NPV), recall rate, F1 score, and the AUCs of receiver operating characteristics (ROC) curves were calculated to evaluate and compare the performance of these models.²⁰

Figure 2 Schematic pipeline of the radiomics and deep learning modeling, as well as their combination for the differentiation of papillary thyroid tumor and papillary thyroid microcarcinoma. Abbreviations: Fc2, 2 fully connected layer; R_score, radiomics score.

Statistical Analysis

Detailed clinical differences between PTC and PTMC were compared by t-test, chi-square test, and Mann–Whitney U-test. LASSO regression model building was done using the “glmnet” package. Glmnet function in R language was applied for n cross validation (n = 10), which means that data was separated into 10 subsets. All statistics were two-sided and p-values less than 0.05 were considered to be statistically significant. Statistical analysis was performed using the R analysis platform (version 3.6.0), OriginPro2018, MedCalc (version 19.3.0), SPSS 19 software, and Python 3.7.

Results

A total of 549 patients with an average age of 46.55±11.33 years (ranges from 14 to 81 years) were enrolled in this study with confirmed 180 PTC and 436 PTMC nodules from Hospital One, as shown in Figure 1 the flowchart of patient selection. Patients were randomly divided into training cohort and validation cohort at a ratio of 8:2. There were 56 patients from Hospital One used as independent testing set 1. Fifty patients from Hospital Two with confirmed 25 PTC and 25 PTMC nodules were enrolled as independent testing set 2 for the external validation of these models. There were 205 PTC and 461 PTMC nodules with an average diameter of 15.80±7.15 mm and 6.89±3.25 mm, respectively. Detailed clinical characteristics of these patients were presented in Table 1. A total of 612 PTMC and 311 PTC US images were analyzed.

Table 1 The Clinical Characteristics of Enrolled Patients for Training and Testing

A total of 203 radiomics features were selected out of the extracted 1777 features according to the Mann–Whitney U-test with a p <0.05. A final 10 features were further screened out from the 203 features to build the radiomics signature using the LASSO logistic regression model, as shown in Supplementary Figure 3. These features included 5 first order features, 2 grey level run length matrix (GLRLM) features, 1 Gray Level Dependence Matrix (GLDM), and 2 Gray Level Size Zone Matrix (GLSZM). Detailed features and their corresponding non-zero coefficients were presented in Table S1. The ROC evaluation of radiomics signature in the differentiation of PTC and PTMC was shown in Figure 3 with an AUC of 0.908 (95% CI: 0.887–0.928), 0.826 (95% CI: 0.734–0.918), and 0.822 (95% CI: 0.710–0.936) in the validation cohort, independent testing set 1 and independent testing set 2, respectively.

Figure 3 The performance of radiomics signature with ROC curves in (a) training, (b) independent testing set 1, and (c) independent testing set 2.

Table 2 shows the performance of different deep learning models in the differentiation of PTC and PTMC with an accuracy of 0.800 (95% CI: 0.700–0.837), 0.850 (95% CI: 0.775–0.875), 0.850 (95% CI: 0.737–0.873), 0.850 (95% CI: 0.763–0.875) and 0.863 (95% CI: 0.787–0.932) in the validation cohort for AlexNet, VGG13, VGG16, VGG19, and EfficientNet, respectively. The ROC curves and the corresponding confusion matrices of different deep learning models were shown in Figure 4. As shown in Figure 4b the AUCs of AlexNet, VGG13, VGG16, VGG19, and EfficientNet were 0.800 (95% CI: 0.708–0.892), 0.850 (95% CI: 0.772–0.928), 0.846 (95% CI: 0.765–0.927), 0.890 (95% CI: 0.818–0.962), and 0.867 (95% CI: 0.789–0.945), respectively. Accordingly, VGG19 and EfficientNet were selected to combine with radiomics signature for further analysis. To further interpret the results of deep learning models, a heatmap of output features was generated by activating the visualization class. The generated heatmap was then overlaid onto the original image to generate the final deep learning visualization heatmap, as shown in Figure 4a the US images of 3 cases of PTMC and PTC, and their activated heatmap with VGG19. The heatmap could produce a coarse localization map highlighting the import regions for the classification targets.

Table 2 The Performance of Different Deep Learning Models in the Training and Validation Cohorts

Figure 4 (a) The network features and heatmaps of 3 cases of papillary microcarcinoma of the thyroid and papillary thyroid carcinoma. The ROC curves and the corresponding confusion matrices of different deep learning models, (b) AlexNet, VGG13, VGG16, VGG19, EfficientNet; (c) ROCs comparison.

Detailed comparison of the performance of radiomics signature, deep learning models, and combined deep learning radiomics models with the independent testing set 1 and set 2 were presented in Table 3. The accuracy of VGG19, EfficientNet, Radiomics, combined radiomics VGG19 (R_V_combined), and radiomics EffiecientNet combination (R_E_combined) were 0.829, 0.798, 0.766, 0.904, 0.851 and 0.680, 0.680, 0.740, 0.780, 0.900 with the independent testing set 1 and set 2, respectively. Figure 5 shows the ROC curves of these models with an AUC of 0.826 (95% CI: 0.734–0.918), 0.890 (95% CI: 0.818–0.962), 0.867 (95% CI: 0.790–0.945), 0.931 (95% CI: 0.870–0.993), 0.908 (95% CI: 0.849–0.966) and 0.822 (95% CI: 0.709–0.936), 0.698 (95% CI: 0.546–0.849), 0.899 (95% CI: 0.806–0.993), 0.874 (95% CI: 0.778–0.969), 0.946 (95% CI: 0.885–1.000) for Radiomics, VGG19, EfficientNet, R_V_combined, and R_E_combined with the independent testing set 1 and set 2, respectively.

Table 3 The Performance of Radiomics Signature, Deep Learning Models and Combined Deep Learning Radiomics Models with Independent Test 1 and 2

Figure 5 The performance of radiomics, deep learning, and combined deep learning radiomics models in differentiating PTMC from PTC with (a) independent testing set 1; (b) independent testing set 2.

Discussion

In this study, the feasibility and accuracy of radiomics, deep learning models, and combined deep learning radiomics models were investigated in the differentiation of PTMC from PTC using US images. The models were further verified with external validation cohorts from a second hospital. An AUC of 0.826, 0.890, 0.867, 0.931, 0.908 and 0.822, 0.698, 0.899, 0.874, 0.946 for Radiomics, VGG19, EfficientNet, R_V_combined, and R_E_combined models was achieved with the independent testing set 1 and set 2, respectively.

With the increasing prevalence of high-resolution US and other imaging modalities, the diagnosis of thyroid cancers has increased remarkably and continuously.²¹ Most of the newly diagnosed thyroid cancers are small PTCs including PTMCs.²² Studies reported that the average size of thyroid tumors decreased from 1.51 cm in the year 2000 to 1.02 cm in 2005 with 36.9% of thyroid cancers being small than 1 cm in 2000, but PTMCs account for 61.48% of all thyroid cancers in 2005.²³ Similarly, PTMCs account for about 70% of all cases enrolled in this study. However, more evidence demonstrated that most PTMCs have a very indolent nature and excellent outcomes.²⁴ The increasing awareness of the impact of overtreatment on PTMCs also changed dramatically the international guidelines.⁶ Therefore, non-invasive methods are urgently needed to differentiate PTMC from PTC preoperative to avoid overtreatment for patients with PTMC regardless of the increased diagnosis of thyroid cancers.

US is widely applied for preoperative imaging of PTC for diagnosis and staging. US features, such as irregular border, halo sign microcalcifications, macrocalcifications, isoechoic, and hypoechoic appearance, had been investigated to differentiate PTC from PTMC clinically.²⁵ An AUC of 0.97, a sensitivity of 88.6%, and a specificity of 94.6% were reported in the prediction of PTMC with combined conventional US, CEUS, and RTE.⁹ However, the accuracy of US diagnosis is easily affected by image quality and the experience of US technician and radiologists, who handles the probe and interprets the US images.²⁴ With the emergence of radiomics, the radiomics model had been proposed as a promising method to assess the risk of PTC metastasis using US images and achieved an AUC of 0.782 and an accuracy of 0.710, respectively.²⁶ The presence of extrathyroidal extension (ETE) of PTC was preoperatively predicted with a radiomics model using CT images and achieved an AUC of 0.812 in the validation cohort.²⁷ Radiomics models were also investigated to predict the presence of B-Raf proto-oncogene, serine/threonine kinase (BRAF) mutation in PTC with an average AUC of 0.651, an accuracy of 64.3%, a sensitivity of 66.8%, and a specificity of 61.8% using US images, respectively.²⁸ However, to the best of our knowledge, this is the first study to investigate the feasibility of US-based radiomics and deep learning models in the differentiation of PTC and PTMC. After manual target segmentation, feature extraction, and selection, an AUC of 0.826 and 0.822 was achieved with radiomics signature in the independent testing set 1 and set 2, respectively, in this study. The performance of our radiomics model was promising in comparison with other radiomics studies for PTC as mentioned previously.^28–30 However, it was inferior to the direct prediction of PTC and PTMC with combined US, CEUS, and RTE.⁹

Five deep learning networks were adapted and trained in this study to further investigate the differentiation of PTMC from PTC in this study. As shown in Table 2 and Figure 4, VGG19 achieved a best AUC of 0.890 and EfficientNet achieved a best accuracy of 0.867, respectively. As shown in Table 3, the performance of radiomics was inferior to two deep learning models in accuracy and AUC with independent testing set 1; however, the accuracy of the radiomics model was better than those of two deep learning models in accuracy with an external validation set (independent testing set 2). However, advanced deep machine learning approaches have been developed to handle the challenging health problems and to diagnose thyroid cancer.^31,32 In this study, the application of deep learning networks did not guarantee superior differentiation ability for PTC and PTMC. On the other hand, the interpretability of deep learning models was weaker in comparison with radiomics features.³³ In this study, visualization class was activated to generate a heatmap after the differentiation with deep learning networks to increase the interpretability of our deep learning models. As shown in Figure 4a the heatmap corresponding to the boundary of the tumor and the low echo area inside the tumor were highlighted, which are the features corresponding to deep learning network features.

In this study, the differentiation ability and accuracy of these models were further improved by combining radiomics and deep learning networks, as shown in Table 3 and Figure 5. The best accuracy and AUC of 0.904, 0.900, and 0.931, 0.946 were achieved with the combination of VGG + radiomics (R_V_Combined) and EffiecientNet + radiomics (R_E_Combined) in the independent testing set 1 and set 2, respectively. Consistently, deep learning radiomics had been reported to improve the prediction ability using US images for breast cancers.^34,35 The combination of radiomics scores and deep learning prediction scores was usually fused by information fusion, which includes three strategies of early fusion, mid-term fusion, and late-stage fusion.¹⁹ However, only the late-stage fusion strategy was applied in this study. Although automatic segmentation on US images for cervical cancer and ovarian cancer had been intensively investigated, further study is still necessary to transfer the automatic segmentation methods to PTC and PTMC on US images.^29,30 On the other hand, the stability and reproducibility of radiomics features may be easily affected by the type of scan machines and the different automatic segmentation algorithms.^36,37 Another limitation of this study is that the influence of clinical factors in the differentiation of PTMC and PTC was not fully investigated.

Conclusions

Deep learning and radiomics combination models are promising in the noninvasively preoperative differentiation of PTMC and PTC to decrease the overtreatment of patients with PTMC and to minimize the complications caused by overtreatment.

Ethics Approval and Informed Consent

This study was approved by the ethics committee in Clinical Research (ECCR) of authors’ hospital and conducted following the Declaration of Helsinki (ECCR no. 2019059). The requirement of informed consent was waived by the ECCR.

Acknowledgments

We would like to thank Jianping Wu, who made a significant contribution in the revision of this manuscript.

Funding

This research was supported partially by a National Natural Science Foundation (12475352), a key project of Zhejiang Natural Science Foundation (LZ24A050008), a Key project of Zhejiang Provincial Health Science and Technology Program (WKJ-ZJ-2437), a Major project of Wenzhou Science and Technology Bureau (ZY2022016, ZY2020011), Zhejiang Engineering Research Center for innovation and application of Intelligent Radiotherapy Technology, Zhejiang-Hong Kong Precision Theranostics of Thoracic Tumors Joint Laboratory, and Wenzhou key Laboratory of basic science and translational research of radiation oncology, Zhejiang Key Laboratory of Intelligent Cancer Biomarker Discovery and Translation, Discipline Cluster of Oncology, Wenzhou Medical University.

Disclosure

No potential conflict of interest relevant to this article was reported. This study had been presented in the 2023 conference of the American Association of Physicists in Medicine.

References

1. Seib CD, Sosa JA. Evolving understanding of the epidemiology of thyroid cancer. Endocrinol Metab Clin North Am. 2019;48(1):23–35. doi:10.1016/j.ecl.2018.10.002

2. Vaccarella S, Franceschi S, Bray F, Wild CP, Plummer M, Dal Maso L. Worldwide thyroid-cancer epidemic? The increasing impact of overdiagnosis. N Engl J Med. 2016;375(7):614–617. doi:10.1056/NEJMp1604412

3. Vigneri R, Malandrino P, Vigneri P. The changing epidemiology of thyroid cancer: why is incidence increasing? Curr Opin Oncol. 2015;27(1):1–7. doi:10.1097/CCO.0000000000000148

4. Lloyd RV, Buehler D, Khanafshar E. Papillary thyroid carcinoma variants. Head Neck Pathol. 2011;5(1):51–56. doi:10.1007/s12105-010-0236-9

5. Wang W, Kong L, Guo H, Chen X. Prevalence and predictor for malignancy of contralateral thyroid nodules in patients with unilateral PTMC: a systematic review and meta-analysis. Endocr Connect. 2021;10(6):656–666. doi:10.1530/EC-21-0164

6. Sugitani I, Ito Y, Takeuchi D, et al. Indications and strategy for active surveillance of adult low-risk papillary thyroid microcarcinoma: consensus statements from the Japan association of endocrine surgery task force on management for papillary thyroid microcarcinoma. Thyroid. 2021;31(2):183–192. doi:10.1089/thy.2020.0330

7. Li M, Dal Maso L, Vaccarella S. Global trends in thyroid cancer incidence and the impact of overdiagnosis. Lancet Diabetes Endocrinol. 2020;8(6):468–470. doi:10.1016/S2213-8587(20)30115-7

8. Chen HY, Liu WY, Zhu H, et al. Diagnostic value of contrast-enhanced ultrasound in papillary thyroid microcarcinoma. Exp Ther Med. 2016;11(5):1555–1562. doi:10.3892/etm.2016.3094

9. Ma HJ, Yang JC, Leng ZP, Chang Y, Kang H, Teng LH. Preoperative prediction of papillary thyroid microcarcinoma via multiparameter ultrasound. Acta Radiol. 2017;58(11):1303–1311. doi:10.1177/0284185117692167

10. Gillies RJ, Kinahan PE, Hricak H. Radiomics: images are more than pictures, they are data. Radiology. 2016;278(2):563–577. doi:10.1148/radiol.2015151169

11. Kwon MR, Shin JH, Hahn SY, et al. Histogram analysis of greyscale sonograms to differentiate between the subtypes of follicular variant of papillary thyroid cancer. Clin Radiol. 2018;73(6):591.e1–591.e7. doi:10.1016/j.crad.2017.12.008

12. Akkus Z, Cai J, Boonrod A, et al. A survey of deep-learning applications in ultrasound: artificial intelligence-powered ultrasound for improving clinical workflow. J Am Coll Radiol. 2019;16(9 Pt B):1318–1328. doi:10.1016/j.jacr.2019.06.004

13. Li X, Zhang S, Zhang Q, et al. Diagnosis of thyroid cancer using deep convolutional neural network models applied to sonographic images: a retrospective, multicohort, diagnostic study [published correction appears in Lancet Oncol. 2020 Oct;21(10):e462. doi: 10.1016/S1470-2045(20)30546-5]. Lancet Oncol. 2019;20(2):193–201. doi:10.1016/S1470-2045(18)30762-9

14. Peng S, Liu Y, Lv W, et al. Deep learning-based artificial intelligence model to assist thyroid nodule diagnosis and management: a multicentre diagnostic study [published correction appears in Lancet Digit Health. 2021 Jul;3(7):e413]. Lancet Digit Health. 2021;3(4):e250–e259. doi:10.1016/S2589-7500(21)00041-8

15. Tessler FN, Middleton WD, Grant EG. Thyroid imaging reporting and data system (TI-RADS): a user’s guide [published correction appears in Radiology. 2018 Jun;287(3):1082]. Radiology. 2018;287(1):29–36. doi:10.1148/radiol.2017171240

16. Friedman J, Hastie T, Tibshirani R. Regularization paths for generalized linear models via coordinate descent. J Stat Softw. 2010;33(1):1–22. doi:10.18637/jss.v033.i01

17. Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. arxiv:14091556[csCV].

18. He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. arxiv:151203385[csCV].

19. Meurer WJ, Tolles J. Logistic regression diagnostics: understanding how well a model predicts outcomes. JAMA. 2017;317(10):1068–1069. doi:10.1001/jama.2016.20441

20. Hanley JA, McNeil BJ. A method of comparing the areas under receiver operating characteristic curves derived from the same cases. Radiology. 1983;148(3):839–843. doi:10.1148/radiology.148.3.6878708

21. Jung KW, Won YJ, Kong HJ, Lee ES; Community of Population-Based Regional Cancer Registries. Cancer statistics in Korea: incidence, mortality, survival, and prevalence in 2015. Cancer Res Treat. 2018;50(2):303–316. doi:10.4143/crt.2018.143

22. Ahn HS, Kim HJ, Welch HG. Korea’s thyroid-cancer “Epidemic” — screening and overdiagnosis. N Engl J Med. 2014;371(19):1765–1767. doi:10.1056/NEJMp1409841

23. Cordioli MI, Canalli MH, Coral MH. Increase incidence of thyroid cancer in Florianopolis, Brazil: comparative study of diagnosed cases in 2000 and 2005. Arq Bras Endocrinol Metabol. 2009;53(4):453–460. doi:10.1590/s0004-27302009000400011

24. Haugen BR, Alexander EK, Bible KC, et al. 2015 American thyroid association management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: the American thyroid association guidelines task force on thyroid nodules and differentiated thyroid cancer. Thyroid. 2016;26(1):1–133. doi:10.1089/thy.2015.0020

25. Zhang XL, Qian LX. Ultrasonic features of papillary thyroid microcarcinoma and non-microcarcinoma. Exp Ther Med. 2014;8(4):1335–1339. doi:10.3892/etm.2014.1910

26. Liu T, Zhou S, Yu J, et al. Prediction of lymph node metastasis in patients with papillary thyroid carcinoma: a radiomics method based on preoperative ultrasound images. Technol Cancer Res Treat. 2019;18:1533033819831713. doi:10.1177/1533033819831713

27. Chen B, Zhong L, Dong D, et al. Computed tomography radiomic nomogram for preoperative prediction of extrathyroidal extension in papillary thyroid carcinoma. Front Oncol. 2019;9:829. doi:10.3389/fonc.2019.00829

28. Kwon MR, Shin JH, Park H, Cho H, Hahn SY, Park KW. Radiomics study of thyroid ultrasound for predicting braf mutation in papillary thyroid carcinoma: preliminary results. AJNR Am J Neuroradiol. 2020;41(4):700–705. doi:10.3174/ajnr.A6505

29. Jin J, Zhu H, Zhang J, et al. Multiple U-net-based automatic segmentations and radiomics feature stability on ultrasound images for patients with ovarian cancer. Front Oncol. 2021;10:614201. doi:10.3389/fonc.2020.614201

30. Jin J, Zhu H, Teng Y, Ai Y, Xie C, Jin X. The accuracy and radiomics feature effects of multiple U-net-based automatic segmentation models for transvaginal ultrasound images of cervical cancer. J Digit Imaging. 2022;35(4):983–992. doi:10.1007/s10278-022-00620-z

31. Tutsoy O, Koç GG. Deep self-supervised machine learning algorithms with a novel feature elimination and selection approaches for blood test-based multi-dimensional health risks classification. BMC Bioinf. 2024;25(1):103. doi:10.1186/s12859-024-05729-2

32. Tutsoy O, Sumbul HE. A novel deep machine learning algorithm with dimensionality and size reduction approaches for feature elimination: thyroid cancer diagnoses with randomly missing data. Brief Bioinform. 2024;25(4):bbae344. doi:10.1093/bib/bbae344

33. Gandin I, Scagnetto A, Romani S, Barbati G. Interpretability of time-series deep learning models: a study in cardiovascular patients admitted to Intensive care unit. J Biomed Inform. 2021;121:103876. doi:10.1016/j.jbi.2021.103876

34. Zheng X, Yao Z, Huang Y, et al. Deep learning radiomics can predict axillary lymph node status in early-stage breast cancer [published correction appears in Nat Commun. 2021 Jul 12;12(1):4370]. Nat Commun. 2020;11(1):1236. doi:10.1038/s41467-020-15027-z

35. Jiang M, Li CL, Luo XM, et al. Ultrasound-based deep learning radiomics in the assessment of pathological complete response to neoadjuvant chemotherapy in locally advanced breast cancer. Eur J Cancer. 2021;147:95–105. doi:10.1016/j.ejca.2021.01.028

36. Yi J, Lei X, Zhang L, et al. The influence of different ultrasonic machines on radiomics models in prediction lymph node metastasis for patients with cervical cancer. Technol Cancer Res Treat. 2022;21:15330338221118412. doi:10.1177/15330338221118412

37. Teng Y, Ai Y, Liang T, et al. The effects of automatic segmentations on preoperative lymph node status prediction models with ultrasound radiomics for patients with early stage cervical cancer. Technol Cancer Res Treat. 2022;21:15330338221099396. doi:10.1177/15330338221099396

Within months of Pennsylvania’s medical cannabis program adding anxiety as a qualifying condition, that diagnosis quickly rose to become the most common for cannabis certifications, according to a study by researchers at the University of Pittsburgh and Johns Hopkins University. The study was published today in Annals of Internal Medicine.

To date, 39 states have medical cannabis programs, with chronic pain and post-traumatic stress disorder (PTSD) historically being the most common and second-most common qualifying diagnoses, respectively, among participants. In recent years, several states, including Pennsylvania, have incorporated anxiety into their programs.

We found that adding anxiety as a qualifying condition fundamentally changed the makeup of Pennsylvania’s medical cannabis program.”

Coleman Drake, Ph.D., corresponding author, associate professor in the Department of Health Policy and Management at Pitt’s School of Public Health

The researchers analyzed Pennsylvania Department of Health data on 1,730,600 medical cannabis certifications issued from November 2017, when anxiety disorders were added, to December 2023. These certifications require a physician visit and annual renewal.

The team found that the number of certifications issued each month nearly tripled during that time. Before the list of qualifying conditions was expanded, chronic pain comprised the lion’s share of diagnoses, at 67%, followed by 16% for PTSD. After anxiety was added, these numbers dropped to 41% and 11%, respectively, and anxiety quickly became the most common diagnosis, at 60%. Some certifications listed multiple conditions. The team was not able to determine how the overall size of the program was affected, nor how many participants had added or switched to anxiety from another diagnosis or were enrolling in the program for the first time.

Evidence supporting cannabis as an effective treatment for anxiety disorders is scant in comparison to other qualifying conditions, notably chronic pain, notes Drake.

“Adding anxiety to the program may inadvertently signal to patients that cannabis is effective for treating it, despite the lack of evidence, which is concerning,” he said. “At the same time, cannabis may improve some health outcomes, relative to alternative treatments, depending on the individual and their circumstances.” But, due to the red tape and funding scarcity that has historically restricted research on cannabis-as well as the lack of more granular data available from medical and adult-use cannabis programs-these unknowns persist, he said.

“The urgency in filling these knowledge gaps is pretty clear, given increases in cannabis use over the past decade, and the large changes in cannabis markets, like those we observed in this study,” said Drake.

Other authors on the study were Linh Tran and Matthew Eisenberg, Ph.D., of Johns Hopkins University Bloomberg School of Public Health.

Specialty

Please chooseI’m not a medical professional.Allergy and ImmunologyAnatomyAnesthesiologyBiostatisticsCardiac/Thoracic/Vascular SurgeryCardiologyCritical CareDentistryDermatologyDiabetes and EndocrinologyEmergency MedicineEpidemiology and Public HealthFamily MedicineForensic MedicineGastroenterologyGeneral PracticeGeneticsGeriatricsHealth PolicyHematologyHIV/AIDSHospital-based MedicineI’m not a medical professional.Infectious DiseaseIntegrative/Complementary MedicineInternal MedicineInternal Medicine-PediatricsMedical Education and SimulationMedical PhysicsMedical StudentNephrologyNeurological SurgeryNeurologyNuclear MedicineNutritionObstetrics and GynecologyOccupational HealthOncologyOphthalmologyOptometryOral MedicineOrthopaedicsOsteopathic MedicineOtolaryngologyPain ManagementPalliative CarePathologyPediatricsPediatric SurgeryPharmacologyPhysical Medicine and RehabilitationPlastic SurgeryPodiatryPreventive MedicinePsychiatryPsychologyPulmonologyRadiation OncologyRadiologyRheumatologySubstance Use and AddictionSurgeryTherapeuticsTraumaUrologyMiscellaneous

Introduction

Diabetes Mellitus (DM) is a chronic metabolic disorder characterized by elevated blood glucose levels due to either insulin deficiency or insulin resistance.¹ It is commonly classified into Type 1 (T1DM) and Type 2 (T2DM). T1DM is primarily caused by an autoimmune-mediated destruction of pancreatic β-cells, leading to a complete deficiency of insulin, while T2DM is characterized by insulin resistance and a progressive decline in insulin secretion.² Several risk factors increase the likelihood of developing T2DM, including a strong family history of diabetes, advancing age, obesity, and physical inactivity.³ In Saudi Arabia, the prevalence of diabetes is notably high, with 17.7% of adults living with the condition, according to the International Diabetes Federation.⁴ Alarmingly, the prevalence of T2DM has risen from 15.8% in 2016 to 18.2% in 2021, and if the trend continues, it is projected to exceed 20% by 2026.⁵

Effective management of DM requires significant lifestyle modifications, including structured meal planning and regular physical activity.⁶ In patients who are overweight or obese, even a modest reduction in body weight has been shown to improve glycemic control, lipid profiles, and blood pressure regulation.⁶ However, inadequate disease management and poor adherence to prescribed treatment regimens can lead to serious chronic complications. These complications are broadly categorized into macrovascular complications, such as coronary artery disease, stroke, and myocardial infarction, and microvascular complications, including diabetic retinopathy, nephropathy, and neuropathy.⁶ Among patients with T2DM in Saudi clinics, nephropathy was the most common microvascular complication (80.2%), followed by retinopathy (32.7%) and neuropathy (8.4%).⁷ Moreover, according to an analysis from the Saudi Health Interview Survey (SHIS), 3.5% of patients with diabetes had experienced a myocardial infarction, and 1.2% had suffered a stroke.⁸

Medication non-adherence is a well-documented challenge in the management of chronic diseases, particularly diabetes and hypertension, with global estimates indicating that up to 50% of patients do not adhere to their prescribed regimens.⁹ In Al-Ahsa, Saudi Arabia, previous studies have reported medication non-adherence rates as high as 65% among diabetic patients receiving primary care.⁹ Poor adherence to antidiabetic medications significantly compromises glycemic control, accelerates disease progression, and increases the risk of adverse health outcomes and complications.^9,10 Moreover, inadequate adherence is frequently associated with poor metabolic control, contributing to both immediate and long-term complications.¹¹ Despite the clinical significance of adherence, accurately assessing medication adherence remains a persistent challenge in diabetes management.¹²

The Adherence to Refills and Medications Scale (ARMS), comprising 12 items, is a widely utilized tool for assessing medication adherence. Its psychometric robustness, including reliability and validity, has been demonstrated in English-speaking populations,¹³ and it has been successfully translated and validated in several languages, such as Turkish,¹⁴ Korean,¹⁵ Chinese,¹⁶ Polish,¹⁷ and Arabic,¹⁸ particularly among patients with chronic conditions like diabetes and hypertension. Due to the lack of studies on refill adherence and the underrepresentation of certain populations, this is the first study to apply the ARMS tool to assess medication adherence among patients with T2DM in the Al-Ahsa region of Saudi Arabia. Using a stratified random sampling technique, the study aims to identify key factors influencing adherence in this population.

Methodology

Study Design and Sample Size

This is a cross-sectional study conducted over a three-month period, from May 2024 to July 2024, among T2DM patients in Al-Ahsa, Saudi Arabia. The primary aim was to assess adherence to refills and medications among patients with T2DM. Data were collected through phone call interviews using a structured questionnaire. The target population consisted of adult T2DM patients who visited governmental primary healthcare centers (PHCs), which provide free healthcare services, between October 2023 and March 2024. Patients diagnosed with T1DM and individuals who refused to participate or did not complete the study requirements were excluded.

A stratified random sampling method was employed to gather data. Initially, all PHCs in the Al-Ahsa region were categorized into four strata—Eastern, Middle, Northern, and Southern—based on their geographical locations. A specific sample size was allocated to each stratum, and participants were then randomly selected within each stratum using simple random sampling via Excel software (Figure 1). Table 1 shows the estimated sample size for each stratum, calculated using the formula with Z = 1.96 (95% confidence interval), E = 0.05 (5% margin of error), and P = 50%.

Table 1 Calculated Sample Size for Each Geographic Stratum

Figure 1 A Stratified Random Sampling Framework for Selecting Study Participants.

Data Collection

In this cross-sectional study, questionnaires were administered through phone call interviews, while weight, height, and HbA1C measurements were obtained from system records. For illiterate participants, the survey was completed with the assistance of their relatives. The self-reported questionnaire included sections on sociodemographic characteristics (age, gender, income level, marital status, level of education, occupation, and residency) and clinical profiles (family history of DM, age of onset of DM, disease duration, type and number of medications prescribed, duration of exercise, complications of DM, frequency of blood sugar monitoring, whether the physician provided information about medications, whether the physician provided adequate care, availability of assistance with medication, reasons for non-adherence to medication, number of doctor visits per year, and comorbid conditions and medications prescribed). The final section utilized the ARMS to assess medication adherence.¹⁸ Internal validity was evaluated using Cronbach’s alpha (α = 0.74), indicating acceptable internal consistency and reliability of the questionnaire within the study population.

Statistical Analysis

Data analysis for this study was conducted using SPSS version 27 (IBM Corp., 2020). Descriptive statistics were used to display the characteristics of the study population, including demographic, diabetes-related, health profile, and adherence-related variables. Continuous variables such as age and duration of diabetes were described using mean and standard deviation, while categorical variables such as gender, region, marital status, and diabetes treatment were summarized using frequency and percentage. For the analysis of factors influencing medication adherence and refill behaviors, bivariate tests, including Pearson’s Chi-square test, were used to examine relationships between independent variables (for example, demographics, diabetes control, and healthcare access) and the dependent variable of adherence. Significant associations were further explored using multiple logistic regression to identify the predictors of non-adherence to medication and refill behavior. Adjusted Odds ratios (OR) and confidence intervals (CI) were reported to assess the strength and direction of these associations. A P-value less than 0.05 was considered for statistical significance.

Ethical Statement

The confidentiality of all participants will be strictly maintained. Ethical approval has been obtained from the King Fahad Hospital-Al-Hofuf Ethics Committee (IRB-KFHH No. H-05-HS-065). The study will be thoroughly explained to potential respondents, and an informed consent will be obtained from all participants. The study will adhere to the principles outlined in the Declaration of Helsinki.

Results

Table 2 presents the bio-demographic characteristics of 732 type 2 diabetic patients in Saudi Arabia. Geographically, the study population was relatively evenly distributed across the regions, with the Eastern region at 27.9% (No=204), the Middle region at 26.5% (No=194), the Northern region at 25.7% (No=188), and the Southern region at 19.9% (No=146). Regarding gender, exact of 378 (51.6%) were females. The age distribution showed 223 (30.5%) aged 50–59 years and a mean age of 54.5 ± 12.7 years. In terms of body mass index, the majority of participants were either obese class I (30.7%, No=225) or overweight (27.7%, No=203), while only 15.0% (No=110) were within the normal weight range. A significant proportion of the participants reported a monthly income of less than 5000 SR (66.8; 489). The vast majority were married, accounting for 85.4% (No=625). Educational levels were varied, with basic education being the most common at 39.3% (No=288). Occupationally, nearly half were not working or were students, at 47.4% (No=347). Most participants resided in downtown areas, representing 71.4% (No=523). Notably, a substantial 86.6% (No=634) reported a family history of diabetes mellitus.

Table 2 Bio-Demographic Characteristics of the Study Type 2 Diabetic Patients in Saudi Arabia (N=732)

Table 3 shows the distribution of diabetes-related variables among 732 type 2 diabetic patients in Saudi Arabia. The age of diabetes onset was most frequently 50 years or older, comprising 31.4% (No=230) of the cases, while the lowest frequency was observed in those with onset between 30–39 years at 18.9% (No=138). Regarding treatment, oral pills were the most common modality, utilized by 48.2% (No=353) of patients, whereas diet and exercise alone represented the lowest at 0.7% (No=5). As for the duration of diabetes, the highest percentage being less than 5 years at 31.4% (No=230). About glycemic control, well-controlled HbA1c levels (<7%) were observed in 36.2% (No=265) of patients, while poor control (>8%) was seen in 31.6% (No=231). The most frequent blood glucose monitoring was more than once daily, at 28.7% (No=210), and the least frequent was every 3 months, at 0.4% (No=3). Notably, 57.8% (No=423) reported no diabetic complications, while retinopathy was the most prevalent complication at 31.3% (No=229), and stroke was the least at 1.8% (No=13).

Table 3 Distribution of Diabetes-Related Variables Among Type 2 Diabetic Patients in Saudi Arabia (N=732)

Table 4 illustrates the health profile of Type 2 diabetic patients in Saudi Arabia. Considering other comorbidities, the most common condition among the participants was hypertension, affecting 357 individuals (49.0%). Other comorbidities included rheumatic diseases, osteoarthritis, and others (132, 18.1%) and diseases of the digestive system such as inflammatory bowel disease (96, 13.2%). Regarding medication use, most participants were on multiple medications. A significant portion (192, 26.2%) received three medications, while six or more medications were prescribed to 142 participants (19.4%). Regarding smoking habits, 105 participants (14.3%) reported that they smoke. Finally, exercise habits show that a few numbers of participants (201, 27.5%) do not engage in any physical activity. Among those who do exercise, 294 participants (40.2%) reported engaging in less than 60 minutes of physical activity weekly, while 147 participants (20.1%) met the recommended 60–150 minutes of exercise per week.

Table 4 Health Profile of Type 2 Diabetic Patients in Saudi Arabia: Comorbidities, Medication, and Lifestyle Behaviors (N=732)

Table 5 clarifies data on healthcare access and medication support among 732 type 2 diabetic patients. A significant majority, 85.9% (No=629), reported receiving information about anti-diabetic medications from their physicians. Regarding the frequency of doctor visits, the highest percentage, 36.5% (No=267), reported visiting their doctor more than three times a year, and only 3.7% (No=27) reported visiting their doctor once a year. The vast majority of patients, 88.5% (No=648), confirmed they were continuing to receive healthcare from their doctors, with only 11.5% (No=84) reporting otherwise. In terms of medication assistance, a total of 79.9% (No=585) of patients reported managing their medications independently, while 20.1% (No=147) indicated they had someone who helped them.

Table 5 Healthcare Access and Medication Support in Type 2 Diabetic Patients (N=732)

Table 6 examines medication adherence and refill behaviors among 732 type 2 diabetic patients. A significant 33.9% of patients reported forgetting to take their medications at least sometimes. Similarly, 20.1% decided not to take their medications at least sometimes. Regarding medication refills, 13.4% forgot to refill their medications, and 14.2% ran out of medications at least sometimes. Furthermore, 16.7% skipped a dose without consulting their doctor, and 20.5% missed medications when feeling better. Notably, 6.6% missed medications when feeling sick, indicating a potential vulnerability during illness. Carelessness led to missed medications for 11.7%. A substantial 33.7% changed their medication dose to suit their needs, highlighting potential issues with medication management. Additionally, 20.1% forgot medications that they were supposed to take more than once a day. Finally, 4.1% put off their medications due to high costs. Conversely, a combined 58.6% reported refilling their medication before running out, indicating a reasonably high rate of proactive refill behavior.

Table 6 Adherence to Medication and Refill among Type 2 Diabetic patients (N=732)

Figure 2 displays the adherence rate to refills and medications among 732 type 2 diabetic patients. A total of 52.5% (n=384) of patients were classified as adherent and 47.5% (n=348) were classified as non-adherent.

Figure 2 Type 2 Diabetic Patients’ Adherence Rate to Re-fill and Medications (n=732).

Figure 3 presents the reasons that reduce medication adherence among Type 2 diabetic patients. The majority of participants (62.4%) reported no reasons for reducing their commitment to taking medication. Among those who reported barriers to adherence, the most common reason was forgetting to take medication (22.1%). Other reasons included stopping medications due to side effects (8.1%) and stopping medication when feeling well (8.6%). A smaller proportion of patients stopped taking medications because of their multiplicity (5.6%) or taking medication only when the disease worsens (5.1%). Financial constraints were less common, with only 1.4% of patients reporting inability to afford medication as a reason for non-adherence. Psychological factors and concerns about kidney and liver failure were reported by just 0.13% of patients each.

Figure 3 Reasons That Reduce Commitment to Taking Medication among Type 2 Diabetic Patients.

Table 7 examines factors associated with refill and medication adherence among type 2 diabetic patients. A highly significant association (p = 0.001) was observed with region. Patients residing in the Southern region demonstrated the lowest adherence rate, with only 22.6% being adherent, compared to higher adherence rates in the Eastern (62.3%), Middle (60.3%), and Northern (56.9%) regions. Monthly income also significantly impacted adherence (p = 0.001). Patients with a monthly income of less than 5000 SR exhibited the highest adherence rate at 61.1%, while those in higher income brackets showed substantially lower adherence. Marital status significantly influenced adherence (p = 0.002). Single patients had the lowest adherence rate, with only 32.6% being adherent, compared to 55.0% for married patients. Occupation showed a significant association with adherence (p = 0.008). Patients not working or students had a higher adherence rate of 56.5%, while retired patients had a lower rate of 42.9%. Residence type also showed a significant association (p=0.016). Patients who live in rural areas had a higher adherence rate of 56.1% compared to those who live in immigration areas (15.4%) or downtown (52%).

Table 7 Factors Associated with Diabetic Patient’s Re-fill and Medication Adherence

Table 7 presents other factors associated with diabetic patients’ medication refill and adherence. Significant factors influencing adherence include the age of onset of diabetes, with those diagnosed at age 50 or older showing the highest adherence (61.3%) compared to younger groups (p=0.001). Diabetic complications were also a key factor, with those without complications having better adherence (57.0%) than those with complications (46.3%, p=0.004). Diabetic control, as measured by HbA1c levels, showed that patients with well-controlled diabetes (<7%) had better adherence (58.1%) compared to those with poorly controlled diabetes (>8%, 45.9%, p=0.025). Monitoring blood glucose levels more frequently was positively associated with adherence. Co-morbidities also played a role; patients without other co-morbidities demonstrated better adherence (60.1%) compared to those with additional health conditions (48.2%, p=0.002). The number of medications prescribed influenced adherence as well. Patients on 3 medications showed the highest adherence (67.2%), while those on 6 or more medications had lower adherence (43.0%, p=0.068). Receiving information from a physician about anti-diabetic medications was another significant factor; those who received guidance had higher adherence (56.8%) compared to those who did not (26.2%, p=0.001). Doctor visits also correlated with adherence. Patients who visited their doctor more frequently (three or more times a year) had better adherence (66.3%) compared to those who only visited once a year (37.0%, p=0.001). Finally, the presence of a helper for medication was not significantly associated with adherence (p=0.134).

Table 8 presents the results of a multiple logistic regression analysis that identifies the predictors of medication non-adherence among type 2 diabetic patients. Patients from the Southern region were significantly more likely to be non-adherent compared to those from the Eastern region (OR=6.59, p=0.001). Additionally, higher monthly income was associated with better adherence (OR=1.40, p=0.021), indicating that patients with greater financial resources were more likely to follow their prescribed medication regimens. Regarding occupation, non-medical field occupations were linked to higher odds of non-adherence (OR=1.13, p=0.049. Patients living in downtown areas had a lower likelihood of non-adherence (OR=0.71, p=0.038). Furthermore, age at onset of diabetes was a significant factor, with older patients or those diagnosed later in life being less likely to be non-adherent (OR=0.98, p=0.009). The presence of diabetic complications was another significant predictor of non-adherence, as patients with complications were more likely to be non-adherent (OR=1.44, p=0.047). A key finding was the lack of physician-provided information about anti-diabetic medications, which significantly increased the odds of non-adherence (OR=2.15, p=0.005). More frequent doctor visits were associated with better adherence (OR=1.53, p=0.001). Finally, infrequent healthcare access was strongly associated with non-adherence (OR=2.34, p=0.003).

Table 8 Multiple Logistic Regression Model for Predictors of Type 2 Diabetic Patients Medications Non-adherence

Discussion

Medication adherence is defined by the Food and Drug Administration (FDA) as the extent to which patients take their medications as prescribed, in agreement with their healthcare provider. Optimal adherence to diabetes medication and lifestyle modifications has been shown to reduce emergency room visits, hospitalizations,¹⁹ and diabetes-related complications.²⁰ There are several factors that contribute to medication adherence in patients with DM, including family support, routine, side effects, complexity of the medication regimen, blood sugar levels, and forgetfulness. Additionally, one of the major factors influencing adherence is the relationship between healthcare professionals and their patients.⁹

Our study found that adherence to medication and refills was 52.5%, as measured by ARMS. In comparison, a previous study conducted in the Al Hasa region of Saudi Arabia in 2012 reported a non-compliance rate of 67.9%,⁹ which is significantly higher than our findings. The difference in adherence rates may be influenced by the passage of time and the impact of the COVID-19 pandemic on people’s behavior regarding medication adherence and refills. Nonetheless, the medication adherence rate observed in our study was satisfactory and higher than those reported in studies conducted in the Eastern Province of Saudi Arabia, Egypt, Switzerland, and Ethiopia, where adherence rates were 34.7%, 38.9%, 40%, and 51.3%, respectively.^21–24 However, it was lower than the rates observed in Tabuk²⁵ (76.4%). These variations in adherence levels could be attributed to several factors, including the availability of free medical services, varying levels of awareness regarding the importance of medication adherence, regional strategies to promote adherence, and differences in the measurement tools used across studies.²⁵

In this study, significant factors influencing medication adherence and refills included income level, marital status, occupation, and residency area. Similarly, a prior study conducted in Ethiopia identified financial difficulties as the primary external barrier to medication adherence.²⁴ Despite the availability of free healthcare in Saudi Arabia, issues such as transportation challenges⁹ and difficulties in scheduling clinic visits still impact adherence. Marital status also emerged as a significant factor, with a study by Alfulayw et al identifying a significant association between marital status and medication adherence.²¹ However, a study conducted in Al-Ahsa, Saudi Arabia, found no statistically significant relationship between marital status and adherence.⁹ Regarding residency, several studies have reported significant differences between rural and urban populations, with rural residents exhibiting higher adherence and refill rates.^9,26 However, a study by Alfulayw et al did not observe a significant association between place of residence and medication adherence.²¹ Consistent with our findings, occupation was also identified as a significant factor influencing adherence in the Saudi population, as reported by Alfulayw et al.²¹

This study found that age was not a significant factor in medication adherence and refills. However, earlier studies have suggested that adherence tends to increase with age.^21,23 On the other hand, Shams et al reported lower adherence among elderly patients, with good adherence rates of 28.1% in the elderly, 36.2% in middle-aged adults, and 51.8% in younger individuals.²² Gender was not significantly associated with medication adherence in our study, a finding consistent with several previous reports.^21,22,26 However, some studies have suggested that women are more likely to adhere to treatment regimens,^9,27 while others have reported higher adherence among men.^23,28 Additionally, our study found no significant association between educational level and adherence, which aligns with the findings of multiple prior studies.^21,26

Our study found a non-statistically significant association between smoking status and adherence to medications and refills. This result contrasts with findings from a study conducted in Shiraz, Iran, which reported a significant relationship between smoking and medication adherence.²⁹ Similarly, no significant association was observed between physical activity and medication adherence in our study. However, a study from Eastern Nepal demonstrated that daily jogging was significantly associated with improved adherence,³⁰ supporting the hypothesis that regular physical activity may enhance mood and motivation, thereby promoting better adherence—particularly among individuals following a structured exercise regimen.²¹ In terms of diabetes information, factors such as the age of onset, having diabetic complications, lower HbA1c levels, daily blood glucose level (BGL) checks, other comorbidities, receiving sufficient information from physicians about antidiabetic medications, the frequency of annual doctor visits, and the continuity of healthcare have a statistically significant relationship with adherence to medications and refills. For instance, a study conducted in Johor, Malaysia, found that medication adherence was significantly associated with good glycemic control.³¹ Similarly, a study in Shiraz, Iran, reported the same findings.²⁹ Additionally, a study in the Kurdistan Region of Iraq indicated that non-adherent participants were more likely to have higher HbA1c levels, reflecting poorer glycemic control.³² Nonadherence to medications and refills has been attributed to factors such as polypharmacy, financial constraints, or concerns about side effects.³³ Participants who were prescribed a glucometer for self-monitoring of BGLs experienced positive outcomes, such as lower BGLs.³⁴ Furthermore, a good relationship with the physician helped reduce concerns and issues related to medications, leading to more frequent visits to the same physician for regular check-ups, thereby ensuring continuity of care.^35–37

Our study found a significant association between the presence of chronic diseases and medication adherence and refill rates, whereas the number of medications was not statistically associated with adherence. In contrast, a study conducted at a tertiary hospital in Saudi Arabia involving 8932 adults with DM found that approximately 78% experienced polypharmacy, with a higher prevalence observed among women and patients with multiple comorbid conditions, including cardiovascular disease, chronic kidney disease, musculoskeletal disorders, respiratory issues, and mental health conditions.³⁸ Due to the numerous complications associated with DM, patients often require more medications, and factors such as the cost of polypharmacy, pill burden, and drug side effects contribute to polypharmacy, ultimately worsening the patient’s condition.³⁹

Limitations

Our study has several limitations, including recall bias, as participants may not accurately remember past treatments, and non-response bias, which may affect data accuracy. Additionally, selection bias is a concern since all participants were recruited from governmental PHCs offering free healthcare, excluding the private sector, which may impact the external validity and generalizability of the findings. The cross-sectional design prevents establishing causal relationships, requiring longitudinal studies for better insight into temporal associations. Furthermore, as the study was questionnaire-based, it is subject to self-reporting bias, which may lead to overestimation or underestimation of adherence behaviors.

Conclusion

This study highlights that nearly half of patients with T2DM in Al-Ahsa were non-adherent to their prescribed medications and refills, as measured by the ARMS, despite the availability of free healthcare services. The findings emphasize the multifactorial nature of medication adherence, with significant associations observed across sociodemographic, clinical, and healthcare-related variables. Key factors influencing adherence included income level, marital status, occupation, residency area, diabetic complications, glycemic control, frequency of blood glucose monitoring, and continuity of physician care. Importantly, multivariable analysis identified residence in the Southern region, presence of complications, lack of physician counseling, and limited access to healthcare as strong predictors of non-adherence.

The study underscores the need for targeted, patient-centered interventions aimed at improving medication adherence, especially among high-risk groups. Educational initiatives, enhanced communication between patients and healthcare providers, and better continuity of care are essential to address barriers to adherence. Future research should further explore behavioral and psychosocial determinants of adherence and assess the effectiveness of tailored interventions in improving long-term treatment outcomes for diabetic patients in different regions of Saudi Arabia.

Data Sharing Statement

All relevant data supporting the findings of this study are included within the manuscript.

Acknowledgment

The authors would like to express their sincere gratitude to all the patients who participated in this study.

Disclosure

The authors declare no conflicts of interest in this work.

References

1. Antar SA, Ashour NA, Sharaky M, et al. Diabetes mellitus: classification, mediators, and complications; A gate to identify potential targets for the development of new effective treatments. Biomed Pharmacother. 2023;168:115734. doi:10.1016/j.biopha.2023.115734

2. Tan SY, Mei Wong JL, Sim YJ, et al. Type 1 and 2 diabetes mellitus: a review on current treatment approach and gene therapy as potential intervention. Diabetes Metab Syndr. 2019;13(1):364–372. doi:10.1016/j.dsx.2018.10.008

3. Fletcher B, Gulanick M, Lamendola C. Risk factors for type 2 diabetes mellitus. J Cardiovasc Nurs. 2002;16(2):17–23. doi:10.1097/00005082-200201000-00003

4. IDF diabetes atlas. 2021. IDF Diabetes Atlas. Available from: https://diabetesatlas.org/atlas/tenth-edition/. Accessed February 27, 2024.

5. Jarrar M, Abusalah MAH, Albaker W, et al. Prevalence of type 2 diabetes mellitus in the general population of Saudi Arabia, 2000-2020: a systematic review and meta-analysis of observational studies. Saudi J Med Med Sci. 2023;11(1):1–10. doi:10.4103/sjmms.sjmms_394_22

6. Chandrasekaran P, Weiskirchen R. The role of obesity in type 2 diabetes mellitus-an overview. Int J Mol Sci. 2024;25(3):1882. doi:10.3390/ijms25031882

7. Alnaim MM, Alrsaheed A, Alkhateeb AA, Aljaafari MM, Alismail A. Prevalence of microvascular complications among patients with type 2 diabetes mellitus who visited diabetes clinics in Saudi Arabia. Saudi Med J. 2023;44(2):211–217. doi:10.15537/smj.2023.44.2.20220719

8. Ahmad MS, Alslamah T, Alannaz SM, et al. Prevalence of micro and macro vascular complications and their risk factors in type 2 diabetes in Saudi Arabian population: an analysis from SHIS. Eur Rev Med Pharmacol Sci. 2021;25(12):4308–4316. doi:10.26355/eurrev_202106_26137

9. Khan A, Al-Abdul Lateef Z, Al Aithan M, Bu-Khamseen M, Al Ibrahim I, Khan S. Factors contributing to non-compliance among diabetics attending primary health centers in the Al Hasa district of Saudi Arabia. J Family Community Med. 2012;19(1):26. doi:10.4103/2230-8229.94008

10. Kleinsinger F. The Unmet Challenge of Medication Nonadherence. Perm J. 2018;22:18–033. doi:10.7812/TPP/18-033

11. Mitiku Y, Belayneh A, Tegegne BA, et al. Prevalence of medication non-adherence and associated factors among diabetic patients in a tertiary hospital at Debre Markos, Northwest Ethiopia. Ethiop J Health Sci. 2022;32(4):755–764. doi:10.4314/ejhs.v32i4.12

12. Khattab MS, Aboifotouh MA, Khan MY, Humaidi MA, al-Kaldi YM. Compliance and control of diabetes in a family practice setting, Saudi Arabia. East Mediterr Health J. 1999;5(4):755–765. doi:10.26719/1999.5.4.755

13. Kripalani S, Risser J, Gatti ME, Jacobson TA. Development and evaluation of the adherence to refills and medications scale (ARMS) among low-literacy patients with chronic disease. Value Health. 2009;12(1):118–123. doi:10.1111/j.1524-4733.2008.00400.x

14. Gökdoğan F, Kes D. Validity and reliability of the Turkish adherence to refills and medications scale. Int J Nurs Pract. 2017;23(5). doi:10.1111/ijn.12566

15. Kim CJ, Park E, Schlenk EA, Kim M, Kim DJ. Psychometric evaluation of a Korean version of the Adherence to Refills and Medications Scale (ARMS) in adults with type 2 diabetes. Diab Educ. 2016;42(2):188–198. doi:10.1177/0145721716632062

16. Chen YJ, Chang J, Yang SY. Psychometric evaluation of Chinese version of Adherence to Refills and Medications Scale (ARMS) and blood-pressure control among elderly with hypertension. Patient Prefer Adherence. 2020;14:213–220. doi:10.2147/PPA.S236268

17. Lomper K, Chabowski M, Chudiak A, Białoszewski A, Dudek K, Jankowska-Polańska B. Psychometric evaluation of the Polish version of the adherence to refills and medications scale (ARMS) in adults with hypertension. Patient Prefer Adherence. 2018;12:2661–2670. doi:10.2147/PPA.S185305

18. Alammari G, Alhazzani H, AlRajhi N, et al. Validation of an Arabic version of the adherence to refills and medications scale (ARMS). Healthcare. 2021;9(11):1430. doi:10.3390/healthcare9111430

19. Roebuck MC, Liberman JN, Gemmill-Toyama M, Brennan TA. Medication adherence leads to lower health care use and costs despite increased drug spending. Health Aff. 2011;30(1):91–99. doi:10.1377/hlthaff.2009.1087

20. Shams N, Amjad S, Kumar N, Ahmed W, Saleem F. Drug non-adherence in type 2 diabetes mellitus; predictors and associations. J Ayub Med Coll Abbottabad. 2016;28(2):302–307.

21. Alfulayw MR, Almansour RA, Aljamri SK, et al. Factors contributing to noncompliance with diabetic medications and lifestyle modifications in patients with type 2 diabetes mellitus in the eastern province of Saudi Arabia: a cross-sectional study. Cureus. 2022;14(11):e31965. doi:10.7759/cureus.31965

22. Shams ME, Barakat EA. Measuring the rate of therapeutic adherence among outpatients with T2DM in Egypt. Saudi Pharm J. 2010;18(4):225–232. doi:10.1016/j.jsps.2010.07.004

23. Huber CA, Reich O. Medication adherence in patients with diabetes mellitus: does physician drug dispensing enhance quality of care? Evidence from a large health claims database in Switzerland. Patient Prefer Adherence. 2016;10:1803–1809. doi:10.2147/PPA.S115425

24. Wabe N, Angamo M, Hussein S. Medication adherence in diabetes mellitus and self management practices among type-2 diabetics in Ethiopia. North Am J Med Sci. 2011;418–423. doi:10.4297/najms.2011.3418

25. Alanazi M, Alatawi AM. Adherence to diabetes mellitus treatment regimen among patients with diabetes in the Tabuk Region of Saudi Arabia. Cureus. 2022;14(10):e30688. doi:10.7759/cureus.30688

26. Alhazmi T, Sharahili J, Khurmi S, et al. Drug compliance among type 2 diabetic patients in Jazan region, Saudi Arabia. Int J Adv Res. 2017b;5(1):966–974. doi:10.21474/ijar01/2838

27. Elsous A, Radwan M, Al-Sharif H, et al. Medications adherence and associated factors among patients with type 2 diabetes mellitus in the Gaza Strip, Palestine. Front Endocrinol. 2017;8. doi:10.3389/fendo.2017.00100

28. Xu N, Xie S, Chen Y, Li J, Sun L. Factors influencing medication non-adherence among Chinese older adults with diabetes mellitus. Int J Environ Res Public Health. 2020;17(17):6012. doi:10.3390/ijerph17176012

29. Mirahmadizadeh A, Khorshidsavar H, Seif M, Sharifi MH. Adherence to medication, diet and physical activity and the associated factors amongst patients with type 2 diabetes. Diabetes Ther. 2020;11(2):479–494. doi:10.1007/s13300-019-00750-8

30. Sarraf DP, Gupta PP. A hospital-based assessment of glycemic control and medication adherence in type 2 diabetes mellitus in Eastern Nepal. J Family Med Prim Care. 2023;12(6):1190–1196. doi:10.4103/jfmpc.jfmpc_90_23

31. Chin SS, Lau SW, Lim PL, Wong CM, Ujang N. Medication adherence, its associated factors and implication on glycaemic control in patients with type 2 diabetes mellitus: a cross-sectional study in a Malaysian primary care clinic. Malays Fam Physician. 2023;18:14. doi:10.51866/oa.88

32. Hamalaw S, Hama Salih A, Weli S. Non-adherence to anti-diabetic prescriptions among type 2 diabetes mellitus patients in the Kurdistan Region of Iraq. Cureus. 2024;16(5):e60572. doi:10.7759/cureus.60572

33. Baryakova TH, Pogostin BH, Langer R, McHugh KJ. Overcoming barriers to patient adherence: the case for developing innovative drug delivery systems. Nat Rev Drug Discov. 2023;22:387–409. doi:10.1038/s41573-023-00670-0

34. Bukhsh A, Goh B-H, Zimbudzi E, et al. Type 2 diabetes patients’ perspectives, experiences, and barriers toward diabetes-related self-care: a qualitative study from Pakistan. Front Endocrinol. 2020;11:534873. doi:10.3389/fendo.2020.534873

35. Presley B, Groot W, Pavlova M. Pharmacy-led interventions to improve medication adherence among adults with diabetes: a systematic review and meta-analysis. Res Social Administrative Pharm. 2019;15(9):1057–1067. doi:10.1016/j.sapharm.2018.09.021

36. Mayberry LS, Berg CA, Ra G Jr, Wallston KA. Assessing helpful and harmful family and friend involvement in adults’ type 2 diabetes self-management. Patient Educ Couns. 2019;102(7):1380–1388. doi:10.1016/j.pec.2019.02.027

37. Tan C, Cheng K, Sum C, Shew J, Holydard E, Wang W. Perceptions of diabetes self-care management among older Singaporeans with type 2 diabetes: a qualitative study. J Nurs Res. 2018;26(4):242–249. doi:10.1097/jnr.0000000000000226

38. Alwhaibi M, Balkhi B, Alhawassi TM, et al. Polypharmacy among patients with diabetes: a cross-sectional retrospective study in a tertiary hospital in Saudi Arabia. BMJ Open. 2018;8(5):e020852. doi:10.1136/bmjopen-2017-020852

39. Kumari S, Kumar S, Jain S. Effects of polypharmacy in elderly diabetic patients: a review. Cureus. 2022;14(9):e29068. doi:10.7759/cureus.29068

Specialty

Please chooseI’m not a medical professional.Allergy and ImmunologyAnatomyAnesthesiologyBiostatisticsCardiac/Thoracic/Vascular SurgeryCardiologyCritical CareDentistryDermatologyDiabetes and EndocrinologyEmergency MedicineEpidemiology and Public HealthFamily MedicineForensic MedicineGastroenterologyGeneral PracticeGeneticsGeriatricsHealth PolicyHematologyHIV/AIDSHospital-based MedicineI’m not a medical professional.Infectious DiseaseIntegrative/Complementary MedicineInternal MedicineInternal Medicine-PediatricsMedical Education and SimulationMedical PhysicsMedical StudentNephrologyNeurological SurgeryNeurologyNuclear MedicineNutritionObstetrics and GynecologyOccupational HealthOncologyOphthalmologyOptometryOral MedicineOrthopaedicsOsteopathic MedicineOtolaryngologyPain ManagementPalliative CarePathologyPediatricsPediatric SurgeryPharmacologyPhysical Medicine and RehabilitationPlastic SurgeryPodiatryPreventive MedicinePsychiatryPsychologyPulmonologyRadiation OncologyRadiologyRheumatologySubstance Use and AddictionSurgeryTherapeuticsTraumaUrologyMiscellaneous

A study published in The Journals of Gerontology: Series A reports that a novel combination of brain stimulation and personalized coaching significantly increased physical activity in older adults and held steady for months. The results offer hopeful news for inactive older adults living in subsidized housing, who may experience several barriers to increased activity, including depression and a lack of motivation.

Regular physical activity is an extremely effective, safe, and modifiable means to improve health, especially in older adults. But more than 85% of adults aged 65 and above regularly fail to meet federal physical activity guidelines, and insufficient physical activity remains a global health issue.

In the randomized trial, researchers found that inactive older adults who received transcranial direct current stimulation (tDCS), a noninvasive technique that delivers low-level electrical currents to targeted areas of the brain, along with individualized behavioral coaching, increased their daily step count an average of 1,179 steps per day.

That’s more than twice the increase seen in those who received coaching and placebo stimulation. Importantly, this boost persisted for three months after the program ended, and adherence to both the stimulation and coaching was exceptionally high.

Participants in the tDCS group received 10 brief 20-minute sessions over two weeks, targeting the left dorsolateral prefrontal cortex, a region associated with motivation, planning, and goal-directed behavior. They also received a personalized behavioral program that continued for two months. The coaching, delivered by physical therapists, included regular phone check-ins, individualized step goals, and practical strategies – like marching during commercials or walking with friends – to help participants gradually increase their daily movement. Researchers tracked daily steps using Fitbits and continued monitoring participants for 12 weeks after the intervention ended.

Adherence was significant: 97% of tDCS sessions and 93% of coaching sessions were completed, with Fitbit usage remaining strong throughout the intervention. Even during the no-contact retention phase, many participants maintained increased activity, suggesting the behavioral changes had taken hold.

Beyond the physical gains, participants in the tDCS group also reported improved motivation and greater perceived walking ability. The study’s findings suggest that tDCS may enhance not just the drive to move but also the stickiness of behavior change, particularly when paired with accessible, goal-driven coaching.

While tDCS has been explored in clinical and lab settings to improve mood, memory, and motor function, this study is among the first to test its use to support real-world health behavior change in older adults, particularly those with limited resources.

Helping older adults build and maintain healthy habits is notoriously difficult, especially in underserved communities. This study provides early but exciting evidence that a short course of brain stimulation can ‘prime the pump’ – enhancing motivation and helping new behaviors stick – and is encouraging, especially given the setting. The program was delivered entirely within participants’ housing facilities, which removed barriers to access. That model could be a blueprint for future community-based interventions.”

On-Yee (Amy) Lo, PhD, assistant scientist II at the Hinda and Arthur Marcus Institute for Aging Research at Hebrew SeniorLife

The authors caution that larger trials are needed to confirm the findings and explore how tDCS might be used to amplify other types of behavioral health programs. They also note the importance of exploring how cognitive function, baseline activity level, and social support might affect outcomes.

Still, for a population at high risk for inactivity-related decline, this study offers hope – and a potential new approach – for getting and keeping older adults moving.

The researchers included Levi Ask; Melike Kahya, PT, PhD, assistant professor of physical therapy at High Point University; Thomas Travison, PhD, senior scientist at the Marcus Institute; Lewis Lipsitz, MD, director of the Marcus Institute and chief academic officer of the Irving and Edyth S. Usen and Family Chair in Medical Research, Hebrew SeniorLife; and Brad Manor, PhD, senior scientist at the Marcus Institute.

The study, Modulating Brain Activity to Improve Goal-directed Physical Activity in Older Adults: A Pilot Randomized Controlled Trial, is published in The Journals of Gerontology: Series A, Volume 80, Issue 6, June 2025, glaf039.

Introduction

Cancer ranks as a leading cause of death and an important barrier to increasing life expectancy in every country of the world, according to Global Cancer Statistics 2022.¹ With the diagnostic and curative level of cancer improved in recent years, the 5-year survival rate among cancer patients in China has increased to 40.5% from 30% 10 years ago.² Despite the improved survival rate, a few of cancer patients in China still live less after treatment as real-world experience confirmed. This therefore reminds us of the importance to suggest personalized therapy strategies for different patients with different clinical characteristics. The established prognostic factors were tumor, node, tumor-node-metastasis (TNM) stage, pathological type, and so on.^3,4 However, even the same cancer type patients with same stage may have distinct survival outcomes. It is critical to identify reliable biomarkers to predict patients’ prognosis and guide their individualized treatment.

There is growing evidence that systematic immune inflammation plays a part in the mechanism of tumor initiation, progression, and metastasis.^5,6 Though the concept of inflammation-based scores, such as the neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and lymphocyte-to-monocyte ratio (LMR) has been revealed as negative prognostic factors in various types of solid tumors,^7–9 no specific factors were recognized as reliable biomarkers. New available and non-invasive prognostic indicators were looked for. The systemic immune-inflammatory index (SII) is calculated based on peripheral blood neutrophils, platelets and lymphocytes. It is a novel indicator that can predict the clinical outcomes of cancer patients confirmed by a few of studies.^10,11

It is generally accepted that cancer patients with malnutrition have a lower tolerance to treatment, as well as worse prognosis and short life span. At the same time, nutritional status is also an essential part of the immune status of cancer patients.^12,13 The prognostic nutritional index (PNI), which is calculated based on the serum albumin and circulating peripheral blood lymphocyte count, has been used to assess the immunonutritional status of cancer patients.^14,15 It is now also used to predict the prognosis of various malignancies, including lung cancer,¹⁶ breast cancer¹⁷ and liver cancer.¹⁸ Our previous study has confirmed that PNI was an independent prognostic factor for gastric cancer.¹⁹

Most of the studies focused on the value of either SII or PNI, but a single marker may not precisely predict the prognosis of cancer patients. We conducted this study to evaluate the combined effects between PNI, SII and clinical outcomes in cancer patients.

Materials and Methods

Patients

Five hundred and eight cancer patients were enrolled from June 2013 to June 2022 in the Affiliated Kunshan Hospital of Jiangsu University. The following inclusion criteria were applied: histologically or cytologically confirmed stage I–IV cancer patients; more than 18 years old; the Eastern Cooperative Oncology Group (ECOG) activity status score of <2; the expected survival should more than 12 months. Patients with the second primary tumor or active concurrent infection as well as incomplete follow-up data were excluded. The study was conducted in accordance with the Declaration of Helsinki, and all the patients provided written informed consent. This observational study was reviewed and approved by the Institutional Review Board of Affiliated Kunshan Hospital of Jiangsu University (2013-03-020-H04).

Data Collection and Follow-up

The detailed clinical characteristics including age, sex, pathologic type, smoking history, body mass index (BMI), TNM stage (AJCC 8th ed., 2018), ECOG PS, peripheral blood count and liver function were got from the electronic medical record system, which was authentic and reliable. One of the researchers collected the survival time by phone contact, with a follow-up deadline of June 30, 2023. The PNI was calculated as albumin level (g/L) + 5×total lymphocyte count (10⁹ /L). The SII was defined as platelet × neutrophil/ lymphocyte counts. The AGR was calculated using the following equations: AGR = ALB/(total protein-ALB). The overall survival (OS) was defined as time from the date of diagnosis to the date of death or last contact. The data were double-checked.

Statistical Analysis

SPSS 16.0 software (SPSS, Chicago, IL, USA) was utilized to perform statistical analyses. The receiver operating characteristic (ROC) curves were carried out to get the optimal cutoff values for AGR, SII and PNI. Comparisons between groups were performed using chi-squared test. Survival analysis was performed using the Kaplan–Meier method and comparisons between survival curves were performed by the Log rank test. Univariate and multivariable analyses were investigated by the Cox proportional hazards regression model. The Cox proportional hazards model was also used to check proportional hazard assumption. The hazard ratio (HR) and 95% confidence interval (CI) were used to assess relative risks. Statistically significance was defined as p values (two sides) <0.05.

Results

Clinicopathologic Characteristics of the Patients

The baseline characteristics of the 508 patients enrolled in the study are summarized in Table 1. There were 239 males (47.05%) and 269 females (52.95%). The median age of the patient was 61 years old, ranging from 25 to 89, of which 191 (37.6%) were ≥65 years old. The most common cancer type was lung cancer (44.69%). 167 (32.88%) patients with stage I–II and 341 (67.12%) patients were diagnosed at stage III–IV. One hundred and seventy-three cases (34.06%) had a history of smoking. There were 68 patients with BMI < 18.5kg/m² (13.39%), 379 patients with 18.5 to 24.9 kg/m² (74.61%), and 61 patients with BMI > 24.9 kg/m² (12.0%). Two hundred and eighty-three (55.71%) patients had PS score of 0 and 225 (44.29%) had PS score of 1.

Table 1 Association of the Patients’ Characteristics with the SII and PNI

Relationships Between SII, PNI and Clinicopathological Features

The optimal SII and PNI cutoff values were analyzed by ROC curves for the OS of patients. According to the ROC curve and the Youden index, the ideal preoperative PNI and SII cutoff values were 792.0 (Youden index is 0.214) and 49.825 (Youden index is 0.356), respectively. The SII level before treatment was elevated in 149 (29.33%) patients and a total of 238 (46.85%) patients had lower PNI levels. As shown in Table 1, increased SII level was significantly associated with smoking history (p = 0.02), ECGO PS (p = 0.004) and AGR level (p < 0.001). The high PNI and low PNI groups showed significant differences in gender, age, smoking history, BMI group, cancer type, TNM stage, ECGO PS and AGR. Consider both of SII and PNI, smoking history, ECGO PS and AGR level may be the major related factors.

Univariate and Multivariable Analyses

We performed Cox regression analyses for OS. Table 2 demonstrated the univariate and multivariable analyses for OS. After multivariable analyses, the tumor stage of III/IV (p < 0.001), BMI<18.5kg/m2 (p = 0.042), high SII (p = 0.001, Figure 1) and low AGR (p = 0.047, Figure 2) were independently negative prognostic markers for OS.

Table 2 Univariate and Multivariate Analyses of Factors for the Prediction of Overall Survival

Figure 1 Kaplan-Meier survival curves of overall survival according to SII (p=0.001).

Figure 2 Kaplan-Meier survival curves of overall survival according to AGR (p=0.047).

The patients were also divided into four groups based on both the SII and PNI levels: high SII and low PNI (n = 101); high SII and high PNI (n = 48); low SII and high PNI (n = 222); low SII and low PNI (n = 137). At the last follow-up in this study, 377 patients (74.21%) were still alive. The median OS for all patients were 24 months. We performed joint analysis and showed in Figure 3. The results presented that both high SII and low PNI group had the worst prognosis (p < 0.001). The median OS of high SII and low PNI group was 21 months.

Figure 3 Kaplan-Meier survival curves of overall survival according to both SII and PNI (p<0.001).

Discussion

Despite significant advancements in cancer treatment in recent years, not all patients benefit equally, primarily due to variations in their baseline health status, such as nutrition status and inflammatory conditions. Many blood-derived markers were applied for their cost-effectiveness and prognostic reliability. A great number of research studies have found that the PNI and SII play an important role in the cancer development and prognosis.^10,20–22 Our results have added evidence that SII is an independent influencing factor of overall survival. Moreover, the joint analysis showed both high SII and low PNI had the lowest OS rate. This adds to the growing body of evidence supporting the utility of SII and PNI as prognostic markers in cancer patients, underscoring the importance of considering both inflammatory and nutritional status in prognostic prediction models.

SII derives from peripheral lymphocyte, neutrophil and platelet counts, which could provide a comprehensive reflection of the local immune status and systemic inflammation in the whole body at the same time.²³ SII alone has been proven to predict prognosis of various malignant tumors.^24–26 PNI, a simple and feasible nutritional factor, has been used to assess the immunonutritional status of cancer patients.^27,28 However, single factor may not reflect the complicated mechanism of tumor micro-microenvironment. More and more studies have focused on the joint predictive value of SII and PNI. Fan et al evaluated the value of SII combined with PNI to predict outcomes in non-small cell lung cancer (NSCLC) patients treated with platinum-doublet chemotherapy, and they found that patients with a higher SII-PNI score had a worse prognosis.²⁹ Another prospective study showed lower SII-PNI scores were associated with better efficacy of chemotherapy combined with immunotherapy in patients with locally advanced gastric cancer.¹⁰ Yang et al developed a nomogram that incorporated the PIIN score, which includes SII and PNI, for predicting overall survival in postoperative pancreatic cancer patients.³⁰ These studies collectively demonstrate the significance of PNI and SII in cancer development and prognosis, highlighting their potential as valuable biomarkers in clinical practice. The results could be explained that chronic inflammation associated with malnutrition could paradoxically suppress activation of the adaptive immune system, which is a vicious cycle.³¹ We could also tell that the patients with severe inflammatory reaction and poor nutritional condition had poor response to treatment and the overall survival was short. As an accessible, simple, and cost-effective marker derived from blood tests, preliminary evidence supports the potential clinical utility of PNI and SII.

Going forward, it will be important to comprehend the interaction of nutritional status, inflammatory reaction and cancer survival. Malnutrition usually occurs in patients with malignant tumors and gradually leads to cachexia.³² The incidence of cachexia is particularly high in patients with tumors and lung cancer. Patients with digestive tract tumors are naturally more prone to malnutrition and even cachexia due to the decline or loss of their own digestion and absorption function, coupled with the serious depletion of the body’s nutrient reserves by cancer.³³ Cytokines secreted by tumors are one of the causes of cachexia.³⁴ These cytokines including IL-6, TGF-β and heat shock proteins (HSPs) which directly causes the catabolism and metabolism of the target tissue. Some studies have confirmed that nutritional intervention can improve the quality of life of patients with cachexia, and even prolong the survival of patients.^35,36 Our findings revealed that patients with cancers and low BMI (<18.5) have short overall survival. In clinical practice, individualized nutrition intervention can effectively improve the nutritional status, life quality and the survival prognosis of locally advanced carcinoma patients.

A few limitations of current study also should be explained. First of all, this retrospective analysis was conducted in a single center. It means the sample size is relative small and selection bias is inevitable Another aspect should be pointed out was that we assessed only the pretreatment level of these factors but did not focus on the dynamic change of them. However, the serum levels of these factors were easily affected by nutritional status and side effects of chemotherapy and radiotherapy. Moreover, other factors related to nutrition, inflammation, and immunity, such as weight, waist-to-hip ratio, C-reactive protein (CRP), procalcitonin, even treatments were not included in the final analysis. Though we have tried our best to minimize the risk of bias and unmeasured confounders by applying strict inclusion criteria (complete medical records) and dual-data verification, further research should aim to address these limitations and explore the full potential of SII and PNI as valuable biomarkers in clinical practice.

Conclusion

In conclusion, cancer patients with both high SII and low PNI had poor survival outcome. Pretreatment level of SII may be an independent prognostic factor for cancer patients.

Funding

This work was supported by the National Natural Science Foundation (Grant numbers: 82403069), Jiangsu Province Natural Science Foundation (Grant numbers: BK20240488), Kushan Science and Technology project (KS2207), Suzhou Science and Technology project (SLT2023020).

Disclosure

The authors report no competing interests for this work.

References

1. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024;74(3):229–263. doi:10.3322/caac.21834

2. Maomao C, He L, Dianqin S, et al. Current cancer burden in China: epidemiology, etiology, and prevention. Cancer Biol Med. 2022;19(8):1121–1138. doi:10.20892/j.issn.2095-3941.2022.0231

3. Benetatos N, Hodson J, Marudanayagam R, et al. Prognostic factors and survival after surgical resection of pancreatic neuroendocrine tumor with validation of established and modified staging systems. Hepatobiliary Pancreatic Dis Int. 2018;17(2):169–175. doi:10.1016/j.hbpd.2018.03.002

4. Wiksten JP, Lundin J, Nordling S, Kokkola A, Haglund C. Comparison of the prognostic value of a panel of tissue tumor markers and established clinicopathological factors in patients with gastric cancer. Anticancer Res. 2008;28(4C):2279–2287.

5. Zhao H, Wu L, Yan G, et al. Inflammation and tumor progression: signaling pathways and targeted intervention. Signal Transduct Target Ther. 2021;6(1):263. doi:10.1038/s41392-021-00658-5

6. Gonzalez H, Hagerling C, Werb Z. Roles of the immune system in cancer: from tumor initiation to metastatic progression. Genes Dev. 2018;32(19–20):1267–1284. doi:10.1101/gad.314617.118

7. Zhang Y, Lu JJ, Du YP, Feng CX, Wang LQ, Chen MB. Prognostic value of neutrophil-to-lymphocyte ratio and platelet-to-lymphocyte ratio in gastric cancer. Medicine. 2018;97(12):e0144. doi:10.1097/MD.0000000000010144

8. Kim JY, Jung EJ, Kim JM, et al. Dynamic changes of neutrophil-to-lymphocyte ratio and platelet-to-lymphocyte ratio predicts breast cancer prognosis. BMC Cancer. 2020;20(1):1206. doi:10.1186/s12885-020-07700-9

9. Kim D, Bae SJ, Ahn SG, et al. RT-induced dynamic changes in the lymphocyte-to-monocyte ratio in patients with breast cancer indicate poor prognosis. Breast Cancer Res Treat. 2022;193(3):637–647. doi:10.1007/s10549-022-06601-8

10. Ding P, Guo H, Sun C, et al. Combined systemic immune-inflammatory index (SII) and prognostic nutritional index (PNI) predicts chemotherapy response and prognosis in locally advanced gastric cancer patients receiving neoadjuvant chemotherapy with PD-1 antibody sintilimab and XELOX: a prospective study. BMC Gastroenterol. 2022;22(1):121. doi:10.1186/s12876-022-02199-9

11. Huang W, Luo J, Wen J, Jiang M. The relationship between systemic immune inflammatory index and prognosis of patients with non-small cell lung cancer: a meta-analysis and systematic review. Front Surg. 2022;9:898304. doi:10.3389/fsurg.2022.898304

12. Bossi P, De Luca R, Ciani O, D’Angelo E, Caccialanza R. Malnutrition management in oncology: an expert view on controversial issues and future perspectives. Front Oncol. 2022;12:910770. doi:10.3389/fonc.2022.910770

13. Rovesti G, Valoriani F, Rimini M, et al. Clinical implications of malnutrition in the management of patients with pancreatic cancer: introducing the concept of the nutritional oncology board. Nutrients. 2021;13(10):3522. doi:10.3390/nu13103522

14. Keskinkilic M, Semiz HS, Ataca E, Yavuzsen T. The prognostic value of immune-nutritional status in metastatic colorectal cancer: Prognostic Nutritional Index (PNI). Support Care Cancer. 2024;32(6):374. doi:10.1007/s00520-024-08572-6

15. Wang N, Xi W, Lu S, et al. A novel inflammatory-nutritional prognostic scoring system for stage III gastric cancer patients with radical gastrectomy followed by adjuvant chemotherapy. Front Oncol. 2021;11:650562. doi:10.3389/fonc.2021.650562

16. Matsubara T, Hirai F, Yamaguchi M, Hamatake M. Immunonutritional indices in non-small-cell lung cancer patients receiving adjuvant platinum-based chemotherapy. Anticancer Res. 2021;41(10):5157–5163. doi:10.21873/anticanres.15333

17. Qu F, Luo Y, Peng Y, et al. Construction and validation of a prognostic nutritional index-based nomogram for predicting pathological complete response in breast cancer: a two-center study of 1,170 patients. Front Immunol. 2023;14:1335546. doi:10.3389/fimmu.2023.1335546

18. Liu C, Zhao H, Zhang R, Guo Z, Wang P, Qu Z. Prognostic value of nutritional and inflammatory markers in patients with hepatocellular carcinoma who receive immune checkpoint inhibitors. Oncol Lett. 2023;26(4):437. doi:10.3892/ol.2023.14024

19. Zhang Y, Zhu JY, Zhou LN, Tang M, Chen MB, Tao M. Predicting the prognosis of gastric cancer by albumin/globulin ratio and the prognostic nutritional index. Nutr Cancer. 2020;72(4):635–644. doi:10.1080/01635581.2019.1651347

20. Lv X, Xu B, Zou Q, Han S, Feng Y. Clinical application of common inflammatory and nutritional indicators before treatment in prognosis evaluation of non-small cell lung cancer: a retrospective real-world study. Front Med. 2023;10:1183886. doi:10.3389/fmed.2023.1183886

21. Atasever Akkas E, Erdis E, Yucel B. Prognostic value of the systemic immune-inflammation index, systemic inflammation response index, and prognostic nutritional index in head and neck cancer. Eur Arch Otorhinolaryngol. 2023;280(8):3821–3830. doi:10.1007/s00405-023-07954-6

22. Yi J, Xue J, Yang L, Xia L, He W. Predictive value of prognostic nutritional and systemic immune-inflammation indices for patients with microsatellite instability-high metastatic colorectal cancer receiving immunotherapy. Front Nutr. 2023;10:1094189. doi:10.3389/fnut.2023.1094189

23. Huang Y, Chen Y, Zhu Y, et al. Postoperative Systemic Immune-Inflammation Index (SII): a superior prognostic factor of endometrial cancer. Front Surg. 2021;8:704235. doi:10.3389/fsurg.2021.704235

24. Tian BW, Yang YF, Yang CC, et al. Systemic immune-inflammation index predicts prognosis of cancer immunotherapy: systemic review and meta-analysis. Immunotherapy. 2022;14(18):1481–1496. doi:10.2217/imt-2022-0133

25. Yan X, Li G. Preoperative systemic immune-inflammation index predicts prognosis and guides clinical treatment in patients with non-small cell lung cancer. Biosci Rep. 2020;40(3). doi:10.1042/BSR20200352

26. Yang X, Wu C. Systemic immune inflammation index and gastric cancer prognosis: a systematic review and meta‑analysis. Exp Ther Med. 2024;27(3):122. doi:10.3892/etm.2024.12410

27. Ishiguro T, Aoyama T, Ju M, et al. Prognostic nutritional index as a predictor of prognosis in postoperative patients with gastric cancer. In Vivo. 2023;37(3):1290–1296. doi:10.21873/invivo.13207

28. Nagashima Y, Funahashi K, Kagami S, et al. Which preoperative immunonutritional index best predicts postoperative mortality after palliative surgery for malignant bowel obstruction in patients with late-stage cancer? A single-center study in Japan comparing the modified Glasgow prognostic score (mGPS), the prognostic nutritional index (PNI), and the controlling nutritional status (CONUT). Surg Today. 2023;53(1):22–30. doi:10.1007/s00595-022-02534-3

29. Fan R, Chen Y, Xu G, Pan W, Lv Y, Zhang Z. Combined systemic immune-inflammatory index and prognostic nutritional index predict outcomes in advanced non-small cell lung cancer patients receiving platinum-doublet chemotherapy. Front Oncol. 2023;13:996312. doi:10.3389/fonc.2023.996312

30. Yang J, Zhou H, Li H, Zhao F, Tong K. Nomogram incorporating prognostic immune-inflammatory-nutritional score for survival prediction in pancreatic cancer: a retrospective study. BMC Cancer. 2024;24(1):193. doi:10.1186/s12885-024-11948-w

31. Arihara Y, Takada K, Murase K, et al. Inflammation and malnutrition as markers of poor outcomes in head and neck cancer patients treated with nivolumab. Acta Otolaryngol. 2023;143(8):714–720. doi:10.1080/00016489.2023.2240372

32. Arends J. Malnutrition in cancer patients: causes, consequences and treatment options. Eur J Surg Oncol. 2024;50(5):107074. doi:10.1016/j.ejso.2023.107074

33. Gliwska E, Glabska D, Zaczek Z, Sobocki J, Guzek D. Influence of enteral nutrition on quality of life in head and neck cancer and upper gastrointestinal tract cancer patients within a pair-matched sample. Nutrients. 2023;15(21):4698. doi:10.3390/nu15214698

34. Kwon YY, Hui S. IL-6 is dispensable for causing cachexia in the colon carcinoma 26 model. bioRxiv. 2023.

35. Gomez-Serna MI, Lopez D, Perez-Garcia YE, Montoya Restrepo ME. [Nutritional assessment of cancer patients in palliative care is a key element in comprehensive care and survival]. Nutr Hosp. 2022;39(4):814–823. Czech. doi:10.20960/nh.03828

36. Sugiyama K, Shiraishi K, Motohashi T, et al. The impact of nutritional support on survival outcomes in patients with advanced gastric adenocarcinoma treated with chemotherapy. Nutr Cancer. 2023;75(3):867–875. doi:10.1080/01635581.2022.2162090

Introduction

Diabetes distress (DD) refers to the negative emotional impact of living with diabetes, including feelings of guilt, anxiety, and concerns about the self-management of the condition.¹ A previous study established DD as a clinically significant risk factor for suboptimal health outcomes in patients with diabetes.² Previous studies have demonstrated that elevated DD is associated with biological markers, including higher HbA1c³ and lower heart rate variability.⁴ Additionally, there is evidence that DD is associated with a higher mortality rate^5,6 and that elevated DD is associated with delayed medical care,⁷ impaired diabetes self-management,^8–10 and lower quality of life.¹¹ While DD is known to complicate diabetes management, its connection to DS and SS is unclear.

For the past few decades, researchers^12,13 have focused on the association between DD and DS, which frequently occur together.¹⁴ According to survey results, 19.6% of adults with diabetes have experienced DD and DS.¹⁵ A longitudinal study demonstrated the persistent coexistence of DD and DS for 18 months.¹⁶ Gastrointestinal symptoms exhibited independent associations with DD and DS in individuals with type 2 diabetes (T2D).¹⁷ The coexistence of DD and DS increases the risk of death, poor disease management, diabetes-related complications, and a lower quality of life, which is a challenge to the care of patients with T2D.¹⁸ The American Diabetes Association and other researchers agree that routine screening for DD and DS should be performed in all adults with diabetes due to comorbidity, persistence over time, and impact on health outcomes.^19–21 Therefore, establishing a link between DD and DS is critical for developing effective interventions.²² Ehrmann et al demonstrated that higher DD predicted more DS 6 months later. Conversely, a higher DS at baseline indicated an increase in DD at the 6-month follow-up date.²³ Burns et al reported a bidirectional association between DD and DS in a follow-up study on a group of nearly 1700 patients with T2D living in the community.²⁴ This indicates that DD was associated with concurrent and subsequent DS, and DS, in turn, was associated with concurrent and subsequent DD. These studies demonstrate an intricate reciprocal association between DD and DS. However, the exact mechanisms of the interaction are unclear.

SS is another external factor closely associated with DD in individuals with T2D. SS is a multidimensional construct that refers to objective support, subjective support, and support utilization.²⁵ Previous studies have confirmed that SS buffers the impacts of DD on health-related quality of life.^26,27 A previous study has demonstrated the potential direct effects of SS in diabetes and reported that higher levels of SS were associated with lower DD, better adoption of diabetes self-management behaviors, and better diabetes-related clinical outcomes, including glycemic control.²⁸ Moreover, effective patient-centered communication has been indicated to buffer the effects of diabetes burden on distress levels, highlighting the importance of supportive interactions in diabetes care.²⁹ A previous study reported that perceived SS can alleviate feelings of distress, potentially reducing the risk of developing DS.³⁰ There could be a negative correlation between DD and SS. However, the mechanisms through which SS influences DD are unclear.

Previous studies on DD were primarily focused on its prevalence, instruments, and consequences.^31–33 Studies have investigated the association between DD and DS/SS, often utilizing traditional statistical methods, including regression or factor analysis.^34–36 While these methods effectively assess the association between specific predictive and outcome variables, they fail to capture the interdependencies and complex interactions among multiple variables.³⁷ This limitation is particularly pronounced when investigating complex phenomena, including DD in patients with T2D. Consequently, a more nuanced statistical approach is needed to investigate the association between them, including central and bridging symptoms, thereby enhancing the understanding of the complex psychopathological mechanisms associated with DD and DS/SS.

The Network Theory of Mental Disorder (NTMD) suggests that the development and maintenance of mental disorders are influenced by dynamic causal relationships among various symptoms within the disorder.³⁸ The network analysis, a cutting-edge approach for analyzing psychiatric disorders, aligns with the principles of NTMD and addresses this complexity by examining the correlation between specific symptoms.³⁹ This method elucidates the relationships among individual symptoms and, through the centrality metrics of the network, facilitates the identification of core and bridge symptoms, providing a more comprehensive perspective on exploring the connection between DD and DS/SS.

Incorporating emotional and social factors in diabetes management may lead to improved health outcomes and enhanced quality of life for patients with T2D.⁴⁰ Further exploration of these associations is essential, as understanding the dynamics of DD and DS/SS could inform more effective interventions for individuals with T2D. This study employed a network analysis method to construct a symptom network among DD and DS/SS to investigate their interactions, aiming to establish a theoretical foundation for future interventions by identifying critical nodes with cascading effects within the network.

Methods

Design

A cross-sectional design was employed in this research. Figure 1 illustrates the study flow chart.

Figure 1 The flowchart of the research.

Setting and Sample

The study was conducted at two diabetes centers in densely populated areas of southwest China, where the prevalence of T2D is among the highest in the country.⁴¹ One of the centers is within a large general hospital that provides outpatient and inpatient care for adults with diabetes. The other center is in a primary care facility that mainly provides outpatient care and home visits. The two centers serve patients with T2D in various medical settings in southwest China, including outpatients, inpatients, community patients, and home care patients, ensuring the representativeness of our T2D samples. The inclusion criteria for participants were as follows: (a) Patients diagnosed with T2D, (b) patients ≥ 18 years of age, and (c) patients who had an average score > 2 points on the Diabetes Distress Scale (DDS). The exclusion criteria were as follows: (a) Patients with a history of severe dementia, psychosis, or serious neurologic disease, and (b) patients refusing to participate in the study. We invited 912 patients with T2D from the two diabetes centers to participate, and 886 consented to enroll.

Variables and Measurements

Demographic and Clinical Information

The participants self-reported their information, including their age, gender, educational background, marital status, family history of diabetes, smoking, and alcohol consumption.

Diabetes Distress

Diabetes distress was assessed using the DDS, developed by Polonsky to evaluate the distress of patients with diabetes.⁴² Zhang et al⁴³ translated the scale into Chinese and reported that the Cronbach’s alpha for the overall scale was 0.88, while the subscales ranged from 0.76 to 0.81 in Chinese adults with T2D. The Chinese DDS comprises 17 items that measure four dimensions: Emotional burden (EB, five items), physician-related distress (PD, four items), regimen-related distress (RD, five items), and diabetes-related interpersonal distress (ID, three items). These items employ a six-point Likert scale that ranges from 1 (no distress) to 6 (high distress). A total score was calculated by adding the 17 items. The higher the scores, the more significant the distress. According to the revised rating system developed by Fisher, a mean item score < 2 indicates little or no distress; 2.0–2.9 indicates moderate distress, and ≥ 3 indicates high distress.

Depressive Symptoms

Depressive symptoms were assessed using the Patient Health Questionnaire (PHQ-9), a short questionnaire. The internal reliability of the PHQ-9 was excellent, with a Cronbach’s alpha of 0.870 among patients with T2D.⁴⁴ The scale consists of 9 questions with response options: including “no problem” (0 points), “a few days, sometimes” (1 point), “more than 7 days” (2 points), or “almost every day” (3 points). The total score is calculated by adding the points for each response, resulting in a score range of 0 to 27. Scores from 0 to 4 indicate the absence of DS, 5 to 9 indicate mild DS (subsyndromal depression), and ≥ 10 indicate a high probability of a depressive episode, which can be classified as moderate (10 to 14), moderately severe (15 to 19), and severe depression (20 and above).

Social Support

Social support was assessed using the Social Support Rating Scale (SSRS), designed for the Chinese population by Xiao.²⁵ SSRS comprises three dimensions: Objective support, subjective support, and utilization of support, and has been verified to have favorable reliability and validity in patients with T2D. Chen et al⁴⁵ indicated that the Cronbach’s alpha coefficient of the SSRS was 0.79. A higher score on the SSRS indicates better SS and comprehensively reflects an individual’s SS status.

Data Analysis

All analyses were performed using R software (Version 4.2.3). We described continuous variables as mean (standard deviation, SD), and presented categorical variables as frequencies and percentages.

Network Estimation

We computed polychoric correlations between all nodes to examine the edges of the network. We estimated the Graphical Gaussian Model (GGM) using the graphical least absolute shrinkage and selection operator.⁴⁶ This study aimed to estimate two network structures: The first was the network structure of DD, which will help us investigate its core symptoms; the second was the network structure of DD-DS-SS, which will help us identify the bridge symptoms between DS and DD, and between SS and DD. In the network model, each symptom is represented as a “node”, and the association between symptoms is defined as an “edge”.⁴⁷ Thicker edges represent stronger correlations between two nodes.

Centrality Estimation

The importance of each node in the item network of DD was quantified using the centrality of strength, which is the sum of the absolute value of the edge weights attached to a node for each node. The strength indicates the network connectivity used to identify the central nodes.⁴⁸ To investigate the interconnections between DS, SS, and DD, we categorized nodes into three distinct communities: The DS community (items from PHQ-9), the SS community (items from SSRS), and the DD community (items from DDS). The bridge expected influence (BEI) was calculated to identify bridge components. The BEI of a node is the sum of its edge weights from all other communities. A higher positive BEI indicates a greater activation capacity to other communities, while a higher negative BEI indicates a greater deactivation capacity to other communities.⁴⁹

Accuracy and Stability

The accuracy of the edge weights was confirmed by calculating 95% confidence intervals (CIs) for all edges using a nonparametric bootstrap approach with 500 bootstrap samples.⁵⁰ Additionally, the stability of the correlation (CS) coefficient for the strength/BEI was thoroughly assessed using a case-dropping subset bootstrap approach with 500 bootstrap samples. The CS coefficient must be greater than 0.25, ideally surpassing 0.5, to maintain the integrity and reliability of the results.

Ethics Approval and Consent to Participate

This study was approved by the Ethics Committee of the Chengdu Jinniu District People’s Hospital (QYYLL-2022-011), and all procedures followed relevant guidelines and regulations. Informed consent was obtained from all subjects. As stated on the information sheet in the questionnaire packet, consent to participate was obtained by participants returning a completed survey. Participants could decide whether or not to participate and could withdraw at any time without repercussions. This study was conducted in accordance with the ethical principles of the Declaration of Helsinki.

Results

Characteristics of the Participants

The final sample comprised 886 participants with T2D, ranging from 20 to 80 years at the time of assessment. There were 562 (62.4%) male and 324 (36.6%) female participants. Of the 886 participants, 519 (58.6%) reported a family history of diabetes, while 367 (41.4%) did not. More demographic details about the participants are presented in Table 1.

Table 1 Summary of Participants’ Characteristics (N = 886)

Score Results of DD, DS, and SS

The means and standard deviations of all variables in the network are presented in Table 2 as indicated by the statistical description results.

Table 2 Mean Scores and Standard Deviations for Items of DDS, PHQ-9, and SSRS

Structure of the DD Network

The structure of the DD network is depicted in Figure 2A. Centrality analysis was performed to examine the importance of each symptom within the DD network, with the results depicted in Figure 2B. Due to high intercorrelations and the more reliable estimation of strength centrality and closeness (the accuracy analyses below), we will focus our interpretation of the most relevant symptoms on node strength centrality for the rest of the report. The three nodes with the highest node strength centrality were PD4 (Do not have doctor I can see regularly), PD2 (Doctor does not give clear directions), and PD1 (Doctor does not know about diabetes).

Figure 2 Network structure of the DD (A) and centrality index of the DD network (B).

Structure of the DD-DS-SS Network

We estimated the network structures of the DD, DS, and SS. The resulting network is displayed in Figure 3. The nodes between DD and DS were positively connected within the network, and particularly strong connections were between DDS1 (diabetes taking up too much energy)-PHQ4 (tired or little energy), DDS13 (not sticking closely enough to meal plan)-PHQ5 (poor appetite/ overeating), DDS16 (Friends/family do not appreciate difficulty of diabetes)-PHQ2 (feeling down, depressed, or hopeless), and DDS17 (friends/family do not give emotional support)-PHQ6 (Failure). These three dimensions of SS were inversely related to DD, especially between DDS17-SSRS2 (subjective support) and DDS17-SSRS3 (support utilization).

Figure 3 Network structure of the DD-DS-SS.

Accuracy and Stability of the Two Networks

We assessed the accuracy and stability of the estimated networks. Figure 4 illustrates the accuracy of the bootstrap method in obtaining edge weights. The narrow confidence interval indicates that the edge weights possess sufficient accuracy. The subset bootstrap (Figure 5) indicates that the centrality of node strength and closeness had good stability, with a decrease in sample size. Meanwhile, coefficients of 0.7 signify adequate stability in centrality of strength and closeness.

Figure 4 Bootstrapped confidence intervals of the edge weights in the DD network (A) and DD-DS-SS network (B).

Figure 5 Subsetting bootstrap for DD network (A) and DD-DS-SS network (B).

Discussion

This is the first study to investigate the interconnections among components of DD and the correlations between DS, SS, and DD constructs in patients with T2D using network analysis, to the best of our knowledge. We performed a network analysis of DD to identify its core symptoms, followed by another analysis that included DS and SS to uncover key connections between them. The principal findings of this study were systematically delineated in Figure 6, which graphically elucidates the core symptoms of diabetes DD and its bridge symptoms between DS/SS. By assessing the stability and accuracy of these networks, we gained insights into the complex association between DD and DS/SS, which helped to provide a focus for the psychological care of people with T2D.

Figure 6 Summary of key findings.

The observed clustering pattern of DD items, illustrated in Figure 1, corresponds closely with the four subscales of DDS-17: Emotional burden, physician-related distress, regimen-related distress, and interpersonal distress.⁵¹ In the DD item network, the three nodes with the highest node strength centrality were PD4 (Do not have doctor I can see regularly), PD2 (Doctor does not give clear directions), and PD1 (Doctor does not know about diabetes). Highly central nodes in a cross-sectional network were indicated to predict the correlation between changes in one node and other network symptoms.⁵² A Canadian cross-sectional survey identified physician-related distress as a core symptom of DD.⁵³ The findings indicated that while diabetes management primarily falls on the patient, healthcare professionals play a crucial role. Previous studies indicate that the involvement of healthcare professionals—including doctors, nurses, and dietitians—enhances patient self-management and compliance and reduces the risk of complications, particularly cardiovascular ones.^54,55 Moreover, medical personnel are instrumental in setting individualized treatment goals and monitoring progress, which is essential for achieving optimal glycemic control.⁵⁶ Although there are several treatment options available, many patients struggle to manage their condition effectively due to factors including a lack of support and low health literacy.⁵⁷ The findings underscored the need for medical professionals to engage in open communication with their patients, to help them understand their condition and the importance of adherence to treatment plans.⁵⁸ This dependence on medical professionals has become the primary source of DD in T2D patients and an essential part of psychological care. Similar evidence was reported in other interventional studies. Psychological interventions provided by nursing staff,⁵⁹ integrating nurse counseling with mobile health technologies,⁶⁰ and nurse-administered mindfulness-based stress reduction programs⁶¹ have all demonstrated significant positive effects on self-efficacy, self-management capabilities, and DD in patients with T2D. A focus group interview revealed favorable responses from patients with T2D toward nurse-physician collaborative care, with participants expressing feelings of empowerment.⁶² Therefore, we recommend incorporating healthcare professional support into psychological interventions for patients with T2D to optimize disease management outcomes.

We observed the link paths between DD and DS/SS in the second network. We analyzed the more microscopic relationship between DD and DS as depicted in Figure 3. Although a previous study suggested that DD and DS overlap with each other,⁵³ the exact overlap is not fully reported. Through network analysis, this study found the exact part of DD and DS duplication, which is of significant help in understanding the differences and connections between DD and DS. A previous study suggested that the scientific debate about the overlap between DD and DS may stem from shared etiological pathways and symptoms,⁶³ and our study demonstrated that DDS1-PHQ4, DDS13-PHQ5, DDS16-PHQ2, and DDS17-PHQ6 have strong positive bridges in terms of their network structure. The DDS1 item addresses the energy expenditure associated with diabetes, while the PHQ4 item addresses the fatigue caused by the disease.⁶⁴ The two focused on the negative emotions associated with the long-term illness. It is therefore not difficult to understand that the diabetes management of DD-positive patients is generally poor. DDS13 and PHQ5 items were focused on understanding the impact of diabetes on the diet of patients.⁶⁵ Diet is a key modifiable factor in the management and prevention of T2D.⁶⁶ This result is consistent with a previous study in which DD and DS were independently associated with gastrointestinal symptoms in patients with T2D.¹⁷ The other pairs of bridging symptoms (DDS16-PHQ2 and DDS17-PHQ6) were associated with inadequate support from family or friends, whether emotional support or dietary help. The above analysis of bridging symptoms summarizes the connection between DD and DS into three aspects: Fatigue, diet, and social interaction. For patients with DD and DS comorbidity, these three aspects may serve as effective intervention targets to sever the connection and comorbidity of DD and DS, representing a significant finding of this study. Based on evidence that dietary management,⁶⁷ peer support,⁶⁸ and family-focused interventions⁶⁹ have independently demonstrated significant benefits for psychological well-being in patients with T2D, we recommend developing a comprehensive intervention package that integrates these approaches to address DD and DS simultaneously.

The second network structure demonstrated the relationship between DD and SS. A strong negative bridge appeared in SSRS2/SSRS3-DDS17. In contrast to the objective support represented by SSR1, the subjective support represented by SSRS2 indicated a negative association with DDS17. This indicates that emotional support from family or friends is more important for patients with T2D than material and financial support and may directly affect patients’ self-cognition. This is consistent with the results of several systematic reviews, where low SS was reported to increase the risk of depression among people with T2D,⁷⁰ and increased SS was inversely associated with emotional distress.⁷¹ More importantly, SS is more linked to the self-management of people with T2D than T1D.⁷² Similarly, the support utilization represented by SSRS3 is equally significant for patients with T2D. This implies that even when subjective and objective support are sufficient, the failure of the patient to perceive or utilize this support may, however, impact the success of their disease management. Few studies have noted this, with only one qualitative study⁷³ examining how adolescents with T2D understand and use SS, indicating that their use of SS is restricted to close friends and family due to fear of disclosing their diabetes to others. Several randomized controlled trials have demonstrated that different SS technologies, including mobile health-enhanced peer support intervention⁷⁴ and peer-led diabetes self-management support intervention,⁷⁵ effectively reduce DD among patients with T2D. Our findings revealed that effective SS must incorporate emotional support components and actively encourage patient engagement with available resources, as interventions limited to offering disease-specific knowledge and skill training are insufficient for comprehensive SS.

Certain limitations must be addressed. First, using cross-sectional data made identifying direct effects between symptoms impossible. Consequently, it is unclear whether the most central symptoms activate other symptoms, are activated by other symptoms, or are the case for both. To examine this causal relationship, longitudinal study data are necessary to provide new insights into the dynamic relationship between DD-DS/SS. Second, our survey was conducted during the COVID-19 pandemic. Therefore, it is impossible to rule out the possibility that the prevalence of the virus influenced the psychological state of people with T2D. Finally, although the sample size of this study is sufficient for network analysis, it is inadequate to support network comparison tests between different subgroups.⁴⁹ Future studies should expand the sample size to more comprehensively investigate the differences in the co-occurrence networks of DD, DS, and SS among different samples.

Conclusions

Our study investigated the interconnections between components of DD and the correlations between constructs of DS, SS, and DD in patients with T2D using network analysis. Our findings from the DD network indicated that physician-related distress may significantly contribute to the development and maintenance of DD. From the DD-DS-SS network, the first significant finding is that the complex link between DD and DS can be summarized in three aspects: Fatigue, diet, and social interaction. Another significant finding is that the subjective support and utilization of support in patients with T2D are closely related to managing their disease. The findings provided more targeted theoretical guidance and a scientific basis for psychological counseling and interventions aimed at alleviating DD in patients with T2D. However, all the above conclusions require more confirmatory studies in the future for validation.

Data Sharing Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Ethics Approval and Consent to Participate

This study was approved by the Ethics Committee of the Chengdu Jinniu District People’s Hospital (QYYLL-2022-011), and written informed consent was obtained from every participant.

Acknowledgments

We would like to thank the study participants, clinicians, and nurses for their unreserved help. We also gratefully acknowledge the financial supports from the Sichuan Province Grassroots Health Development Research Center (SWFZ23-Y-23) and the 2021 Xinglin Scholars Scientific Research Promotion Project of Chengdu University of Traditional Chinese Medicine (MPRC2021021). Meanwhile, we gratefully acknowledge Dr. Jingting Liao from Chengdu Jinniu District People’s Hospital for securing the ethical approvals critical to this study.

Funding

This study was supported by grants from the Sichuan Province Grassroots Health Development Research Center (SWFZ23-Y-23) and the 2021 Xinglin Scholars Scientific Research Promotion Project of Chengdu University of Traditional Chinese Medicine (MPRC2021021).

Disclosure

The authors report no conflicts of interest regarding this manuscript.

References

1. Poole L, Hackett RA. Diabetes distress: the psychological burden of living with diabetes. Lancet Diabetes Endocrinol. 2024;12(7):439–441. doi:10.1016/S2213-8587(24)00126-8

2. Skinner TC, Joensen L, Parkin T. Twenty-five years of diabetes distress research. Diabet Med. 2020;37(3):393–400. doi:10.1111/dme.14157

3. Evans M, Ellis DA, Vesco AT, et al. Diabetes distress in urban Black youth with type 1 diabetes and their caregivers: associations with glycemic control, depression, and health behaviors. J Pediatr Psychol. 2024;49(6):394–404. doi:10.1093/jpepsy/jsad096

4. Ehrmann D, Chatwin H, Schmitt A, et al. Reduced heart rate variability in people with type 1 diabetes and elevated diabetes distress: results from the longitudinal observational DIA-LINK1 study. Diabet Med. 2023;40(4):e15040. doi:10.1111/dme.15040

5. Huang W, Aune D, Ferrari G, et al. Psychological distress and all-cause, cardiovascular disease, cancer mortality among adults with and without diabetes. Clin Epidemiol. 2021;13:555–565. doi:10.2147/CLEP.S308220

6. Lindekilde N, Pouwer F. More work is needed to better understand diabetes distress as a predictor of all-cause mortality in type 2 diabetes. Diabetologia. 2018;61(10):2247–2248. doi:10.1007/s00125-018-4714-z

7. Saseetharran A, Patel SA. COVID-19 pandemic-related healthcare interruptions and diabetes distress: a national study of US adults with diabetes. Bmc Public Health. 2024;24(1):493. doi:10.1186/s12889-024-17921-3

8. Zhang M, Zhang X, Yang Y, et al. Diabetes distress among patients undergoing surgery for diabetic retinopathy and associated factors: a cross- sectional survey. Psychol Res Behav Manag. 2024;17:2039–2040. doi:10.2147/PRBM.S474995

9. Trojanowski PJ, Pardon A, Reynolds C, et al. Body mass index moderates the association between diabetes distress and objective self-management behaviours in adolescents with type 1 diabetes and elevated A1Cs. Diabet Med. 2024;41(7):e15325. doi:10.1111/dme.15325

10. Lewinski AA, Shapiro A, Crowley MJ, et al. Diabetes distress in Veterans with type 2 diabetes mellitus: qualitative descriptive study. J Health Psychol. 2024;29(14):1399771157.

11. Snoek FJ, Bremmer MA, Hermanns N. Constructs of depression and distress in diabetes: time for an appraisal. Lancet Diabetes Endocrinol. 2015;3(6):450–460. doi:10.1016/S2213-8587(15)00135-7

12. Bajl TD. Depression or diabetes distress? Perspect Psychiatr C. 2017;1–4.

13. Nouwen A. Depression and diabetes distress. Diabet Med. 2015;32(10):1261–1263. doi:10.1111/dme.12863

14. Boehmer K, Lakkad M, Johnson C, Painter JT. Depression and diabetes distress in patients with diabetes. Prim Care Diabetes. 2023;17(1):105–108. doi:10.1016/j.pcd.2022.11.003

15. AlOzairi A, Irshad M, AlKandari J, et al. Prevalence and predictors of diabetes distress and depression in people with type 1 diabetes. Front Psychiatry. 2024;15:1367876. doi:10.3389/fpsyt.2024.1367876

16. Fisher L, Skaff MM, Mullan JT, et al. A longitudinal study of affective and anxiety disorders, depressive affect and diabetes distress in adults with Type 2 diabetes. Diabet Med. 2008;25(9):1096–1101. doi:10.1111/j.1464-5491.2008.02533.x

17. Kamruzzaman M, Horowitz M, Polonsky WH, et al. Diabetes distress and depression are independently associated with gastrointestinal symptoms in type 2 diabetes in Bangladesh. Diabet Med. 2024;41(11):e15379. doi:10.1111/dme.15379

18. Owens-Gary MD, Zhang X, Jawanda S, et al. The importance of addressing depression and diabetes distress in adults with type 2 diabetes. J Gen Intern Med. 2019;34(2):320–324. doi:10.1007/s11606-018-4705-2

19. Young-Hyman D, de Groot M, Hill-Briggs F, et al. Psychosocial care for people with diabetes: a position statement of the American diabetes association. Diabetes Care. 2016;39(12):2126–2140. doi:10.2337/dc16-2053

20. McMorrow R, Hunter B, Hendrieckx C, et al. Effect of routinely assessing and addressing depression and diabetes distress on clinical outcomes among adults with type 2 diabetes: a systematic review. BMJ Open. 2022;12(5):e54650. doi:10.1136/bmjopen-2021-054650

21. Albright D, Wardell J, Harrison A, et al. Screening for diabetes distress and depression in routine clinical care for youth with type 1 diabetes. J Pediatr Psychol. 2024;49(5):356–364. doi:10.1093/jpepsy/jsae016

22. Fisher L, Gonzalez JS, Polonsky WH. The confusing tale of depression and distress in patients with diabetes: a call for greater clarity and precision. Diabet Med. 2014;31(7):764–772. doi:10.1111/dme.12428

23. Ehrmann D, Kulzer B, Haak T, et al. Longitudinal relationship of diabetes-related distress and depressive symptoms: analysing incidence and persistence. Diabet Med. 2015;32(10):1264–1271. doi:10.1111/dme.12861

24. Burns RJ, Deschênes SS, Schmitz N. Cyclical relationship between depressive symptoms and diabetes distress in people with type 2 diabetes mellitus: results from the Montreal Evaluation of Diabetes Treatment Cohort Study. Diabet Med. 2015;32(10):1272–1278. doi:10.1111/dme.12860

25. Shuiyuan X. Theoretical basis and research application of the social support rating scale. J Clin Psychiamed. 1994;(2):98–100.

26. Beverly EA, Ritholz MD, Dhanyamraju K. The buffering effect of social support on diabetes distress and depressive symptoms in adults with type 1 and type 2 diabetes. Diabet Med. 2021;38(4):e14472. doi:10.1111/dme.14472

27. Onu DU, Ifeagwazi CM, Prince OA. Social support buffers the impacts of diabetes distress on health-related quality of life among type 2 diabetic patients. J Health Psychol. 2022;27(10):2305–2317. doi:10.1177/1359105320980821

28. Presley CA, Mondesir FL, Juarez LD, et al. Social support and diabetes distress among adults with type 2 diabetes covered by Alabama Medicaid. Diabet Med. 2021;38(4):e14503. doi:10.1111/dme.14503

29. Peimani M, Garmaroudi G, Stewart AL, et al. Type 2 diabetes burden and diabetes distress: the buffering effect of patient-centred communication. Can J Diabetes. 2022;46(4):353–360. doi:10.1016/j.jcjd.2021.11.007

30. Hoogendoorn CJ, Schechter CB, Llabre, et al. Distress and type 2 diabetes self-care: putting the pieces together. Ann Behav Med. 2021;55(10):938–948. doi:10.1093/abm/kaaa070

31. Kenny E, O’Malley R, Roche K, et al. Diabetes distress instruments in adults with type 1 diabetes: a systematic review using the COSMIN (Consensus-based standards for the selection of health status measurement instruments) checklist. Diabet Med. 2021;38(4):e14468. doi:10.1111/dme.14468

32. Tang F-Y, Guo X-T, Zhang L, et al. The prevalence of diabetes distress in Chinese patients with type 2 diabetes: a systematic review and meta-analysis. Diabet Res Clin Pract. 2023;206:110996. doi:10.1016/j.diabres.2023.110996

33. Straton E, Anifowoshe K, Moore H, et al. Associations of coping strategies with glycemic and psychosocial outcomes among adolescents with type 1 diabetes experiencing diabetes distress. Ann Behav Med. 2024;58(9):628–633. doi:10.1093/abm/kaae028

34. Davis WA, Bruce DG, Davis T, et al. The clinical relevance of diabetes distress versus major depression in type 2 diabetes: a latent class analysis from the Fremantle Diabetes Study Phase II. J Clin Med. 2023;12(24):7722. doi:10.3390/jcm12247722

35. Chen SY, Hsu HC, Huang CL, et al. Impact of type d personality, role strain, and diabetes distress on depression in women with type 2 diabetes: a cross-sectional study. J Nurs Res. 2023;31(1):e258. doi:10.1097/jnr.0000000000000536

36. Carper MM, Traeger L, Gonzalez JS, et al. The differential associations of depression and diabetes distress with quality of life domains in type 2 diabetes. J Behav Med. 2014;37(3):501–510. doi:10.1007/s10865-013-9505-x

37. Guan M, Liu J, Li X, et al. The impact of depressive and anxious symptoms on non-suicidal self-injury behavior in adolescents: a network analysis. Bmc Psychiatry. 2024;24(1):229. doi:10.1186/s12888-024-05599-1

38. Denny Borsboom AOJC. Network analysis: an integrative approach to the structure of psychopathology. Ann Rev Clin Psychol. 2013;1(9):91–121.

39. Borsboom D. A network theory of mental disorders. World Psychiatry. 2017;16(1):5–13. doi:10.1002/wps.20375

40. Li D, Dai FM, Xu JJ, et al. Characterizing hotspots and frontier landscapes of diabetes-specific distress from 2000 to 2018: a bibliometric study. Biomed Res Int. 2020;2020(1):8691451. doi:10.1155/2020/8691451

41. Li Y, Teng D, Shi X, et al. Prevalence of diabetes recorded in mainland China using 2018 diagnostic criteria from the American diabetes association: national cross sectional study. BMJ. 2020;369:m997. doi:10.1136/bmj.m997

42. Polonsky WH, Anderson BJ, Lohrer PA, et al. Assessment of diabetes-related distress. Diabetes Care. 1995;18(6):754–760. doi:10.2337/diacare.18.6.754

43. Zhang YY, Li W, Sheng Y. The Chinese version of the revised diabetes distress scale for adults with type 2 diabetes: translation and validation study. Int J Nurs Sci. 2022;9(2):243–251. doi:10.1016/j.ijnss.2022.03.002

44. Tay D, Chua M, Khoo J. Validity of the short-form five-item problem area in diabetes questionnaire as a depression screening tool in type 2 diabetes mellitus patients. J Diabetes Investig. 2023;14(9):1128–1135. doi:10.1111/jdi.14051

45. Chen W, Maimaitituerxun R, Xiang J, et al. Social support, coping strategies, depression, anxiety, and cognitive function among people with type 2 diabetes mellitus: a path analysis. Psychiatry Invest. 2024;21(9):1033–1044. doi:10.30773/pi.2024.0024

46. Friedman J, Hastie T, Tibshirani R. Sparse inverse covariance estimation with the graphical lasso. Biostatistics. 2008;9(3):432–441. doi:10.1093/biostatistics/kxm045

47. Hevey D. Network analysis: a brief overview and tutorial. Health Psychol Behav Med. 2018;6(1):301–328. doi:10.1080/21642850.2018.1521283

48. Opsahl T, Agneessens F, Skvoretz J. Node centrality in weighted networks: generalizing degree and shortest paths. Soc Networks. 2010;32(3):245–251.

49. Jones PJ, Ma R, McNally RJ. Bridge centrality: a network approach to understanding comorbidity. Multivar Behav Res. 2021;56(2):353–367. doi:10.1080/00273171.2019.1614898

50. Epskamp S, Borsboom D, Fried EI. Estimating psychological networks and their accuracy: a tutorial paper. Behav Res Methods. 2018;50(1):195–212. doi:10.3758/s13428-017-0862-1

51. Polonsky WH, Fisher L, Earles J, et al. Assessing psychosocial distress in diabetes: development of the diabetes distress scale. Diabetes Care. 2005;28(3):626–631.

52. Rodebaugh TL, Tonge NA, Piccirillo ML, et al. Does centrality in a cross-sectional network suggest intervention targets for social anxiety disorder? J Consult Clin Psychol. 2018;86(10):831–844. doi:10.1037/ccp0000336

53. McInerney AM, Lindekilde N, Nouwen A, et al. Diabetes distress, depressive symptoms, and anxiety symptoms in people with type 2 diabetes: a network analysis approach to understanding comorbidity. Diabetes Care. 2022;45(8):1715–1723. doi:10.2337/dc21-2297

54. Laranjo L, Neves AL, Costa A, et al. Facilitators, barriers and expectations in the self-management of type 2 diabetes–a qualitative study from Portugal. Eur J Gen Pract. 2015;21(2):103–110. doi:10.3109/13814788.2014.1000855

55. Mameli C, Smylie GM, Galati A, et al. Safety, metabolic and psychological outcomes of Medtronic MiniMed 670G in children, adolescents and young adults: a systematic review. Eur J Pediatr. 2023;182(5):1949–1963.

56. Soni A, Ng SM. Intensive diabetes management and goal setting are key aspects of improving metabolic control in children and young people with type 1 diabetes mellitus. World J Diabetes. 2014;5(6):877–881. doi:10.4239/wjd.v5.i6.877

57. van Dijk CE, Verheij RA, Swinkels IC, et al. What part of the total care consumed by type 2 diabetes patients is directly related to diabetes? Implications for disease management programs. Int J Integr Care. 2011;11:e140.

58. Winocour PH. Diabetes and chronic kidney disease: an increasingly common multi-morbid disease in need of a paradigm shift in care. Diabet Med. 2018;35(3):300–305. doi:10.1111/dme.13564

59. Ismail K, Winkley K, de Zoysa N, et al. Nurse-led psychological intervention for type 2 diabetes: a cluster randomised controlled trial (diabetes-6 study) in primary care. Br J Gen Pract. 2018;68(673):e531–e540. doi:10.3399/bjgp18X696185

60. Young HM, Miyamoto S, Dharmar M, et al. Nurse coaching and mobile health compared with usual care to improve diabetes self-efficacy for persons with type 2 diabetes: randomized controlled trial. JMIR mHealth uHealth. 2020;8(3):e16665. doi:10.2196/16665

61. Guo J, Wang H, Ge L, et al. Effectiveness of a nurse-led mindfulness stress-reduction intervention on diabetes distress, diabetes self-management, and HbA1c levels among people with type 2 diabetes: a pilot randomized controlled trial. Res Nurs Health. 2022;45(1):46–58.

62. Taylor KI, Oberle KM, Crutcher RA, et al. Promoting health in type 2 diabetes: nurse-physician collaboration in primary care. Biol Res Nurs. 2005;6(3):207–215.

63. Abdoli S, Hessler D, Smither B, et al. New insights into diabetes burnout and its distinction from diabetes distress and depressive symptoms: a qualitative study. Diabet Res Clin Pract. 2020;169:108446. doi:10.1016/j.diabres.2020.108446

64. Nano J, Carinci F, Okunade O, et al. A standard set of person-centred outcomes for diabetes mellitus: results of an international and unified approach. Diabet Med. 2020;37(12):2009–2018. doi:10.1111/dme.14286

65. Lin CL, Chang YT, Liu WC, et al. Exploring and developing a new culturally-appropriate diabetes distress scale in Taiwan. Front Public Health. 2022;10:888661.

66. Forouhi NG. Embracing complexity: making sense of diet, nutrition, obesity and type 2 diabetes. Diabetologia. 2023;66(5):786–799.

67. Kakoschke N, Zajac IT, Tay J, et al. Effects of very low-carbohydrate vs. high-carbohydrate weight loss diets on psychological health in adults with obesity and type 2 diabetes: a 2-year randomized controlled trial. Eur J Nutr. 2021;60(8):4251–4262. doi:10.1007/s00394-021-02587-z

68. Ju C, Shi R, Yao L, et al. Effect of peer support on diabetes distress: a cluster randomized controlled trial. Diabet Med. 2018;35(6):770–775. doi:10.1111/dme.13625

69. Roddy MK, Spieker AJ, Nelson LA, et al. Well-being outcomes of a family-focused intervention for persons with type 2 diabetes and support persons: main, mediated, and subgroup effects from the FAMS 2.0 RCT. Diabet Res Clin Pract. 2023;204:110921.

70. Azmiardi A, Murti B, Febrinasari RP, et al. Low social support and risk for depression in people with type 2 diabetes mellitus: a systematic review and meta-analysis. J Prev Med Public Health. 2022;55(1):37–48. doi:10.3961/jpmph.21.490

71. Almubaid Z, Alhaj Z, Almosa O, et al. The impact of social support on health outcomes of diabetic patients: a systematic review. Cureus. 2024;16(8):e67842. doi:10.7759/cureus.67842

72. Song Y, Nam S, Park S, et al. The impact of social support on self-care of patients with diabetes: what is the effect of diabetes type? Systematic review and meta-analysis. Diabetes Educ. 2017;43(4):396–412. doi:10.1177/0145721717712457

73. Brouwer AM, Salamon KS, Olson KA, et al. Adolescents and type 2 diabetes mellitus: a qualitative analysis of the experience of social support. Clin Pediatr. 2012;51(12):1130–1139. doi:10.1177/0009922812460914

74. Presley C, Agne A, Shelton T, et al. Mobile-enhanced peer support for African Americans with type 2 diabetes: a randomized controlled trial. J Gen Intern Med. 2020;35(10):2889–2896.

75. Tang TS, Afshar R, Elliott T, et al. From clinic to community: a randomized controlled trial of a peer support model for adults with type 2 diabetes from specialty care settings in British Columbia. Diabet Med. 2022;39(11):e14931. doi:10.1111/dme.14931