Category: 3. Business

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

Depression is a prevalent clinical mental illness characterized by a persistent low mood accompanied by varying degrees of cognitive and behavioral changes, often resulting in functional impairment. It can affect a patient’s education, work, and social life, and in severe cases, may lead to suicide.¹ A Chinese mental health survey reported a lifetime prevalence of depression disorder in China of 6.8%, with 53.1% of patients experiencing suicidal ideation, and 15% ultimately dying by suicide.^2–4 Suicide attempts pose a significant threat to patients’ safety and impose a profound psychological and economic burden on families.

Assessment scales are among the most effective tools for evaluating the severity of depression in patients. The Hamilton Depression Scale (HAMD-24) is one of the most widely used tools for depression screening. It helps assess the severity of depression and evaluate treatment effects. It also includes items indicative of suicidal ideation, such as feelings of guilt, despair, and hopelessness. Recent studies suggest that these factors are strongly correlated with suicide risk in patients with depression. Despair, in particular, is identified as a significant predictor, reflecting a core feature of depression that contributes to the development of suicidal thoughts. The HAMD-24’s role in early suicide risk assessment is critical, as it can help identify individuals who may benefit from timely intervention.^5,6

Building on recent ambivalence-focused research,^7,8 there is growing consensus that suicidal desire fluctuates dynamically. Ecological momentary assessment and digital phenotyping now capture real-time swings between the wish to live and the wish to die. Integrating such ambivalence markers with conventional scales like the HAMD-24 could therefore enhance early detection beyond static symptom snapshots.

However, emerging research highlights the ambivalent nature of suicidal ideation—where patients simultaneously experience desire for death and connection to life.⁷ This complexity underscores limitations of static assessments like the HAMD-24. Novel approaches capturing dynamic triggers (eg, real-time affective shifts) may enhance early detection.⁸ Recent computational psychiatry studies advocate integrating passive digital phenotyping (eg, speech patterns, actigraphy) with clinical scales to detect subtle risk signatures preceding crises.^9–12

In fact, most suicides are preventable, particularly by analyzing the factors strongly associated with suicidal ideation on depression scales like the HAMD-24.¹³ This analysis is valuable for the early prevention and intervention of suicidal behaviors. Such insights can serve as a theoretical reference for developing new scales focused on depression and suicide risk assessment.^14–16

Machine learning is a subset of artificial intelligence that selects the optimal algorithm from complex datasets, effectively addressing the limitations of traditional statistical methods.¹⁷ It can automatically prioritize variables based on relevance, incorporate them into the most efficient model, and identify and manage nonlinear relationships and interactions between variables.¹⁸ Walsh et al found that machine learning models predicted suicide risk more accurately than traditional logistic regression analysis methods using databases, with random forest (RF) and support vector machine (SVM) algorithms demonstrating the highest accuracy.^19–22

In recent years, machine learning has been widely applied to population-based suicide predictions, yet critical gaps remain in applying these techniques specifically to clinical depression scales for suicidal ideation prediction. The existing research combining machine learning with depression scales has the following issues:^23–25 1. Lack of comparative analysis of diverse algorithms, potentially overlooking optimal model performance; 2. There is limited exploration of the relative importance and interactions of characteristic symptoms in mature clinical assessment scales such as HAMD-24, which hinders clinical interpretability and the identification of core mechanistic pathways; 3. The main focus is on population level risks, lacking granular symptom profiles for validated clinical evaluations and failing to leverage the rich, standardized symptom data inherent in widely used scales like the HAMD-24 for individualized risk prediction within clinical workflows. Subsequently, it employs and rigorously compares four distinct machine learning algorithms selected for their complementary strengths in clinical prediction tasks: an SVM model (effective in high-dimensional spaces and robust to outliers),²⁶ a naive Bayes classification (NBC) model (computationally efficient and well suited for probabilistic inference),²⁷ an RF model (excellent handling of complex interactions and feature importance estimation),²⁸ and an extremely random tree classification (ERTC) model (enhances diversity and robustness through extreme randomization, potentially reducing overfitting).²⁹ The classification and prediction performances of these four models were compared, and the most predictive model was selected to analyze factors associated with suicidal ideation. This dual approach aims to both characterize associated factors and enable early interventions and reduce the occurrence of suicidal behavior while providing transparent insights into key predictive features within the HAMD-24. Prior machine learning studies have relied on heterogeneous data sources (EHRs, social media) rather than standardized clinical scales, limiting bedside utility. Moreover, no study has systematically compared SVM, NBC, RF, and ERTC on HAMD-24 items while excluding the direct suicide item to avoid circularity.

Based on the previous research results, we have put forward the following hypotheses:

Hypothesis 1: An ensemble tree model (ERTC) will outperform SVM, NBC, and RF in predicting suicidal ideation (AUC difference ≥ 0.05).

Hypothesis 2: Among HAMD-24 items (excluding Item 3), despair, guilt, inferiority, work and interests, depressive emotions may be server as the five strongest predictors (top quartile of feature importance).

Population and Sample

Inclusion and Exclusion Criteria

A total of 374 patients diagnosed with depression were involved in this study, who received treatment at the psychological clinic of the Changzhou Second People’s Hospital between March 2022 and October 2023, via consecutive sampling of all eligible out-patients during the study window. A priori power analysis was conducted using G*Power 3.1. Based on an expected medium effect size (Cohen’s f² = 0.15), α = 0.05, power (1-β) = 0.80, and 23 predictor variables (demographics + HAMD-24 items excluding Item 3), the minimum required sample size was 190. Our final sample (N = 374) exceeds this threshold by >95%, ensuring robust statistical power for both traditional analyses and machine learning modeling. This sample size also adheres to the heuristic of ≥10 events per predictor variable (EPV) for regression-based models (23 predictors × 10 = 230; our suicidal ideation group had 233 cases) and meets computational requirements for complex ensemble algorithms like ERTC.^28,29

The inclusion criteria were as follows: ① Fulfillment of the diagnostic criteria for depression as outlined in the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), confirmed by two psychiatrists at the attending level or higher;³⁰ ② No gender restrictions, with ages ranging from 18 to 60 years; ③ HAMD-24 scores of ≥8 points; ④ Ability to understand the questionnaire and cooperate with its completion; ⑤ Obtaining informed consent from the patient or their legal guardian. The exclusion criteria were as follows: ① Coexisting schizophrenia, mania, bipolar disorder, or other mental illnesses; ② Other neurological or severe physical conditions; ③ Secondary depression caused by other medical conditions.

To minimize confounding effects:

Antidepressant use (type/dose) was recorded and included as a covariate in statistical analyses;

Demographic variables (age, gender, education) showing group differences (Factors Associated with Suicidal Ideation: DemoFigureic Characteristics) were controlled via multivariate regression;

Stratified sampling preserved class balance (suicidal ideation yes/no) during data splitting for machine learning.

Demographic covariates (age, gender, education) and antidepressant status were entered as covariates in the ERTC model to control residual confounding.

This study received approval from the Medical Ethics Committee of Changzhou Second People’s Hospital (Approval No.: 2020KY04-01). All patients were informed about the research objectives and provided written informed consent. Data anonymization was implemented by assigning unique identifiers, and access was restricted to the research team.

Data Collection

Instrumentation

This study employed a self-designed questionnaire to gather general information, including age, gender, years of education, smoking and drinking history, and antidepressant use. Antidepressant use: Currently prescribed medication for ≥4 weeks at time of assessment. Smoking/drinking history: Regular use (≥3 times/week) for >6 months.

Scale Evaluation

The HAMD-24 is a widely used scale for assessing depression clinically, consisting of 24 items and 7 factors (anxiety/somatization, weight, cognitive impairment, day/night changes, delays, sleep disorders, and feelings of despair). The total score indicates the severity of depression, with higher scores indicating more severe depression. This scale demonstrates strong reliability and validity, with coefficients of 0.99 and 0.37, respectively.⁶ In this study, the scale was used to assess depression severity, with a score of less than 8 points indicating no depressive symptoms, less than 20 points indicating mild symptoms, 20 to 35 points indicating moderate symptoms, and above 35 points indicating severe symptoms. Item 3 of the HAMD-24 assesses suicide risk, and to avoid interference with this item, previous studies have typically excluded it when investigating the correlation between suicide risk and depressive symptom severity.³¹ Therefore, this study excluded item 3 in assessing the correlation between HAMD-24 factors and suicide risk. The BSI-CV consists of 19 items, each scored on a scale of 0 to 2. Items 1 to 5 assess suicidal ideation, while items 6 to 19 evaluate suicidal tendencies. If the answers to items 4 and 5 are “none”, the patient is considered not to have suicidal ideation and does not need to complete the remaining items. If the answers to items 4 or 5 are “weak” or “moderate to strong”, the patient is considered to have suicidal ideation and must complete the remaining items. The total score on the BSI-CV ranges from 0 to 38 points, with higher scores indicating stronger suicidal ideation and a higher suicide risk. This study used only the first 5 items of the BSI-CV to evaluate suicidal ideation. The scale’s Cronbach α coefficient was 0.78.³²

Model Framework

The machine learning prediction model involved several steps:

Data preprocessing: Exclusion of HAMD-24 Item 3 (suicide item) to avoid target leakage;

Train-test split: Random stratified partitioning (80% training, 20% testing) to maintain class distribution;

Model development: Implementation of SVM, NBC, RF, and ERTC using scikit-learn (v1.2.2);

Hyperparameter tuning: 5-fold cross-validation on training set with randomized search (100 iterations);

Model validation: Evaluation on held-out test set;

Final evaluation: Performance metrics calculation.

The flowchart of the machine learning prediction model is shown in Figure 1.

Figure 1 Flow chart of machine learning prediction model.

Key parameters:

Cross-validation: 5 folds with stratified sampling; Tree-based models (RF/ERTC): nestimators=200, maxdepth=10 (optimized via cross-validation); SVM: RBF kernel, C=1.0, gamma=“scale”; NBC: Gaussian priors.

After comprehensive evaluation, the ERTC model demonstrated superior prediction performance.

Sampling Rationale for Machine Learning: The dataset was randomly split into training (80%, n=299) and testing (20%, n=75) sets using stratified sampling to preserve the distribution of the target variable (suicidal ideation). This partitioning ratio optimizes model training while retaining sufficient independent samples for unbiased validation, aligning with best practices in clinical machine learning.^17,19

Reproducibility: Random seed 42 was fixed throughout all randomized procedures (data splitting, model initialization). Complete code and preprocessing scripts are available from corresponding author upon request.

Statistical Methods

Statistical analysis was conducted using the SPSS 25.0 software. Continuous data were assessed for normality using Shapiro–Wilk tests. Non-normally distributed clinical data are expressed as medians (M) with interquartile ranges (P25, P75). Non-parametric rank sum tests (Mann–Whitney U-tests) were applied to two independent samples, and rank sum tests were used for comparisons between groups. Comparisons of count data were conducted using chi-squared (χ²) tests or Fisher’s exact test where cell counts were <5, with results presented as cases (%). A Bonferroni correction was applied for multiple comparisons of HAMD-24 items (adjusted P < 0.0022 (0.05/23 items)). Statistical significance was determined at a threshold of P <0.05. Non-parametric tests were selected because (i) age and HAMD-24 items were non-normally distributed (Shapiro–Wilk p < 0.001) and (ii) ordinal or binary variables precluded parametric alternatives. These tests served only to guide feature inclusion, not to test causal hypotheses.

Machine learning models were implemented in Python 3.9 using scikit-learn (v1.2.2). All models underwent identical preprocessing and evaluation protocols to ensure fair comparison. The SVM method, designed based on optimization strategies, minimizes influence from outliers and demonstrates strong generalization ability to achieve global optimal solutions.²⁶ NBC is a well-established probabilistic algorithm used for text classification, providing efficient, stable classification, rapid computation, and high accuracy.²⁷ RF is an ensemble learning algorithm based on decision trees, capable of balancing classification errors and maintaining prediction performance, with faster learning and training efficiency.²⁸ ERTC is an ensemble learning technique that constructs multiple decision trees using the Gini index for data splitting and aggregates different results to output classification outcomes.²⁹

Model performance was assessed using metrics including accuracy, precision, recall, F1-scores, Kappa coefficients, Matthews correlation coefficient (MCC), and area under the curve (AUC) values. The optimal model was selected based on these performance indicators.

Results

Factors Associated with Suicidal Ideation: Demographic Characteristics

To provide clinicians with baseline information on potential confounders, we first conducted exploratory group comparisons; these analyses were not intended to predict suicidal ideation but to inform feature selection for the machine-learning pipeline.

This study included a total of 374 patients, who were divided into two groups: 141 without suicidal ideation and 233 with suicidal ideation. As demonstrated in Table 1, no statistically significant differences were observed between the two groups in terms of the first/follow-up visit (χ² = 1.15, P = 0.28), gender (χ² = 1.98, P = 0.16), educational level (χ² = 6.27, P = 0.099), smoking history (χ² = 1.18, P = 0.277), and drinking history (χ² = 2.66, P = 0.103). However, the median age of those with suicidal ideation was significantly lower than that of those without (Z = −6.62, P < 0.001), and a statistically significant difference was found in the use of antidepressants (χ² = 4.21, P < 0.05). These variables were subsequently included as covariates in machine learning feature engineering to mitigate confounding.

Table 1 General Information of Depressed Patients with and without Suicidal Ideation

Factors Associated with Suicidal Ideation: HAMD-24 Symptom Profiles

Statistically significant differences (Table 2) were identified for the HAMD-24 total score and 17 sub-domains: depression emotions (Z = −6.28., P < 0.001), guilt (Z = −5.67, P < 0.001), difficulty falling asleep (Z = −3.83, P < 0.001), work and interests (Z = −7.07, P < 0.001), psychomotor retardation (Z = −2.17, P < 0.05), agitation (Z = −2.06, P < 0.05), somatic anxiety (Z = −3.48, P < 0.001), gastrointestinal symptoms (Z = −4.05, P < 0.001), systemic symptoms (Z = −3.82, P < 0.001), hypochondriasis (Z = −2.01, P < 0.05), weight loss (Z = −3.73, P < 0.001), self-awareness (Z = −3.15, P < 0.05), diurnal variation (Z = −3.53, P < 0.001), paranoid symptoms (Z = −5.79, P < 0.001), reduced abilities (Z = −3.49, P < 0.001), despair (Z = −8.01, P < 0.001), and inferiority complex (Z = −6.49, P < 0.001). The remaining 6 sub-domains—lack of deep sleep (Z = −0.38, P = 0.707), early awakening (Z = −1.78, P = 0.075), psychological anxiety (Z = −1.50, P = 0.134), sexual symptoms (Z = −1.47, P = 0.141), personality disintegration or reality disintegration (Z = −1.43, P = 0.152), and compulsive symptoms (Z = −1.53, P = 0.127)—showed no significant differences. The negative z-value indicates that the depression score among individuals without suicidal ideation is lower than among those with suicidal ideation. The total HAMD-24 score (excluding item 3) was significantly higher in the suicidal ideation group (Z = −8.81, P < 0.001).

Table 2 Comparison of HAMD-24 Total Scores Between Depressed Patients with and without Suicidal Ideation [M (P25, P75)]

Analysis of Machine Learning Results

Accuracy refers to the proportion of correctly classified samples in relation to the total number of samples in a given class. The AUC is an evaluation metric that assesses the quality of a binary classification model, indicating the probability that the predicted value of a positive example exceeds that of a negative example. The recall rate, (true examples (TP)/(TP + false negative examples (FN))), represents the ratio of correctly predicted positive samples to the total number of actual positive samples.

Precision, (TP/(TP + false positive examples (FP)), represents the ratio of correctly predicted positive samples to the total number of predicted positive samples. The F1 score is defined as F1 = 2/(1/recall + 1/precision).

The Kappa coefficient is a consistency measure used to assess the effectiveness of classifications, specifically to determine whether the model’s predicted results align with the actual classification outcomes. A value greater than 0.5 indicates strong classification performance.

The MCC coefficient is used to evaluate the performance of binary classification models, especially when dealing with imbalanced datasets. It incorporates TP, TN, FP, and FN, offering a single value that summarizes classification quality, with values greater than 0.5 indicating strong classification performance.

As demonstrated in Table 3, the performance evaluation of the four machine learning models revealed that the ERTC model achieved excellent classification performance, with a prediction accuracy of 77.75% on the independent test set. Additional indicators demonstrated that this model exhibited strong classification performance in analyzing suicidal ideation.

Table 3 Prediction Results of Four Machine Learning Algorithm Models

The area under the receiver operating characteristic (ROC) curve, known as the AUC, serves as an indicator for evaluating classifier performance. A larger AUC reflects a better classifier performance. An AUC greater than 0.9 is considered very good, 0.8 to 0.9 is good, 0.7 to 0.79 is median, and less than 0.7 is poor. The closer the ROC curve is to the upper left corner, the better the classification performance of the algorithm. Based on the ROC curves and AUC values shown in Figure 2, the ERTC model demonstrated high classification performance in predicting suicidal ideation.

Figure 2 ROC Curves for ExtraTreesClassifier.

Based on the confusion matrix presented in Figure 3, the ERTC model demonstrated good classification accuracy, achieving a correct classification rate of 77%.

Figure 3 ExtraTreesClassifier Confusion Matrix.

Notably, this represents the first machine learning study to identify symptom-level predictors of suicidal ideation exclusively from HAMD-24 items (excluding Item 3) with clinically actionable accuracy (AUC=0.80). The ERTC model’s superior performance—outperforming established algorithms like SVM and RF—validates its utility for precision psychiatry applications.

The feature analysis of the ERTC model, demonstrated in Figure 4, identified the most important features as feelings of despair, guilt, inferiority, work and interest, and depression.

Figure 4 Feature Importance Plot.

Discussion

Unlike prior studies that combined heterogeneous clinical notes or social-media data,^19,25 our study uniquely isolates symptom-level predictors within a single, widely adopted scale (HAMD-24) and demonstrates that ensemble tree methods outperform SVM and RF under these constraints.

This study pioneers the integration of machine learning with granular HAMD-24 symptom profiling to predict suicidal ideation in depression, addressing a critical gap in translating population-level suicide risk models into clinically actionable tools. While numerous studies have confirmed associations between depression severity and suicidality,^4,33,34 and others have applied machine learning to suicide prediction using diverse data sources (eg, electronic health records, social media, multi-omics data),^16,17,19,25 our work specifically addresses the under-explored potential of leveraging “routine clinical scale data (HAMD-24)” for early suicide ideation detection within a standardized diagnostic framework.

Hypothesis 1: Model Performance

Our work bridges this gap through three key advances: First, by employing an ensemble-based Extremely Randomized Trees Classifier (ERTC), we not only confirmed known associations but also quantified the hierarchical symptom contributions (despair > guilt > inferiority > work/interest > depression) to suicidal ideation with unprecedented precision, providing symptom-level weights rarely captured in traditional analyses or population-level machine learning models. Second, we achieved clinically statistically significant prediction accuracy (AUC=0.80) using only routine HAMD-24 items—excluding the direct suicide query (Item 3) to avoid tautology—thereby demonstrating the scale’s latent capacity for risk stratification beyond its original design and distinct from studies relying on explicit suicide items or non-scale data. This approach mitigates response bias inherent in direct suicide questioning. Third, we demonstrated ERTC’s superior robustness over established algorithms (SVM, RF) in handling psychiatric symptom data,^16,21 attributable to its extreme random subspace sampling and node splitting, which mitigates overfitting and captures complex feature interactions inherent in depression psychopathology. This finding contrasts with studies where SVM or RF often performed best,^25,27 suggesting ERTC’s particular suitability for modeling complex symptom interactions within constrained scale data. This methodological synergy not only advances mechanistic understanding of suicide risk but delivers a scalable framework for real-world clinical decision support rooted in existing assessment practices.

Demographic patterns reinforced known epidemiological trends while underscoring clinical nuances. The heightened vulnerability of younger patients aligns with developmental psychopathology models positing that impulsivity, identity instability, and maladaptive coping peak in early adulthood, amplifying suicide risk when comorbid with depression.³⁵ The significant reduction in suicidal ideation among antidepressant users highlighted treatment’s protective role, potentially mediated by neurotransmitter modulation (eg, enhanced serotonin signaling reducing impulsivity) or psychological mechanisms (eg, restored agency).³⁶ This reinforces guidelines advocating prompt pharmacotherapy initiation in moderate-to-severe depression. Furthermore, a higher total HAMD-24 score (excluding item 3) was strongly associated with an increased likelihood of suicidal ideation, corroborating previous research linking the severity of depressive symptoms to suicide risk.³³

Hypothesis 2: Key Symptom Predictors

Consistent with Beck’s hopelessness theory, despair—conceptualised as negative future expectancy—was the single strongest predictor in our model (feature importance = 0.27), corroborating meta-analytic evidence that hopelessness accounts for 76% of the explained variance in suicidal ideation.^34,37 Theoretical frameworks position despair: conceptualized as a future-oriented cognitive schema characterized by negative expectations (“nothing will improve”) and perceived inescapability (“no way out”)³⁷—as an essential proximal cause of suicidal thoughts.^38–41 Notably, contemporary models integrate ambivalence—recognizing suicidal urgency often coexists with “countervailing motivations” (eg, fear of death, responsibility to family). This may explain why despair alone is insufficient to predict attempts; our model could be augmented by ambivalence metrics. Neurocognitively, this may reflect dysfunction in prefrontal cortical circuits responsible for future prospection and problem-solving, trapping individuals in a cycle of hopelessness. Our findings solidify despair’s critical role within the HAMD-24 framework, while also underscoring the clinical necessity of addressing closely ranked co-factors (guilt, inferiority), given the marginal quantitative difference in their importance scores. Mechanistically, despair may propagate suicidal ideation through two synergistic pathways: 1. Direct pathway: By inducing cognitive constriction—a narrowing of perceived options—where suicide is misappraised as the sole viable solution to unbearable psychological pain;^38,42 2. Indirect pathway: By eroding psychological resilience—the capacity to adaptively cope with adversity—through diminished self-efficacy, impaired access to positive memories, and reduced motivation to seek support, thereby disabling protective mechanisms.^43–45 This dual-pathway model underscores despair’s pernicious role in both motivating suicidal escape and disabling natural buffers against it.

The secondary prominence of guilt and inferiority complex the interpersonal dimension of suicide risk. Guilt—operationalized as “burdening guilt” involving irrational self-condemnation and perceived liability to others⁴⁶ and inferiority—rooted in chronic low self-worth and social comparison.⁴⁷ Align precisely with Joiner’s Interpersonal Theory of Suicide (IPTS).⁴⁰ IPTS posits that suicidal desire emerges from co-occurring “thwarted belongingness” (social alienation) and “perceived burdensomeness” (self-hate as a liability). Our findings validate this model within the HAMD-24 architecture: guilt embodies burdensomeness (“I am a drain on my family”), while inferiority fuels belongingness deficits (“I am unworthy of connection”). These factors likely contribute to suicidal ideation by fostering cognitive distortions, social isolation, and reinforcing feelings of hopelessness and worthlessness.^39,48 Notably, guilt’s strong predictive power persists despite potential scale-limited assessment in HAMD-24 versus BDI, where it exhibits denser symptom-network connectivity.⁴⁹ This suggests that guilt’s clinical relevance may be underestimated in HAMD-centric evaluations, warranting supplemental assessment when risk is elevated.

Work and interests loss and depressive emotions, though lower-ranked, reveal cyclical psychobehavioral mechanisms. Anhedonia-driven social withdrawal initiates a self-perpetuating cascade:^49,50 Isolation → eroded social support → exacerbated despair guilt → deeper depression → reinforced withdrawal. This “depression-withdrawal feedback loop” is neurobiologically scaffolded by reward system dysfunction (eg, ventral striatal hypoactivity), diminishing motivation for social engagement.⁵¹ Critically, withdrawal’s predictive value extends beyond functional impairment; it signifies disengagement from protective social anchors, depriving patients of reality testing, emotional scaffolding, and reasons for living.^52,53 Meanwhile, pervasive depressive mood—while a foundational symptom—may operate partly via its amplification effect on despair and guilt, illustrating how core affective and cognitive symptoms interact multiplicatively to elevate risk.⁵⁴

Clinical and Research Implications

Our ERTC model converts the standard administration process of the HAMD-24 scale into a dynamic risk-stratification tool. Clinicians can prioritize symptom remediation hierarchically (eg, targeting despair via cognitive restructuring before inferiority via social skills training). The quantified feature weights further guide scale refinement: Future depression-suicide risk instruments could amplify item weighting for despair/guilt or add nuanced guilt descriptors (eg, “feelings of being an unforgivable burden”). Methodologically, ERTC’s success advocates for algorithmic pluralism in psychiatric machine learning—avoiding overreliance on single models (eg, SVM-dominated literature) and leveraging ensemble methods for complex symptom interactions. In clinical work. In clinical practice applications, for early detection, we recommend complementing scale-based screening with: 1. Ecological Momentary Assessment (EMA) to capture real-time symptom fluctuations linked to triggers;^55,56 2. Natural language processing of clinical notes to identify linguistic markers of ambivalence (eg, coexisting hopelessness and future planning);^9,10 3.Actigraphy monitoring to detect behavioral correlates (eg, social withdrawal spikes).^57,58

Limitations and Future Directions

While impactful, our study has constraints: Single-center sampling limits generalizability; antidepressant type/dose was recorded but not standardized across participants, potentially confounding medication effects; the sample size, though sufficient for model development, warrants validation in larger cohorts; modest specificity risks false positives, necessitating secondary screening; and exclusive reliance on HAMD-24 omits biomarkers (eg, inflammatory markers, fMRI connectivity). Future work should: 1. Validate the model in multicentric, culturally diverse cohorts; 2. Incorporate ambivalence measures (eg, Death/Suicide Implicit Association Test) and multimodal digital phenotyping 3. Track temporal symptom dynamics to capture risk flux; 4. Improve specificity via hybrid models (eg, ERTC + neural networks).

Conclusion

Based on machine learning analysis of HAMD-24 data from 374 depressed patients, this study demonstrates that: 1. The ERTC model outperformed SVM, NBC, and RF in predicting suicidal ideation (accuracy: 77.75%, AUC: 0.80). Its superior robustness arose from enhanced randomization during tree construction, mitigating overfitting and improving generalization for capturing the complex interactions among HAMD-24 symptoms. This finding highlights ERTC as a particularly suitable algorithm for modeling complex symptom interactions within clinical scale data, contrasting with some prior studies favoring SVM or RF for similar tasks.^13,15 2. Critical predictors of suicidal ideation within the HAMD-24 (excluding Item 3) included feelings of despair, guilt, inferiority, loss of work/interest, and overall depressive mood. This granular identification and ranking of specific depressive symptoms as primary drivers of suicide ideation, validated by machine learning, provides a more nuanced understanding than simply associating overall depression severity (HAMD total score) with risk.^29,30 While despair ranked highest in feature importance, its quantitative lead over guilt and inferiority was marginal, suggesting these core symptoms collectively drive risk within the scale framework. 3. The quantification of the relative importance of specific HAMD-24 symptoms offers a novel, data-driven theoretical foundation for refining existing depression-suicide risk assessment scales or developing new, more targeted ones, moving beyond simply confirming associations to informing how specific symptoms contribute mechanistically to suicide ideation within the depression construct. Clinicians should prioritise cognitive-restructuring interventions targeting despair, guilt and inferiority; risk-stratify patients using an 80% probability cut-off from the ERTC model; and consider supplementing HAMD-24 screening with momentary ambivalence assessments for high-risk individuals.

Data Sharing Statement

The de-identified datasets (demographics, HAMD-24 item scores excluding Item 3, BSI-CV group labels) and Python code used for preprocessing, model training, and evaluation are available from the corresponding author on reasonable request under a data sharing agreement that ensures participant confidentiality and compliance with ethical guidelines.

Ethics Approval and Consent to Participate

This study was conducted in accordance with the declaration of Helsinki. This study was conducted with approval from the Ethics Committee of Nanjing Medical University Affiliated Changzhou Second People’s Hospital. A written informed consent was obtained from all participants.

Author Contributions

Conception and design of the research: Yun Chen; Guan-Zhong Dong; Acquisition of data: Yun Chen; Guan-Zhong Dong; Wei-Yuan Zhang; Ke Wang; Analysis and interpretation of the data: Zhong-Yi Jiang, Ke Wang; Statistical analysis: Zhong-Yi Jiang, Wei-Yuan Zhang; Obtaining financing: Hai-Yan Yang; Ke Wang; Writing of the manuscript: Yun Chen; Critical revision of the manuscript for intellectual content: Hai-Yan Yang. All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

None of the authors have any financial disclosure or conflicts of interest to report for this work.

References

1. Wang G, Feng Y, Liu J, et al. Chinese guidelines for the prevention and treatment of depression (2024 Edition) proposal. Chin Psychiatry Miscellaneous Zhi. 2023;56(6):407–412. doi:10.3760/cma.j.cn113661-20230919-00076

2. Huang Y, Wang YU, Wang H, et al. Prevalence of mental disorders in China: a cross-sectional epidemiological study. Lancet Psychiatry. 2019;6(3):211–224.

3. Dong M, Wang SB, Li Y, et al. Prevalence of suicidal behaviors in patients with major Depressive disorder in China: a comprehensive meta-analysis. J Affect Disord. 2018;225:32–39.

4. Orsolini L, Latini R, Pompili M, et al. Understanding the complex of suicide in depression: from research to clinics. Psychiatry Invest. 2020;17(3):207–221.

5. Fava GA, Kellner R, Munari F, Pavan L. The Hamilton depression rating scale in normals and depressives. Acta Psychiatr Scand. 1982;66(1):26–32. PMID: 7124430. doi:10.1111/j.1600-0447.1982.tb00911.x

6. Zhang M. Handbook of Psychiatric Assessment Scales. 2nd ed. Changsha: Hunan Science and Technology Press; 1998:121–126.

7. Teismann T, Siebert AM, Forkmann T. Suicidal ambivalence: a scoping review. Suicide Life Threat Behav. 2024;54(5):802–813. PMID: 38709556. doi:10.1111/sltb.13092

8. Ernst M, Gemke TJ, Olivi LJ, O’Connor RC. Ambulatory assessment of suicidal ambivalence: the temporal variability of the wish to live and the wish to die and their relevance in the concurrent and prospective prediction of suicidal desire. Suicide Life Threat Behav. 2024;54(5):831–843. PMID: 39096098. doi:10.1111/sltb.13120

9. Hogoboom A, Rouch M, Lauerman D, Pauselli L, Compton MT. Initial evidence of vowel space reduction in a subset of individuals with schizophrenia. Schizophr Res. 2023;255:158–164. PMID: 36989674; PMCID: PMC11371129. doi:10.1016/j.schres.2023.03.026

10. Lucarini V, Cangemi F, Daniel BD, et al. Conversational metrics, psychopathological dimensions and self-disturbances in patients with schizophrenia. Eur Arch Psychiatry Clin Neurosci. 2022;272(6):997–1005. PMID: 34476588. doi:10.1007/s00406-021-01329-w

11. Spulber S, Elberling F, Svensson J, Tiger M, Ceccatelli S, Lundberg J. Patterns of activity correlate with symptom severity in major depressive disorder patients. Transl Psychiatry. 2022;12(1):226. PMID: 35654778; PMCID: PMC9163191. doi:10.1038/s41398-022-01989-9

12. Petrosellini C, Eriksson SH, Meyer N, et al. Sleep disturbance as a marker of postpartum psychosis risk: a prospective actigraphy study. BMC Psychiatry. 2025;25(1):569. PMID: 40457330; PMCID: PMC12131421. doi:10.1186/s12888-025-07017-6

13. Lu J, Li L, Xu X. Interpretation of the Chinese guidelines for the prevention and treatment of depression (Second Edition): evaluation and diagnosis. Chin Psychiatry J. 2017;50(3):169–171.

14. Maatoug R, Gorwood P. The psychometrics of the hospital anxiety and depression scale supports a shorter −12 item- version. Psychiatry Res. 2019;274:372–376. PMID: 30852430. doi:10.1016/j.psychres.2019.02.074

15. Kim HJ, Kim S, Son Y, Youn I, Lee K. Reliability and validity of the Korean Version of the Ask Suicide-Screening Questions (ASQ). J Korean Med Sci. 2023;38(6):e41. PMID: 36786085; PMCID: PMC9925327. doi:10.3346/jkms.2023.38.e41

16. Yook V, Choi YH, Gu MJ, et al. Suicide Screening Questionnaire-Self-Rating (SSQ-SR): development, reliability, and validity in a clinical sample of Korean adults. Compr Psychiatry. 2023;121:152360. PMID: 36508776. doi:10.1016/j.comppsych.2022.152360

17. Linthicum KP, Schafer KM, Ribeiro JD. Machine learning in suicide science: applications and ethics. Behav Sci Law. 2019;37(3):214–222.

18. Li R. Data mining and machine learning methods for dementia re- search. Methods Mol Biol. 2018;1750:363–370.

19. Walsh CG, Ribeiro JD, Franklin JC. Predicting risk of suicide attempts over time through machine learning. Clin Psychol Sci. 2017;5(3):1032627260.

20. Miché M, Studerus E, Meyer AH, et al. Prospective prediction of suicide attempts in community adolescents and young adults, using regression methods and machine learning. J Affect Disord. 2020;265:570–578. PMID: 31786028. doi:10.1016/j.jad.2019.11.093

21. Ryu S, Lee H, Lee DK, Kim SW, Kim CE. Detection of suicide attempters among suicide ideators using machine learning. Psychiatry Invest. 2019;16(8):588–593. PMID: 31446686; PMCID: PMC6710424. doi:10.30773/pi.2019.06.19

22. Bhak Y, Jeong HO, Cho YS, et al. Depression and suicide risk prediction models using blood-derived multi-omics data. Transl Psychiatry. 2019;9(1):262. PMID: 31624227; PMCID: PMC6797735. doi:10.1038/s41398-019-0595-2

23. Yang L, Zhang Q, Bai S, et al. Using machine learning algorithms to predict the risk of suicide attempts among college students. Clin Psychol China J. 2023;31(3):525–529634. doi:10.16128/j.cnki.1005-3611.2023.03.005

24. Dong J, Wei W, Wu K, et al. The application of machine learning in the field of depression. Prog Psychol Sci. 2020;28(2):266–274. doi:10.3724/SP.J.1042.2020.00266

25. Qu J, Wu Y, Liu J, et al. Computational psychiatry: a new perspective on depression research and clinical application. Psychol Sci Prog. 2020;28(1):111–127. doi:10.3724/SP.J.1042.2020.00111

26. Lee YW, Choi JW, Shin EH. Machine learning model for predicting malaria using clinical information. Comput Biol Med. 2021;129:104151. PMID: 33290932. doi:10.1016/j.compbiomed.2020.104151

27. Karaismailoglu E, Karaismailoglu S. Two novel nomograms for predicting the risk of hospitalization or mortality due to COVID-19 by the naïve Bayesian classifier method. J Med Virol. 2021;93(5):3194–3201. PMID: 33599308; PMCID: PMC8013381. doi:10.1002/jmv.26890

28. Wallace ML, Mentch L, Wheeler BJ, et al. Use and misuse of random forest variable importance metrics in medicine: demonstrations through incident stroke prediction. BMC Med Res Methodol. 2023;23(1):144. PMID: 37337173; PMCID: PMC10280951. doi:10.1186/s12874-023-01965-x

29. Geng D, An Q, Fu Z, Wang C, An H. Identification of major depression patients using machine learning models based on heart rate variability during sleep stages for pre-hospital screening. Comput Biol Med. 2023;162:107060. PMID: 37290394; PMCID: PMC10229199. doi:10.1016/j.compbiomed.2023.107060

30. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Beijing: Peking University Press; 2015.

31. Lin JY, Huang Y, Su YA, et al. Association between perceived stressfulness of stressful life events and the suicidal risk in Chinese patients with major depressive disorder. Chin Med J. 2018;131(8):912–919. PMID: 29664050; PMCID: PMC5912056. doi:10.4103/0366-6999.229898

32. Li X, Fei L, Zhang Y, et al. The reliability and validity of the Chinese version of the Beck Suicide Intention Scale in college students. Chin psychol J Health. 2011;25(11):862–866. doi:10.3969/j.issn.1000-6729.2011.11.013

33. Li H, Luo X, Ke X, et al. Major depressive disorder and suicide risk among adult outpatients at several general hospitals in a Chinese Han population. PLoS One. 2017;12(10):e0186143. PMID: 29016669; PMCID: PMC5634639. doi:10.1371/journal.pone.0186143

34. Ribeiro JD, Huang X, Fox KR, Franklin JC. Depression and hopelessness as risk factors for suicide ideation, attempts and death: meta-analysis of longitudinal studies. Br J Psychiatry. 2018;212(5):279–286. PMID: 29587888. doi:10.1192/bjp.2018.27

35. Buerke M, Galfalvy H, Keilp JG, et al. Age effects on clinical and neurocognitive risk factors for suicide attempt in depression – findings from the AFSP lifespan study. Affect Disord. 2021;295:123–130. PMID: 34425314; PMCID: PMC8551053. doi:10.1016/j.jad.2021.08.014

36. Simon GE, Moise N, Mohr DC. Management of depression in adults: a review. JAMA. 2024;332(2):141–152. Erratum in: JAMA. 2024;332(15):1306. doi: 10.1001/jama.2024.18427. PMID: 38856993. doi:10.1001/jama.2024.5756

37. Beck AT, Steer RA, Kovacs M, Garrison B. Hopelessness and eventual suicide: a 10-year prospective study of patients hospitalized with suicidal ideation. Am J Psychiatry. 1985;142(5):559–563. PMID: 3985195. doi:10.1176/ajp.142.5.559

38. Beck AT, Kovacs M, Weissman A. Hopelessness and suicidal behavior: an overview. JAMA. 1975;234:1146–1149.

39. Joiner T. Why People Die by Suicide. Harvard University Press; 2005.

40. Van Orden KA, Witte TK, Cukrowicz KC, Braithwaite SR, Selby EA, Joiner TE. The interpersonal theory of suicide. Psychol Rev. 2010;117(2):575–600. PMID: 20438238; PMCID: PMC3130348. doi:10.1037/a0018697

41. Abramson LY, Alloy LB, Hogan ME, et al. The hopelessness theory of suicidality. In: Joiner T, Rudd MD, editors. Suicide Science. Springer US; 2000:17–32.

42. Wenzel A, Beck AT. A cognitive model of suicidal behavior: theory and treatment. Appl Prev Psychol. 2008;12:189–201.

43. Li Y. A study on the relationship between despair, psychological resilience, and suicidal ideation among college students. Chongqing Med J. 2014;2014(5):524–526.

44. Fedina L, Nam B, Jun HJ, et al. Moderating effects of resilience on depression, psychological distress, and suicidal ideation associated with interpersonal violence. J Interpers Violence. 2021;36(3–4):NP1335–1358NP. PMID: 29295024. doi:10.1177/0886260517746183

45. Okechukwu FO, Ogba KTU, Nwufo JI, et al. Academic stress and suicidal ideation: moderating roles of coping style and resilience. BMC Psychiatry. 2022;22(1):546. PMID: 35962365; PMCID: PMC9373522. doi:10.1186/s12888-022-04063-2

46. Leonardi J, Gazzillo F, Gorman B, Bush M. Assessing Burdening Guilt and Its Correlates. Psychodyn Psychiatry. 2023;51(4):479–499. PMID: 38047672. doi:10.1521/pdps.2023.51.4.479

47. Castro NB, Lopes MVO, Monteiro ARM. Low chronic self-esteem and low situational self-esteem: a literature review. Rev Bras Enferm. 2020;73(1):e20180004. English, Portuguese. PMID: 32049223. doi:10.1590/0034-7167-2018-0004

48. Jia H, Min Z, Yiyun C, et al. Association between social withdrawal and suicidal ideation in patients with major depressive disorder: the mediational role of emotional symptoms. J Affect Disord. 2024;347:69–76. PMID: 37992770. doi:10.1016/j.jad.2023.11.051

49. Feiten JG, Mosqueiro BP, Uequed M, Passos IC, Fleck MP, Caldieraro MA. Evaluation of major depression symptom networks using clinician-rated and patient-rated data. J Affect Disord. 2021;292:583–591. PMID: 34147971. doi:10.1016/j.jad.2021.05.102

50. Kato TA, Kanba S, Teo AR. Defining pathological social withdrawal: proposed diagnostic criteria for hikikomori. World Psychiatry. 2020;19(1):116–117. PMID: 31922682; PMCID: PMC6953582. doi:10.1002/wps.20705

51. Porcelli S, Van Der Wee N, van der Werff S, et al. Social brain, social dysfunction and social withdrawal. Neurosci Biobehav Rev. 2019;97:10–33. PMID: 30244163. doi:10.1016/j.neubiorev.2018.09.012

52. Evans GW, Rhee E, Forbes C, Mata allen K, Lepore SJ. The meaning and efficacy of social withdrawal as a strategy for coping with chronic residential crowding. J Environ Psychol. 2000;20:335–342.

53. Zhu S, Lee PH, Wong PWC. Investigating prolonged social withdrawal behaviour as a risk factor for self-harm and suicidal behaviours. BJPsych Open. 2021;7(3):e90. PMID: 33926603; PMCID: PMC8142544. doi:10.1192/bjo.2021.47

54. Iweama CN, Agbaje OS, Lerum NI, Igbokwe CC, Ozoemena LE. Suicidal ideation and attempts among Nigerian undergraduates: exploring the relationships with depression, hopelessness, perceived burdensomeness, and thwarted belongingness. SAGE Open Med. 2024;12:20503121241236137. PMID: 38533197; PMCID: PMC10964440. doi:10.1177/20503121241236137

55. Winstone L, Heron J, John A, et al. Ecological momentary assessment of self-harm thoughts and behaviors: systematic review of constructs from the integrated motivational-volitional model. JMIR Ment Health. 2024;11:e63132. PMID: 39652869; PMCID: PMC11667137. doi:10.2196/63132

56. Wenzel J, Dreschke N, Hanssen E, et al. Ecological momentary assessment (EMA) combined with unsupervised machine learning shows sensitivity to identify individuals in potential need for psychiatric assessment. Eur Arch Psychiatry Clin Neurosci. 2024;274(7):1639–1649. Erratum in: Eur Arch Psychiatry Clin Neurosci. 2025;275(3):959–962. doi: 10.1007/s00406-024-01939-0. PMID: 37715784; PMCID: PMC11422424. doi:10.1007/s00406-023-01668-w

57. Lebowitz ER, François B. Using motion tracking to measure avoidance in children and adults: psychometric properties, associations with clinical characteristics, and treatment-related change. Behav Ther. 2018;49(6):853–865. PMID: 30316485; PMCID: PMC6394864. doi:10.1016/j.beth.2018.04.005

58. Kagawa F, Yokoyama S, Takamura M, et al. Decreased physical activity with subjective pleasure is associated with avoidance behaviors. Sci Rep. 2022;12(1):2832. PMID: 35181696; PMCID: PMC8857298. doi:10.1038/s41598-022-06563-3

Stay informed with free updates

Simply sign up to the Pharmaceuticals sector myFT Digest — delivered directly to your inbox.

Incoming GSK chief executive Luke Miels is under pressure to boost investor confidence in the pharmaceutical group’s drug pipeline, following years of market scepticism under current boss Emma Walmsley.

Miels, a 50-year-old Australian, will take over the UK-listed drugmaker at the start of next year when Walmsley steps down. After eight years as chief commercial officer, he is known for his sharp execution and knack for uncovering “hidden gem” acquisitions. 

One top five investor said Miels had an “opportunity to reward shareholders”.

Walmsley has previously said she is confident GSK can navigate patent expiries in its HIV business towards the end of the decade and that annual revenue will increase from £31bn in 2024 to £40bn by 2031. But she has struggled to convince many investors. GSK’s share price has fallen about 10 per cent during her tenure and analysts forecast annual revenues of £33bn by 2031.

“I do believe [Miels] will be more convincing than Emma with the market,” the shareholder said. “He has had a successful track record before GSK. And he has proper ownership of his pipeline and he will now be in charge of asset allocation,” they added.  

Miels has previously worked at AstraZeneca, Roche and Sanofi. GSK shares rose more than 3 per cent on the day his appointment was announced.

The shareholder added that Miels would be judged on whether he can hit the £40bn sales target, but also needed to deliver “some milestones along the way so that people can gain comfort that GSK is on track”. 

Another shareholder said Miels had enough time before major patent expiries to deliver results, calling this a “strength of the succession process”.

“The key driver remains execution on the drug pipeline,” they said, adding that success in pharma was “essentially binary”. Companies that prove they can discover effective drugs trade on much higher price/earnings multiples. 

Miels came to GSK from AstraZeneca, where he had been close to chief executive Pascal Soriot. AstraZeneca sued for violation of his contract and court documents reported by the Sunday Times at the time said Soriot told Miels he could leave for any company but GSK. 

Since then, GSK has hired other staff from its UK rival, including key members of its M&A team. 

Miels has made dealmaking a major part of his job, devoting at least a day a week to it. He has championed the acquisition of under-appreciated assets that have become successful products, including a blood cancer drug and an antibiotic to treat urinary tract infections.

One executive close to Miels said he is willing to cut programmes that are not performing but happy to commit money to those he thinks have potential, including the antibiotic when there were many internal doubters. “The pace of business development will continue or take another step up in the next two or three years,” the executive said. 

Unlike Walmsley, who was criticised by activist investor Elliott Management for not having a scientific background, Miels has an undergraduate degree in science. He has spent his career on the commercial side of the business, but works closely with GSK’s chief scientific officer Tony Wood, who runs research and development.  

“He loves going toe to toe with R&D, and he loves the science,” the executive said. 

When Miels was appointed, GSK chair Sir Jonathan Symonds commented on his “outstanding global biopharma development and commercial experience”.

But another executive who worked with him earlier in his career said Miels could be “more than hyper focused” on the commercial plans but did not have a “passion for science”.

Announcing the appointment, Walmsley said she was “extremely pleased” with the smooth succession process. She said that when she started putting together her team in late 2016, she wrote in her diary that Miels would be a “dream appointment”. 

“All these years later, I absolutely couldn’t be more delighted that he’s picking up the baton.” 

 

But as expectations rise and the pace of change accelerates, a critical question emerges: at what cost? The drive to keep up with innovation and deliver results can stretch leaders and teams to their limits, putting their well-being and resilience at risk. Burnout is not a distant concern—it’s a real and pressing challenge in today’s boardrooms and beyond.

The forces at play

The modern boardroom is being called to step up with agility and foresight. Directors’ roles are growing in scope and complexity—no longer limited to monitoring and compliance, but demanding decisive, visionary leadership in the face of uncertainty. Their bandwidth to focus on consequential decisions, rather than being consumed by reactive decision-making, is under pressure. 

Perhaps nowhere is this more evident than with the rise of AI, which exemplifies both the opportunity and complexity boards must navigate. Deloitte’s latest survey reveals AI’s rapid rise is fueling a readiness to evolve: 53% of C-suite leaders want to accelerate AI adoption—but 66% say their boards lack sufficient knowledge or experience. The board’s challenge lies in deepening their expertise while being intentional about protecting their bandwidth given the sheer volume of information and pace of change. 

But technology is only part of the story. The future of work is evolving on multiple fronts, and boards must also balance the drive for innovation with the workforce’s desire for stability. Deloitte’s 2025 Global Human Capital Trends report introduces “stagility”—stability and agility—as an essential leadership capability. While 75% of workers hope for more stability, 85% of executives are willing to embrace change and focus on becoming agile as they adapt to rapid transformations. It’s the board’s responsibility to be aware of this tension and provide thoughtful oversight to help organizations strike the right balance. 

The strain and pressures directors are experiencing in the boardroom mirror the broader societal and systemic forces shaping today’s environment. While many factors are outside the board’s control, the opportunity to set the tone for well-being as an imperative—not a nice to have—can start with us. Modern leadership means championing both business and human outcomes. By prioritizing purposeful, resilient governance, we can help safeguard our bandwidth, inspire broader organizational well-being, and enable high-impact decision-making at scale. 

Intent over habit: governing at scale

You may be thinking, “yes, but how?” The answer may lie in anchoring board practices in clear purpose and adaptable structures. Too often, boards fall into the trap of doing things “the way they’ve always been done.” As stewards of the organization, directors often equate tradition with stability—especially when pressure and stakes are high. But many of today’s organizations look very different than at their inception—and governance practices should reflect that evolution.

To unlock the art of the possible, it’s important to commit to governance at scale, moving beyond traditional practices to meet the complexities of modern business. This means zeroing in on what truly drives value: establishing clear priorities. Leveraging technology and streamlining processes can enable boards to run efficient meetings and direct their attention to consequential issues—protecting bandwidth and empowering leaders to embrace “stagility.”

Governing at scale doesn’t require complicated solutions. Streamlining agendas and providing concise pre-read materials can allow directors to prepare thoughtfully and focus on strategic issues. Maximizing schedules by incorporating virtual or hybrid meeting formats can enable directors to stay refreshed and attentive, so they can contribute meaningfully. Bringing in outside experts for focused education sessions can expose directors to fresh perspectives and equip them to navigate emerging challenges with greater confidence. With the right guardrails, integrating AI and other emerging technologies can help boards decode complex issues faster, accelerate upskilling, and enhance decision-making. When directors are supported by intentional, streamlined board processes, they can gain the clarity and confidence to stay engaged and energized—enabling high-impact governance that inspires innovation, nurtures resilience, and drives sustainable growth.

The path forward

Prioritizing these practices at the highest levels is about more than just wellness; it can be a strategic advantage. The health of the boardroom is intrinsically linked to the health of the organization. Especially as AI and other forces reshape the landscape, organizations that invest in their board’s capacity to adapt and govern at scale can be better equipped to navigate disruption and shape the future with decisive, agile oversight.

Let’s commit to showing up authentically, supporting one another, and governing with intention and care. By embracing new ways of working and optimizing bandwidth, boardrooms and organizations can not only endure disruption, but capitalize on change. And while the path to resilience is ongoing, finding ways to track and evaluate it—just as we do with other key performance indicators—may be an essential step toward true accountability and sustained performance. Turn today’s challenges into tomorrow’s opportunities and help ensure your organizations remain resilient, innovative, and ready for whatever comes next. 

This publication contains general information only and Deloitte is not, by means of this publication, rendering accounting, business, financial, investment, legal, tax, or other professional advice or services. This publication is not a substitute for such professional advice or services, nor should it be used as a basis for any decision or action that may affect your business. Before making any decision or taking any action that may affect your business, you should consult a qualified professional advisor. Deloitte shall not be responsible for any loss sustained by any person who relies on this publication.

Copyright © 2025 Deloitte Development LLC. All rights reserved.

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.

Fortune Global Forum returns Oct. 26–27, 2025 in Riyadh. CEOs and global leaders will gather for a dynamic, invitation-only event shaping the future of business. Apply for an invitation.

Welcome to OncLive®’s OncFive!

Every week, we bring you a quick roundup of the 5 top stories from the world of oncology—ranging from pivotal regulatory decisions to key pipeline updates to expert insights on breakthroughs that are moving the needle in cancer care. This resource is designed to keep you informed on the latest updates in the space, in just a matter of minutes.

Here’s what you may have missed this week:

The FDA cleared cemiplimab-rwlc (Libtayo) for adjuvant use in adult patients with cutaneous squamous cell carcinoma (CSCC) at high risk of recurrence after surgery and radiation. The decision was supported by findings from the phase 3 C-POST trial (NCT03969004), in which cemiplimab reduced the risk of recurrence or death by 68% vs placebo (HR, 0.32; 95% CI, 0.20-0.51; P < .0001), with median disease-free survival not reached in the cemiplimab arm. Recurrence occurred in 9% of patients in the cemiplimab arm vs 30% of those in the placebo arm. The toxicity profile was consistent with prior findings, with common adverse effects (AEs) including rash, pruritus, and hypothyroidism. The decision signifies the first and only immunotherapy approved for adjuvant treatment in those with CSCC with a high risk of recurrence following surgery and radiation.

The FDA has issued a complete response letter (CRL) to the new drug application (NDA) for dasatinib (Dasynoc) for use in patients with chronic myeloid leukemia (CML). The CRL stems from Good Manufacturing Practice observations at the company’s contract manufacturing site, prompting a temporary pause in new product approvals until corrective actions are completed. The manufacturer has begun remediation efforts and plans to meet with the regulatory agency later this year to address outstanding issues. The NDA for dasatinib sought approval for a formulation designed to maintain efficacy at lower doses, reduce pharmacokinetic variability, and minimize drug-drug interactions associated with acid-suppressive agents. Despite manufacturing delays, the drug remains under active development, with plans for expedited resubmission once compliance measures are met.

The regulatory agency also accepted and granted priority review to the biologics license application (BLA) seeking approval of Orca-T for select patients with acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL), and myelodysplastic syndrome (MDS). The application is supported by data from the phase 3 Precision-T trial (NCT04013685), in which Orca-T significantly improved 1-year moderate-to-severe chronic graft-vs-host disease (cGVHD)–free survival vs conventional allogeneic hematopoietic stem cell transplant (allo-HSCT; 78% vs 38%; HR, 0.26; P < .00001). Those who received Orca-T also experienced lower rates of moderate-to-severe cGVHD (13% vs 44%) and grade 3/4 acute GVHD (6% vs 17%) vs allo-HSCT. One-year overall survival (OS) was 94% in the Orca-T arm and 83% in the allo-HSCT arm, and non-relapse mortality rates were 3% vs 13%, respectively. The target action date for the BLA is set for April 6, 2026.

Data from the phase 3 TROPION-Breast02 trial (NCT05374512) showed that datopotamab deruxtecan-dlnk (Dato-DXd; Datroway) significantly improved OS and progression-free survival (PFS) vs investigator’s choice of chemotherapy in patients with locally recurrent, inoperable, or metastatic triple-negative breast cancer (TNBC) not able to receive immunotherapy. The safety profile of Dato-DXd was consistent with prior breast cancer studies. AstraZeneca and Daiichi Sankyo will share these results with regulatory authorities, highlighting the potential of the antibody-drug conjugate as the first therapy to significantly improve OS in this population.

The FDA has granted fast track designation to the TEAD autopalmitoylation inhibitor VT3989 for use in patients with unresectable malignant pleural or non-pleural mesothelioma who have progressed after previous immune checkpoint inhibition and platinum-based chemotherapy. Preliminary findings from an ongoing phase 1/2 trial (NCT04665206) demonstrated that the agent led to reductions in target lesion size in both pleural and non-pleural mesothelioma, irrespective of NF2 mutational status. The multicenter trial includes dose-escalation, dose-expansion, and combination cohorts, which are evaluating safety, antitumor activity, and pharmacokinetics of the agent, with oral dosing ranging from 25 mg to 200 mg. Common treatment-related AEs included albuminuria, proteinuria, fatigue, peripheral edema, and gastrointestinal and hepatic effects.

Honorable Mentions

• OncLive spoke with the following leading genitourinary oncology experts to highlight the most anticipated abstracts and data at the 2025 ESMO Congress: David A. Braun, MD, PhD, of Yale Medical School; Alan Tan, MD, of Vanderbilt University Medical Center; and Axel Merseburger, MD, PhD, of University Hospital Schleswig-Holstein. Check out the exclusive preview ahead of the meeting.
• From further clarifying the role of PD-(L)1 inhibition in gastric cancer and colorectal cancer, to demonstrating the potential predictive value of minimal residual disease negativity in late-stage colon cancer, gastrointestinal oncology data presented at the 2025 ESMO Congress are anticipated to challenge current standards of care and bring new questions to the forefront of research in the field. In this exclusive preview, Tanios S. Bekaii-Saab, MD, of Mayo Clinic; Suneel Kamath, MD, of Cleveland Clinic Lerner College of Medicine of Case Western Reserve University; and Kanwal P. S. Raghav, MBBS, MD, of The University of Texas MD Anderson Cancer Center, share their most anticipated abstracts.

Less than 10 years after dropping out of New York University and then starting what would become the prediction market Polymarket in the bathroom of his Lower East Side apartment, Shayne Coplan has been crowned the youngest ever self-made billionaire by capitalizing on Gen Z’s readiness to bet on anything.

On Tuesday, the New York Stock Exchange’s parent company, Intercontinental Exchange, invested $2 billion cash in Polymarket, skyrocketing the company’s valuation to $9 billion post investment and making CEO and founder Coplan a billionaire at the age of 27, according to the Bloomberg Billionaire Index.

Through the partnership, the NYSE will distribute Polymarket’s data and the two companies will work together on tokenization initiatives, according to a press release.

Polymarket has a simple premise: Markets are the best way to source truth. By giving users a stake in predicting literally everything, from the 2025 World Series Champion to when the government shutdown will end (both bets are currently available on its website), Polymarket aims to “harness the power of free markets to demystify real events that matter most to you,” Coplan said in a post on X.

Polymarket matches users with opposing bets and pays out $1 per “share” for every correct guess with the help of a U.S. dollar-backed stablecoin and a blockchain built on top of Ethereum’s infrastructure. This means if you bet “yes” on an outcome at 37 cents and it proves to be true, you’ll net a 63-cent profit. You can also sell your stake in an outcome before the event happens, which can also net you a profit if the price of your shares go up as the outcome you chose becomes more likely.

Aleksandar Tomic, an economist and associate dean for strategy, innovation, and technology at Boston College, said prediction markets like Polymarket have existed before. A similar platform, Intrade, received widespread media attention for its success in predicting the 2008 and 2012 U.S. elections before shutting down in 2013. Polymarket and its competitors seem to be succeeding where others have failed. Polymarket for its part has drawn in younger users with a better platform and by seizing upon the pandemic-era gambling trend especially prevalent in Gen Z men—not to mention with a little help from a newly friendly regulatory environment.

“I think these types of markets are just another place to make a bet,” Tomic told Fortune.

Polymarket did not immediately respond to a request for comment.

Until last month, Polymarket was banned in the U.S., largely due to federal regulators’ objections to its “speed over scrutiny” business model. In 2022, the company paid a $1.4 million fine after the Commodity Futures Trading Commission (CFTC) said it was allegedly operating an unregistered event market. The platform was subsequently barred from the country. As wagers on the 2024 election grew last year, the CFTC renewed its scrutiny, and the FBI raided founder Shayne Coplan’s home in November. Just under a year later—following President Trump’s return to office, and with Donald Trump Jr. now serving as an adviser to the company—the CFTC and Justice Department closed their investigations without filing charges, clearing the way for Polymarket’s return to the U.S.

Despite Polymarket’s many headwinds over the years, Coplan has previously pointed out some of the company’s notable successes on X, including predicting that President Biden would drop out of the 2024 presidential election. In total, Polymarket users put at least $3.2 billion into the presidential election, and one French trader even made around $85 million from his contrarian multimillion-dollar bet that President Donald Trump would return to the Oval Office.

With this week’s announcement of the team-up and investment from the New York Stock Exchange’s parent company, Coplan is riding high, while still reflecting on his journey upwards.

“At the onset of the pandemic, I quite literally had nothing to lose: 21, running out of money, 2.5 years since I dropped out and nothing to show for it,” he wrote on X. “But I knew we were entering an era where ways to find truth would matter more than ever, and Polymarket could play a critical role in that.”

On Sunday, he plans to watch football and beta-test the new Polymarket U.S. app, he wrote on X.

This story was originally featured on Fortune.com

By Christine Ji

Anthropic’s brand-awareness effort drew thousands as the company competes for consumer attention amid increasing competition

Claude’s pop-up at the Air Mail newsstand in New York drew over 5,000 people over the course of a weekend.

The line stretched for blocks down the street in Manhattan’s West Village neighborhood. Hundreds of people stood in the heat of an unseasonably warm October weekend – not for a sample sale or celebrity sighting, but a pop-up hosted by the artificial-intelligence company Anthropic.

For seven days in early October, Anthropic’s large language model Claude was the brand-in-residence at the Air Mail newsstand, the physical outpost for the digital magazine founded by former Vanity Fair editor Graydon Carter.

Like any good pop-up seeking to attract New Yorkers, Claude’s involved a variety of free merchandise: coffee cups, postcards, tote bags and matchbooks adorned with designs from Anthropic’s in-house illustrator; baseball caps embroidered with the word “thinking”; and copies of Chief Executive Dario Amodei’s essay, “Machines of Loving Grace,” wrapped in navy cloth and printed on locally sourced and 100% postconsumer recycled paper.

By the time I stepped into the brick storefront on the pop-up’s last day, the caps and essays were long gone, but the space was still crowded for a Tuesday afternoon.

Claude’s partnership with Air Mail was the physical manifestation of the company’s recent brand campaign called “Keep Thinking,” Sam McAllister, member of staff at Anthropic, told me. We sat on Air Mail’s back patio, surrounded by clusters of people reading, working on their laptops and talking among themselves. Claude is the AI “built to help people think through their hardest problems,” according to McAllister. “We don’t need to distract them with anything else.”

People now use AI to cheat on homework, generate low-quality social-media videos or get responses like “that’s exactly right!” to everything they input. In the increasingly commoditized world of LLMs, Anthropic seems to be positioning Claude as something different – a higher-brow AI tool that encourages users to think of problems that can be solved with more information, rather than as a way to outsource their own thinking.

It’s not a conventional AI campaign, as Claude’s predecessors in the Air Mail location include luxury names like Bottega Veneta and Ralph Lauren. And New York is not the easiest place to market an AI product – something the wearable AI startup Friend.com discovered when its recent subway advertisements were met with anti-AI vandalism.

“New York City is the capital of the world and therefore the top domino to go after in consumer marketing,” Friend.com CEO Avi Schiffman later told MarketWatch.

Wall Street currently views AI as primarily a consumer product and not a business application, according to Bryan Wong, portfolio manager at Osterweis Capital Management. In his opinion, enterprise adoption has been limited beyond areas of coding and customer support.

The path to large-scale profits from business applications will be “gradual,” said Jordan Klein, an analyst on the trading desk at Mizuho Securities. As a result, investors anticipate that most models will need to incorporate some form of ad placement to capitalize on the faster-growing consumer segment, Klein told me.

The major LLMs are trying to promote themselves widely to consumers, and branding has become an important part of the AI battle. Meta Platforms Inc. (META), OpenAI and Alphabet Inc.’s (GOOGL) (GOOG) Google Gemini have recently rolled out video-generation capabilities. Gemini shot to the top of Apple Inc.’s (AAPL) U.S. App Store last month after the launch of its image editor Nano Banana. Earlier this week, OpenAI’s short-form video app Sora reached 1 million downloads in less than five days.

More users and engagement leads to more advertising revenue – an important consideration for AI companies looking to fund the extremely expensive development of the technology, according to Klein.

OpenAI declined to comment for this article, while Google did not immediately respond to a request for comment.

Anthropic – a four-year-old private startup that counts Amazon.com Inc. (AMZN) and Google among its major investors – has doubled down on productivity with its latest model, adding the ability to generate PDFs, PowerPoints and Excel files. And it’s trying to take that message to consumers with a branding campaign that includes its Manhattan pop-up.

“We prioritize outputs like text, files, code and creative applications,” Anthropic’s McAllister told me. In an age where AI “slop” – a term used to refer to low-quality media that’s fast and easy to produce – proliferates on the internet and in real life, Claude aims to be the “antidote to that,” he added.

Also read: OpenAI wants to build a social-media business. Can its Sora app take on Meta and Google?

A ‘thinking space’

“I found out about this event yesterday,” Poorva Patel, a recent engineering graduate, told me at the pop-up. By 6 p.m. on Monday, she had hopped into her car and departed from her hometown of Columbus, Ohio.

“I was exhausted by midnight, but I was like, ‘I have to make it no matter what,’” Patel said.

After 10 hours of driving, she had arrived in New York on Tuesday morning and had spent the day meeting new people, exploring the Air Mail space and enjoying free tea and merchandise.

Later in the afternoon, another attendee, Sanam Ghaneeian, introduced herself to me and asked if I had experience “vibe coding.” (I don’t, but I suppose the point of AI tools is to allow anyone to code – even journalists.) A founder and marketing strategist, Ghaneeian had just moved to New York on Monday from Los Angeles to work on her startup. She had heard about the pop-up online and decided to stop by to do some work.

Ghaneeian rotates through a variety of AI tools in her everyday life, but she told me that Claude has been particularly “transformative” for her. “It’s allowed me to vibe-code websites, build custom dashboards and design entire visual systems in under 24 hours,” Ghaneeian said.

McAllister recounted his feeling of shock and “mild panic” upon seeing a line of over 100 people in front of the pop-up last Saturday, more than an hour before opening. The line only grew longer as the day went on. McAllister and his team handed out ice cream and demonstrated a gadget called Poetry Camera, which uses Claude to analyze photos and print a corresponding poem.

“We definitely didn’t order enough merchandise, that’s for sure,” he laughed.

McAllister and his team’s goal of creating a physical “thinking space” yielded over 5,000 visitors who came to the pop-up during the course of the weekend.

“I think Anthropic really cares about what its users want,” Patel told me. “I love how it’s emphasizing the human side of AI technology.”

Where’s the money?

Forgoing AI-generated visual media is a deliberate choice – one that many on Wall Street characterize as an unwise business move.

But outside of the finance world, the success of Claude’s pop-up shows that there’s a growing dissatisfaction with AI’s role in generating low-effort content in cyberspace. On X, Meta’s announcement of its new short-form AI video app, Vibes, was met with a wave of comments calling the product “slop.”

According to a Meta spokesperson, the company’s aim with Vibes is to provide free AI tools that allow anyone to create and share high-quality content. The spokesperson added that Meta’s current priority is to build a strong consumer experience and gather feedback on how people are using the feature.

By centering AI and real-world experiences, the pop-up catapulted Claude to internet virality, with over 10 million impressions across social media over the course of a weekend. But influencing customer sentiment on Anthropic will require more than free hats.

Anthropic leads the enterprise LLM API market with 32% market share and over 300,000 business customers, surpassing OpenAI’s 25% share. That’s according to a report from Menlo Ventures, one of Anthropic’s leading investors.

“It allows us to have a consumer offering that isn’t beholden to tactics like maximizing app downloads or shipping short-form video just to get more attention and more eyeballs,” McAllister said of Claude’s strong enterprise presence.

However, according to a Morgan Stanley survey of chief information officers, Anthropic is positioned to capture just 2% of incremental enterprise AI spending in 2025 – trailing behind OpenAI, Google, Amazon and Microsoft Corp (MSFT).

Read: IBM’s stock rises toward a record. Why its Anthropic deal symbolizes a new frontier in AI.

In the eyes of investors, Anthropic will need to translate the energy from the pop-up into consumer app downloads, where it’s lagging other players. The odds of Claude becoming the “best AI at the end of 2025” are at 4% on prediction market Kalshi right now, and Claude makes up less than 1% of total mobile-app downloads in the consumer LLM market, according a J.P. Morgan report by analyst Brenda Duverce.

That’s not to say that Anthropic isn’t finding success with its antislop agenda. Its run-rate revenue grew from under $1 billion at the beginning of 2025 to over $5 billion as of August, according to the company.

If Anthropic opts out of the AI visual-media market, J.P. Morgan’s Duverce believes it will be difficult to grow its already low consumer adoption relative to competitors.

“Its more limited consumer features/offerings relative to leading competitors” will lead Anthropic to “struggle to capture meaningful consumer share,” Duverce wrote. Fierce competition among AI models could lead to the commoditization of products, she added.

(MORE TO FOLLOW) Dow Jones Newswires

10-11-25 0800ET

Copyright (c) 2025 Dow Jones & Company, Inc.

Being a dedicated sort, I tasted all 10 of these mustards straight from the jar. With watering eyes, pumping endorphins and overactive salivary glands, I licked each mustard off a spoon, then quickly cleansed my palate with plenty of milk and water to subdue the heat. But the sandwich I had at the end, with my new favourite mustard, was worth every fiery spoonful.

Mustard is intriguingly complex in flavour – powerful, umami, hot – yet easy to make. At its simplest, it is little more than fermented mustard seeds soaked in brine until viscous and bubbling. Sadly, however, many modern brands, especially the cheaper ones, see fit to include unnecessary additives, such as xanthan gum and wheat flour. And, rather predictably, the cheaper the mustard, the less actual mustard it tends to contain.

Much as with mayonnaise, many processed mustards contain spirit vinegar instead of more subtle acids, such as white-wine or cider vinegar. Spirit vinegar is harsh and sharp, dominates the overall flavour and can result in an unappealingly aerated, mousse-like texture. In fact, six of the mustards I tasted were so similar, I wouldn’t be at all surprised if they were all made in the same factory. They were essentially carbon copies of each other – sulphuric, sweet, hot and moussey – even if they did still bear the hallmark fieriness and complexity of a classic English mustard.

So, when looking for a mustard that’s worth your money, seek out those with real vinegar rather than spirit vinegar, and without emulsifiers or texturing agents such as wheat flour; they’re simply unnecessary. Or, better still, make your own: you can easily produce a year’s worth of mustard with just five minutes’ work.

---

The best supermarket English mustard

---

Best overall:
Stokes classic English mustard

£2.90 for 185g at Ocado (£1.57/100g)

★★★★☆

Satin-sheen gold with a sweet, clean aroma. It has that lightly sour and bitter aftertaste typical of English mustard, and a creamy yet slightly granular texture. A simple, well-executed product that delivers on all fronts.

---

Best bargain:
M&S hot English mustard

£1 for 180g at Ocado (56p/100g)

★★★★☆

Buttercup yellow with fresh mustard leaf aroma. Made with mustard, salt, lemon juice concentrate and turmeric – and no emulsifiers. The heat builds gradually, it’s less sweet than most, and has a lovely, smooth pureed texture. Exceptional quality for the price.

---

And the rest …

Tracklements strong English mustard

£2.70 for 140g at Ocado (£1.93/100g)

£3.08 for 140g at Field & Flower (£2.20/100g)

★★★☆☆

School-bus yellow, and super-fiery and intense: this thick, simple paste made with pureed mustard seeds hits the nose like wasabi. Good ingredients, and well worth its Great Taste star. One for mustard purists.

---

Colman’s original English mustard

£2 for 170g at Tesco (£1.18/100g)

£1.70 for 100g at Sainsbury’s (£1.70/100g)

★★★☆☆

Bright yellow with a classic aroma and touch of egg. Really hot, though subdued in a ham sandwich. Quite acidic and powerful, with a thin, puree-like texture. Contains wheat flour and xanthan gum.

---

Stamford Street Co English mustard

55p for 180g at Sainsbury’s (31p/100g)

★★☆☆☆

Turmeric yellow, with a pickled egg and salt-and-vinegar-crisp aroma. Instant strong heat on the tongue, sweet with slight complexity and a powdery texture. Contains acetic acid, wheat flour and xanthan gum. A fiery, budget-friendly mustard.

skip past newsletter promotion

after newsletter promotion

---

Tesco English mustard

55p for 190g at Tesco (29p/100g)

★★☆☆☆

Rubber-duck yellow with a sulphuric aroma. Salty up front, followed by heat and sweet, complex mustard notes. Slightly moussey in texture, and contains wheat flour and xanthan gum. An affordable and reliable everyday mustard.

---

Sainsbury’s English mustard

55p for 185g at Sainsbury’s (30p/100g)

★★☆☆☆

Saffron yellow, with a predominant vinegar aroma, strong sulphuric flavour and bitter notes, all complemented by an enjoyably spiky heat. Aerated and contains wheat flour.

---

Asda English mustard

49p for 180g at Asda (27p/100g)

★★☆☆☆

Honey yellow, with a reassuringly familiar sweet aroma and a heat that develops and grows, with sulphurous mustard notes. Gloopy texture, though, and contains wheat flour and xanthan gum.

---

Waitrose Essential English mustard

80p for 180g at Waitrose (44p/100g)

★★☆☆☆

Classic dark yellow, with a strong spirit vinegar aroma. Extra-hot, with a sweet background and a pronounced eggy mustard flavour. Standard emulsified texture, and contains xanthan gum.

---

Batts English mustard

59p for 185g at Lidl (32p/100g)

★☆☆☆☆

Canary yellow, with a strong spirit vinegar aroma. Very sweet, but with a fiery heat and finishing on bitter mustard notes. Gloopy texture from xanthan gum, and also contains wheat flour. Still, a pretty decent budget mustard.

For more, read the best kitchen knives for every job – chosen by chefs