Introduction
Nontuberculous mycobacteria (NTM) are environmental pathogens with over 190 identified species,1 primarily affecting individuals with preexisting lung disease or immunodeficiency.2 There is a rising trend in incidence of infection and mortality globally, including among younger populations. Among children, the infection may occur with or without pre-existing lung disease.3 Risk factors among adults have been well studied, but there are fewer pediatric studies in this area of study.4 Past research has found that immunodeficiency, the developing immune system, and chronic lung disease heighten the risk of NTM infections in the pediatric population.5 NTM infections are especially prevalent because of their ability to inhabit common soil and water sources such as water distribution systems, allowing repeated environmental exposures to invade the lungs through bioaerosols.5 It is also thought that different species and subspecies of mycobacteria containing genetic differences manifest themselves differently in infection and clinical response.6 For example, two of the most common NTM species Mycobacterium avium and Mycobacterium abscessus differ in their infections with the former being characterized by slow growth, greater general prevalence, and weaker biofilm formation,7 while the latter is associated with significant morbidity/mortality, is clinically resistant to most antibiotics, and shows greater general immune responses throughout the course of infection.8 Infections from different types of these bacteria can vary by geographic location5 with Mycobacterium abscessus being commonly observed in East Asia and Mycobacterium avium complex and Mycobacterium kansasii occurring most commonly in many parts of the US. In Europe, Asia, and partially in Australia, increases in latitude generally see higher rates of Mycobacterium avium complex.
NTM pulmonary disease is frequently misdiagnosed as tuberculosis due to similar symptoms, leading to delays in appropriate treatment. One of the most common signs of NTM infection in healthy children is cervical lymphadenitis, but its clinical presentation is indistinguishable from that of cervical lymphadenitis in regular tuberculosis.9 Pulmonary infection, while less commonly associated may be more difficult to treat. This is especially problematic in countries with high tuberculosis rates, where NTM testing methods are resource limited.10–12 Industrialized nations often report higher NTM incidence than tuberculosis, although pediatric prevalence globally is not well studied.13,14
This study was conducted to investigate pediatric pulmonary NTM infections among the non-cystic fibrosis population. The global incidence of NTM infections is increasing, and more studies are needed to better understand the disease among the pediatric population.15 Unlike Cystic Fibrosis patients, who are known to have higher susceptibility to pulmonary infections,16 the risk factors and clinical manifestations in non-cystic fibrosis pediatric population remains poorly understood. This study aims to investigate the epidemiology, clinical comorbid condition, and five year clinical outcomes of non CF pulmonary NTM infections in distinct pediatric age groups. We hypothesize that there are important differences in these key areas among pediatric age groups. This may lead to the identification of characteristics that could guide clinical decision-making and potentially improve patient outcomes.
The study further hypothesizes that: i) BMI percentiles mediate susceptibility to NTM infections in children, ii) common comorbidities are expected to be significant risk factors for pediatric pulmonary NTM infections, iii) pediatric pulmonary NTM may predispose patients to future lung disease.
Methods
The TriNetX Clinical Data Platform
We utilized the TriNetX research platform to select our patient cohorts from a total of 152,714,105 collected, de-identified patient records on the Global Collaborative Network containing 127 health-care organizations from the following 17 countries: United States, Canada, United Kingdom, Germany, France, Italy, Spain, Netherlands, Denmark, Australia, Singapore, Japan, Brazil, Mexico, Israel, South Korea, and Switzerland. TriNetX deidentifies and aggregates electronic health record (EHR) data from health-care systems, primarily drawing from large academic medical institutions across the USA. It organizes diagnoses under specific ICD codes and stores information such as demographics, medications, lab results, procedures, and vital signs. The platform provides a secure, web-based access to patient-level analyses and interpretation. It reports updated population-level data while ensuring Health Insurance Portability and Accountability Act (HIPAA) compliance.
Data Collection and Stratification
From the TriNetX Global Collaborative Network, the pediatric patient population was stratified into four groups: i) 0–2 years, ii) 3–5 years, iii) 6–12 years, and iv) 13–18 years based on age. The inclusion criteria were: i) subjects with pulmonary NTM infection, and ii) ages 0–18 years. The exclusion criteria were: i) Cystic Fibrosis, ii) Tuberculosis, iii) smoking history, iv) cutaneous non-mycobacterial infections, and v) adult patients. The total cohort (0–18 years) consisted of 2,344 NTM cases from a larger population base of over 23 million pediatric patients collected on July 13, 2024. The distribution among the four groups was expressed both numerically and as a percent of patients in each respective age group with NTM.
Data Analysis
Data analysis focused on patient demographics, clinical characteristics, comorbidities, and outcomes of the 2,344 NTM cases, with particular attention to odds ratios and lab values associated with these infections. Descriptive analysis for this data included prevalence, mean values, and standard deviation within the age cohorts. Demographic information included age, sex, ethnicity, and race. Comorbidities were identified in our analysis. Clinical data were collected including mean values for BMI and oxygen saturation.
Statistical analysis was performed to compare the prevalence and outcomes of NTM across the different age groups, and to evaluate the different comorbidities experienced by each age group. Specifically, the six most common five-year outcomes were compared across the cohorts, assessing risk difference, confidence interval, risk ratio, odds ratio, and statistical significance to determine risk levels between pairs of groups. The analysis was conducted through the built-in analytics tools in the TriNetX software where a p value of <0.05 was defined as statistically significant. Our patients were selected using the inclusion and exclusion criteria outlined in Table 1. Descriptive statistics (mean, SD, proportion) were reported for demographics and clinical features. Comparative analysis across age groups was conducted using Chi-square tests for categorical variables and Student’s t-test or ANOVA for continuous variables. To evaluate associations between NTM and clinical outcomes across age groups, we conducted: Logistic regression analysis for binary outcomes to generate odds ratios (OR), Binomial regression to estimate risk ratios (RR) when appropriate. All models were adjusted for key covariates including age, sex, race, and comorbidity burden. Point estimates were reported with 95% CI.
|
Table 1 Inclusion and Exclusion Criteria Entered into TriNetX
|
Ethics
The University of California, Riverside IRB determined that the current study was exempt from further ethics review as it did not fit under the federal definition of human subjects research [DHHS 45 CFR46.102(e), 46.102(l)] or clinical investigation [FDA 21 CFR 50.3(c), 56.102(c)].
Results
Demographics
The demographic data provided insight into the prevalence and distribution of pediatric pulmonary NTM infections among the age cohorts, and with respect to sex, race and ethnicity. The study cohort comprised 2,344 cases stratified across four age groups: 0–2 years, 3–5 years, 6–12 years, and 13–18 years. The majority of NTM cases were observed in the 6–12 year age group at 1074 (734/100,000), followed closely by the 13–18 year age group at 760 (848/100,000). The 3–5 year age group had 401 patients (1261/100,000) and the 0–2 group had 109 patients (689/100,000) (Table 2). There were gender differences in the prevalence of pediatric pulmonary NTM lung infections (Table 3). There was a higher mean prevalence of NTM infections amongfemales compared to males 53.31% vs 46.29% among all age groups. There was no identified gender for 4.17% of the cohort. The majority of the cohort was identified as white 57.8%, followed by Black/African American, 8.88%, Asian, 4.49%, Native American, 2.83%, Native Hawaiian, 1.19%, another race, 8.52%, and unspecified race 21.6%. BMI percentiles were above the 50th percentile for all age groups, and this was statistically significant (p < 0.05) (Table 4).
 |
Table 2 Age Demographics of Our Study Cohort, Including the Number of Patients in Each Age Cohort
|
 |
Table 3 Sex Demographics of Our Study Cohort, Stratified by Age
|
 |
Table 4 BMI Percentiles of Each Age Cohort
|
Laboratory Findings
The data indicated a varying prevalence of NTM infections across the age groups, with the highest prevalence observed in children aged 6–12 years at 1,074 cases (0.007339%). Lab values, particularly those related to immune function, revealed a trend towards higher inflammatory markers among older children, possibly reflecting a more robust immune response or delayed diagnosis. There was an elevated mean white blood cell count in the 13–18 age group at 18.5×103/μL compared to the expected range of 4.0×103/μL to 11.0×103/μL (Table 5). Similarly, CRP mean levels for this age group were significantly above the expected value of 1 mg/L at 26.8 mg/L. In the 3–5 year and 13–18 year cohorts, basophil levels were above the expected 0–1% at 2.68% and 1.83%, respectively. Ferritin levels were higher in the 6–12 year and 13–18 year groups at 546 ng/mL and 1758 ng/mL, respectively, compared to the normal range of 10–200 ng/mL. Among younger children, there were lower levels of inflammatory markers, which may suggest an underdeveloped immune response or early-stage infection.
 |
Table 5 Comparison of Selected Inflammatory Markers by Age Cohort
|
Comorbid Conditions
Comorbidities such as acute pharyngitis, pneumonia, asthma, malignancy, and immunodeficiency were identified. (Table 6 and Figure 1). The most common comorbidity among the groups was acute pharyngitis in the 13–18 age group at 30% of the cohort. There was a higher burden of NTM infections among the older pediatric group, ages 13 to 18 years. This group also had the highest proportion of each comorbidity: 18% pneumonia, 22% asthma, 12% malignancy, and 11% immunodeficiency. The overall proportion of individuals with comorbidities was lower in the 6–12 year age group, still with acute pharyngitis leading at 19%, pneumonia at 12%, asthma at 13%, malignancy at 6%, and immunodeficiency at 7%. The 3–5 year old cohort had the following comorbid conditions: 11% acute pharyngitis, 7% pneumonia, 11% asthma, 5% malignancy, and 4% immunodeficiency. The 0–2 age groups saw a notable increase in acute pharyngitis, pneumonia, and malignancy at 18%, 9%, and 13%, respectively. The age 0–2 year old group had the highest percent of malignancy. Asthma and immunodeficiency data was not recorded for this age group likely due to reduced diagnosis of these issues at such ages. The most prevalent comorbidity was acute pharyngitis at 78% across all age groups, and no other comorbidity rose above 46%.
 |
Table 6 Comparison of the Top Five Most Common Comorbidities by Age Cohort
|
 |
Figure 1 The five most common comorbidities associated with NTM infection as a percent of each age cohort. *indicates P < 0.05.
|
Discussion
The findings of this study demonstrates an age-based differential presentation of Pulmonary NTM infection among the cohort studied. This study highlights the need for age-specific approaches to the diagnosis and management of pediatric pulmonary NTM infections. Our study adds novel data regarding the global prevalence of NTM pulmonary infections among a Pediatric population. The varying prevalence and outcomes across age groups suggest that different factors, such as immune development and environmental exposure, play significant roles in the susceptibility to and progression of these infections. It is important to understand the mechanisms of these differences. Further studies will be completed to better understand the age based differential of disease prevalence.17 The top five comorbidities we identified were acute pharyngitis, unspecified pneumonia, asthma, malignancy and immunodeficiency. The presence of these comorbidities also correlated with a higher burden of NTM infection in the older pediatric groups, suggesting that these children might have prolonged exposure to risk factors or delayed diagnosis.18 This increased susceptibility is thought to be related to damaged respiratory mucosa, promoting NTM attachment and infection.19 Our observed comorbidity of asthma drew our attention to the fact that children with underlying chronic lung diseases may have a higher likelihood of severe outcomes, including prolonged hospitalizations and the need for intensive care.20 There was a higher odds ratio (OR) of developing more severe lung disease, such as pneumonia, pulmonary fibrosis, lung abscess, bronchiectasis, interstitial lung disease upon NTM infection among 0–2 year olds compared to older age groups (Table 7). This is also supported by several reports of some of these outcomes occurring in various age groups.21,22 Immunodeficiency, whether congenital or acquired, tends to be associated with a more complicated clinical course and higher mortality rates, and we also observed it as a comorbidity.23 Specifically, this immunosuppression allows for extrapulmonary NTM disease to develop and spread by escaping the body’s immune cells and entering the lymphatic system and bloodstream.24 The finding of lung or mediastinal abscess as a five year outcome heightens our clinical concern among this cohort, especially when considering clinical uncertainty in treatment of NTM.25 The analysis of racial and ethnic data was limited in this study due to potential incomplete availability of documentation of some demographics in the electronic health records, which may have introduced bias in the interpretation of these variables. However, our demographic data supports our initial hypothesis that even a mildly elevated BMI may increase susceptibility for pediatric NTM.26 While current literature supports lower BMI as a significant risk factor for NTM infection,27 one possible explanation is the susceptibility of the host in states of malnutrition. Individuals with higher BMI are more likely to be diagnosed with chronic inflammatory or autoimmune conditions28 which are commonly treated using immunosuppressive corticosteroids,29 potentially increasing the risk for NTM infection.30,31 In addition to BMI, other physical conditions and comorbidities, particularly chronic lung diseases and immunodeficiency, underscore the need for vigilant screening and management in at-risk populations. The significant odds ratios associated with future 5-year outcomes in older children (Table 7) suggest that more aggressive diagnostic and therapeutic strategies may be warranted in these age groups.32 Our findings show that the majority of NTM cases were observed in the 6–12 year age group and the highest percent of cases were observed in the 3–5 year age group. This finding shows that school-aged children and adolescents are more likely to be diagnosed with NTM infections, potentially due to increased exposure to environmental reservoirs of NTM, such as water and soil, and higher rates of outdoor activities compared to toddlers.33 Abnormal inflammatory marker values among the 13–18 age group were also a notable age cohort-based finding. Particularly, the elevated white blood cell count, ferritin levels, and CRP in older children suggest a more pronounced inflammatory response in the adolescent age group with NTM infections. These findings are distinct from the data that has been found from a previous study done in Wisconsin. In the Wisconsin study, the isolates were based on statewide data, rather than a global database, and this study included non Pulmonary infections. Our study is the first to our knowledge to analyze a global database, with a focus on Pediatric non CF Pulmonary infections. The data in the Wisconsin study included Cystic Fibrosis patients. Since CF is a known risk factor for Pulmonary NTM, our study design focused on non CF patients.32 From our demographic data, we also found that there were significant differences in the gender of infected patients. This trend aligns with existing adult literature that suggests females may have a higher susceptibility to pulmonary infections due to differences in immune response, hormonal influences, or behavioral factors, such as a greater likelihood of engaging in activities that increase exposure to NTM.34 There were statistically significant differences in NTM infection rates among different racial groups, which has to our knowledge not been studied utilizing a global EHR database. The available data from diverse populations shows that certain racial and ethnic groups might be disproportionately affected by NTM infections, potentially due to disparities in healthcare access, environmental exposures, or genetic predispositions.35 This finding highlights the need for more comprehensive data collection and medical record documentation in future studies to better understand the role of race and ethnicity in the epidemiology of pediatric pulmonary NTM infections.
 |
Table 7 Pulmonary Health Outcomes by Age Cohort
|
Limitations
This study has several limitations, including its retrospective nature and reliance on data from the TriNetX platform, which may introduce selection bias. Additionally, the reliance on electronic health record data may result in incomplete or inaccurate documentation of clinical characteristics and outcomes. The platform’s retrospective nature may introduce selection bias, affecting the generalizability of the results. Additionally, differences in cohort sizes can impact comparison results. TriNetX also anonymizes data by replacing counts of 1–10 with “10” to protect patient privacy, which limits the statistical analysis of rare outcomes. Furthermore, patients who move to non-TriNetX facilities are lost and not included in the outcome comparisons. There may also be comorbidities that were documented in patients’ records in institutions not participating in TriNetX. The study also did not account for the potential impact of socio-economic factors, geographic location, or variations in healthcare access, which could influence the prevalence and outcomes of NTM infections.36,37 Despite these limitations, our study has provided an extensive analysis of global data in this area of study, utilizing a novel EHR based database.38,39 Additionally, more advocacy is needed to ensure that NTM is properly documented, species-wise in the EHR to ensure that future epidemiological studies can be done effectively.
Conclusions
Our study uniquely provides an analysis of global prevalence of non CF Pediatric Pulmonary NTM infections. The study contributes to current knowledge in the field and identifies selected future five year outcomes. We also compared the associated risk for the specific age cohorts studied. This study thus adds to current understanding of the incidence and characteristics of Pediatric NTM non CF pulmonary disease. Future studies to develop treatment strategies and age based considerations are therefore important in the management of Pediatric pulmonary NTM infection.
Disclosure
The authors report no conflicts of interest in this work.
References
1. Daley C, Laccarino J, Lange C, Cambau E. Treatment of nontuberculous mycobacterial pulmonary disease: an official ATS/ERS/ESCMID/IDSA clinical practice guideline. Clin Infect Dis. 2020;71(4:e1–36. doi:10.1093/cid/ciaa241
2. Bai A, Belda O, Dosanjh A. Pulmonary nontuberculous mycobacterial infection in infants: a systematic review. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC68721029/#.
3. Ratnatunga C, Lutzky V, Kupz A, Doolan D, Reid D. The rise of non-tuberculosis mycobacterial lung disease. Front Immunol. 2020. 11. doi:10.3389/fimmu.2020.00303
4. Loebinger M, Quint J, van der Laan R, Chawla R. Risk factors for nontuberculous mycobacterial pulmonary disease. Chest. 2023;164(5):1115–1124. doi:10.1016/j.chest.2023.06.014
5. Honda JR, Virdi R, Chan ED, Sun X, Yarden O, Li S. Global environmental nontuberculous mycobacteria and their contemporaneous man-made and natural niches. Front Microbiol. 2018;9:9. doi:10.3389/fmicb.2018.02029
6. Lamb G, Starke J. Mycobacterium abscessus infections in children: a review of current literature. J Pedia Infect Dis Soc. 2018;7(3):e131–e144. doi:10.1093/jpids/piy047
7. Belardinelli JM, Li W, Avanzi C, Angala SK, Lian E. Unique features of mycobacterium abscessus biofilms formed in synthetic cystic fibrosis medium. Front Microbiol. 2021;12. doi:10.3389/fmicb.2021.743126
8. Sampaio EP, Elloumi HZ, Zelazny A, et al. Mycobacterium abscessus and M. avium trigger toll-like receptor 2 and distinct cytokine response in human cells. Am J Respir Cell Mol Biol. 2008;39(4):431–439. doi:10.1165/rcmb.2007-0413OC
9. Lopez-Varlea E, Garcia-Basteiro A, Santiago B, Wagner D, van Ingen J, Kampmann B. Non-tuberculous mycobacteria in children: muddying the waters of tuberculosis diagnosis. Lancet Resp Med. 2015;3(3):244–256. doi:10.1016/S2213-2600(15)00062-4
10. Bonnet M, Chamroeun San K, Pho Y, Sok C. Nontuberculous mycobacteria infections at a provincial reference hospital, Cambodia. Emerging Infectious Diseases. 2017;23(7):1139–1147. doi:10.3201/eid2307.170060
11. Chindam A, Vengaldas S, Srigiri V, Syed U. Challenges of diagnosing and treating non-tuberculous mycobacterial pulmonary disease [NTM-PD]: a case series. J ClinTubercul Other Mycobacter Dis. 2021;25:100271. doi:10.1016/j.jctube.2021.100271
12. Wu ML, Aziz D, Dartois V, Dick T. NTM drug discovery: status, gaps and the way forward. Drug Discovery Today. 2018;23(8):1502–1519. doi:10.1016/j.drudis.2018.04.001
13. Ryu YJ, Koh WJ, Daley, Daley CL. Diagnosis and treatment of nontuberculous mycobacterial lung disease: clinicians’ perspectives. Tubercul Resp Dis. 2016;79(2):74. doi:10.4046/trd.2016.79.2.74
14. Pennington KM, Vu A, Challener D, et al. Approach to the diagnosis and treatment of non-tuberculous mycobacterial disease. J Clin Tuberc Mycobact Dis. 2021;24:100244. doi:10.1016/j.jctube.2021.100244
15. Raju RM, Raju SM, Zhao Y, Rubin EJ. Leveraging advances in tuberculosis diagnosis and treatment to address nontuberculous mycobacterial disease. Emerging Infectious Diseases Journal – CDC. 2016;22(3). doi:10.3201/eid2203.151643
16. Lyczak J, Cannon C, Pier G. Lung infections associated with cystic fibrosis. Available from: https://pmc.ncbi.nlm.nih.gov/articles/PMC118069/.
17. Adjemian J, Frankland TB, Daida YG, et al. Epidemiology of nontuberculous mycobacterial lung disease and tuberculosis, Hawaii, USA. Emerging Infectious Diseases Journal – CDC. 2017;23(3). doi:10.3201/eid2303.161827
18. Mauch RM, Mansinho AA, Rocha PM, et al. Nontuberculous mycobacterial infections in a Brazilian pediatric population: a seven-year survey. Pathog Glob Health. 2020;114(2):104–108. doi:10.1080/20477724.2020.1725330
19. Bhattacharya J, Mohandas S, Goldman D. Nontuberculous mycobacterial infections in children | pediatrics in review | American academy of pediatrics. Available from: https://publications.aap.org/pediatricsinreview/article-abstract/40/4/179/35278/Nontuberculous-Mycobacterial-Infections-in?redirectedFrom=fulltext.
20. Fritscher L, Marras T, Bradi A. Nontuberculous mycobacterial infection as a cause of difficult-to-control asthma – chest. Available from: https://journal.chestnet.org/article/S0012-3692(11)60010-X/abstract.
21. Park B, Park J. Atypical presentation of nontuberculous mycobacterial pulmonary infection in a patient with interstitial lung abnormality: a case report – European journal of radiology open. Available from: https://www.ejropen.com/article/S2352-0477(21)00033-2/fulltext.
22. Yin H, Gu X, Wang Y, et al. Clinical characteristics of patients with bronchiectasis with nontuberculous mycobacterial disease in Mainland China: a single center cross-sectional study. BMC Infect Dis. 2021;21(1):1216. doi:10.1186/s12879-021-06917-8
23. Ford T, Silcock R, Holland S. Overview of nontuberculous mycobacterial disease in children. J Paedia Child Health. 2025;57(1):15–8. doi:10.1111/jpc.15257
24. Cinicola B, Ottaviano G, Hashim I. Prevalence and characteristics of non-tuberculous mycobacteria (NTM) infection in recipients of allogeneic hematopoietic stem cell transplantation: a systematic review and meta-analysis. Journal of Clinical Immunology. 2025. doi:10.1007/s10875-023-01615-3
25. Kuhajda I, Zarogoulidis K, Tsirgoggiani K. Lung abscess-etiology, diagnostic and treatment options. Available from: https://atm.amegroups.org/article/view/7152/7940.
26. Andersen CJ, Murphy KE, Fernandez ML. Impact of obesity and metabolic syndrome on immunity. Adv Nutr Bethesda Md. 2016;7(1):66–75. doi:10.3945/an.115.010207
27. Kang JY, Han K, Kim. Severity of underweight affects the development of nontuberculous mycobacterial pulmonary disease; a nationwide longitudinal study | scientific Reports. Available from: https://www.nature.com/articles/s41598-022-21511-x.
28. Obesity as a risk and severity factor in rheumatic diseases (autoimmune chronic inflammatory diseases) – PMC. Available from: https://pmc.ncbi.nlm.nih.gov/articles/PMC4227519/.
29. Coutinho A, Chapman. The anti-inflammatory and immunosuppressive effects of glucocorticoids, recent developments and mechanistic insights – PMC. Available from: https://pmc.ncbi.nlm.nih.gov/articles/PMC3047790/.
30. Liu V, Winthrop K, Lu Y. Association between inhaled corticosteroid use and pulmonary nontuberculous mycobacterial infection – PMC. Available from: https://pmc.ncbi.nlm.nih.gov/articles/PMC6321990/.
31. Mohseni M, van der Valk ES, Van der Hurk MJB, Savas M, Boon MR, van Rossum EFC. Corticosteroid use and long-term changes in weight and waist circumference: the lifelines cohort study. J Clin Endocrinol Metab. 2025;dgaf166. doi:10.1210/clinem/dgaf166
32. Vonasek BJ, Gusland D, Tans-Kersten J, Misch EA, Gibbons-Burgener SN. Nontuberculous mycobacterial infection in Wisconsin children and adolescents. J Clin Tuberc Mycobact Dis. 2024;36:100456. doi:10.1016/j.jctube.2024.100456
33. Meoli A, Deolmi M, Iannarella R, Esposito S. Non-tuberculous mycobacterial diseases in children. Pathogens. 2020;9(7):553. doi:10.3390/pathogens9070553
34. Chan ED, Slender IMD. Older women appear to be more susceptible to nontuberculous mycobacterial lung disease. Gend Med. 2010;7(1):5–18. doi:10.1016/j.genm.2010.01.005
35. Henkle E, Hedberg K, Schafer S, Novosad S, Winthrop KL. Population-based incidence of pulmonary nontuberculous mycobacterial disease in oregon 2007 to 2012. Ann Am Thorac Soc. 2015;12(5):642–647. doi:10.1513/AnnalsATS.201412-559OC
36. Zhao Z, Hu H, Wang M. Risk factors and mental health status in patients with non-tuberculous mycobacterial lung disease: a single center retrospective study. Available from: https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2022.912651/full.
37. O’Connell J. Nontuberculous respiratory infections among the homeless. Available from: https://pubmed.ncbi.nlm.nih.gov/1810003/.
38. Ku J, Henkle E, Carlson K. Validity of diagnosis code–based claims to identify pulmonary NTM disease in bronchiectasis patients. Available from: https://wwwnc.cdc.gov/eid/article/27/3/20-3124_article.
39. Mejia-Chew C, Yaeger L, Montes K. Diagnostic accuracy of health care administrative diagnosis codes to identify nontuberculous mycobacteria disease: a systematic review. Available from: https://academic.oup.com/ofid/article/8/5/ofab035/6278496.