The Relation of Multiple Sclerosis to Family History, Lifestyle, and Health Factors in Childhood and Adolescence

Findings of a case–control study nested within the German National Cohort (NAKO) study

Background: Multiple sclerosis (MS) is a neuroinflammatory disease of presumed autoimmune origin. A combination of genetic susceptibility and exposure to certain environmental and lifestyle factors might trigger the onset of MS. The currently known risk factors include a genetic predisposition, infection with the Epstein–Barr virus (EBV), smoking, and an increased body mass index.

Methods: In 2021–22, we carried out a case–control study nested within the German National Cohort (NAKO) to investigate associations of potential risk factors with MS.

Results: The subjects included 576 persons with MS (cases) and 895 without MS (controls). Beyond the known risk factors, we observed associations between MS and the cumulative number of common childhood infections (odds ratio (OR) 1.14 per additional infection, 95% confidence interval (CI): [1.03; 1.25]), major stressful life events (SLE) (OR 1.25 per additional event, [1.06; 1.48]), being the firstborn child of a mother aged 30 or older (OR 2.11, [1.08; 4.13]); higher amounts of physical activity in the teenage years were associated with a lower risk of MS (OR 0.82 per unit increase in activity level, [0.71; 0.95]).

Conclusion: We confirmed known risk factors for MS and found associations with a number of new ones, e.g., the cumulative number of common childhood infections. These findings may shed light on the etiology of MS and merit further study.

Cite this as: Holz A, Obi N, Pischon T, Schulze MB, Ahrens W, Berger K, Bohn B, Brenner H, Emmel C, Fischer B, Greiser KH, Harth V, Holleczek B, Kaaks R, Karch A, Katzke V, Keil T, Krist L, Leitzmann M, Meinke-Franze C, Michels KB, Nimptsch K, Peters A, Riedel O, Schikowski T, Schipf S, Schmidt B, Thierry S, Hellwig K, Riemann-Lorenz K, Heesen C, Becher H: The relation of multiple sclerosis to family history, lifestyle, and health factors in childhood and adolescence: Findings of a case–control study nested within the German National Cohort (NAKO) study. Dtsch Arztebl Int 2025; 122: 348–54. DOI: 10.3238/arztebl.m2025.0069

Multiple sclerosis (MS) is a neuroinflammatory, presumably autoimmune disease causing disability in early adulthood (1), with a considerable impact on the quality of life of those affected (2).

In recent decades, the prevalence and incidence of MS have increased worldwide (1, 3), while its etiology is still only partly known (4, 5, 6). Presumably, exposure to environmental and lifestyle factors in genetically susceptible individuals leads to the manifestation of the disease (7, 8).

The odds ratios (OR) of the previously identified MS-associated risk factors body mass index (BMI), vitamin D deficiency (9), and Epstein–Barr virus (EBV) infection range between 1.14 (BMI) (9) and 3.33 (EBV) (10), indicating small to moderate effect sizes. Moderate to vigorous physical activity (PA) during adulthood has been identified as a protective factor (9). Although the majority of observational studies have shown an association between smoking and an elevated risk of MS (11), a causal relationship has not yet been confirmed (9).

Factors acting prenatally or in (early) childhood/adolescence—i.e., at a time when the immune system is still very susceptible—may be of particular importance in the development of MS (12). These include, for example:

• Number of older siblings
• Maternal age
• Childhood infections
• Passive smoking
• Physical activity (PA)
• Time spent outdoors

A potential association with such factors has already been shown for other autoimmune diseases such as bronchial asthma (13) and type 1 diabetes (T1D) (14). Apart from molecular and environmental factors, psychological factors are increasingly being discussed as potential risk factors for MS. A recently published meta-analysis showed a weak to moderate effect of psychological stressors on the risk of MS (15).

Owing to the limited number of studies on some of these modifiable risk factors for MS, the inconsistent study results, and/or marked methodological heterogeneity, further research on these factors is needed.

To contribute to the understanding of the causes of MS, we conducted a case–control study. Our aim was to identify associations between potential risk factors and the development of MS. The following potential risk factors were investigated:

• Family history of MS
• Infectious diseases in childhood
• Passive smoking
• Number of older siblings
• Maternal age at birth
• Time spent outdoors during childhood and adolescence
• PA during adolescence
• Stressful life events (SLE)
• BMI at the age of 18 years
• Smoking.

Methods

The analyses are based on data from the baseline survey of the German National Cohort (NAKO) and the StERKE study, a case–control study on the effect of risk factors on the course and onset of MS that is nested within NAKO (16). Cases were defined as persons with a self-reported physician-based MS diagnosis. The controls, randomly selected NAKO participants without MS, were individually matched to the cases (matching ratio 2:1) by birth year, sex, and study center. In the NAKO baseline survey, data was collected by means of an interview, while the StERKE participants completed a questionnaire. In the analyses, participants were considered exposed only if the exposure had occurred before MS diagnosis (for cases) or before the age of the matched case at MS diagnosis (for controls). A conditional logistic regression was performed to assess the association between observed exposure variables and MS onset. Results are displayed as odds ratios (OR) with 95% confidence intervals.

All statistical analyses were performed with R version 4.3.1 (2023–06–16) (17). The methods used are described in detail in the eMethods and in eSupplement Tables 1–5.

Results

In total, 576 persons with MS (396 women, 180 men) and 895 controls (638 women, 257 men) participated, corresponding to a response rate of 77.2% for cases and 60.0% for controls.

The study population comprised 70% women and 30% men with a mean age of 50 years. The main characteristics of the StERKE participants and the distribution of exposure variables and covariates by sex and case–control status are shown in Table 1 and eTable 1.

Table 1

Characteristics of StERKE participants by sex and case–control status, Germany, 2021–2022

eTable 1

Characteristics of StERKE participants by sex and case–control status, Germany, 2021–2022

eTable 2

Calculation of physical activity level based on the frequency of light and vigorous physical activity during teenage years (13–19 years)

The median age at MS manifestation was 33 and 35 years for women and men, respectively. The median age at diagnosis was 38 years for both sexes. Relapsing–remitting MS (RRMS) was diagnosed in 80% of cases, and 58% had ever received immunotherapy. Among the StERKE participants, 88 (15.3%) were incident cases (Table 2).

Table 2

Characteristics of StERKE participants with multiple sclerosis by sex, Germany, 2021–2022

Figures 1 and 2 show the distribution of the cumulative number of childhood infections and SLE before MS diagnosis, respectively.

Figure 1

Distribution of cumulative number of childhood infections

Figure 2

Distribution of the cumulative numer of stressful life events

Table 3 summarizes the results of the main analysis. Having a first- or second-degree relative with MS showed an OR of 7.08 [3.90; 12.86] compared with no family member with MS. Maternal age at birth of the participant was associated with MS (OR 1.03 per year [1.00; 1.05]), whereas an inverse association with MS was observed for the number of older siblings (OR 0.85 [0.77; 0.95]). The combination of both factors, having no older siblings and being born to a mother ≥ 30 years of age at delivery, was associated with a higher likelihood of MS (OR 2.11 [1.08; 4.13]). We observed a direct association between the cumulative number of childhood infections (OR 1.14 per additional infection [1.03; 1.25]) and having MS. Persons with the maximum of five such infections had an estimated OR of 1.93 (= 1.14⁵) [1.18; 3.06]) compared with those without any reported infection. Persons who had contracted an EBV infection had a 3.05-fold [1.80; 5.16] likelihood of having MS compared with persons who did not. Further, the higher the PA level during teenage years, the lower the likelihood of having MS (OR 0.82 [0.71; 0.95]). Compared with normal weight, overweight and obesity at the age of 18 years were associated with having MS (OR 1.73 [1.22; 2.44] and OR 2.29 [1.18; 4.46], respectively). Moreover, the cumulative number of SLE (OR 1.25 per additional event [1.06; 1.48]) was associated with MS. Finally, we observed a weak association between smoking prior to the diagnosis of MS and MS (OR 1.19 [0.99; 1.43]).

Table 3

Conditional logistic regression on the association of family history and factors occurring in the prenatal period, childhood, adolescence, and adulthood with multiple sclerosis, Germany, 2021–2022

The remaining factors—own serious illness (other than MS) before diagnosis; passive smoking, i.e., parental smoking during pregnancy and in the participant’s childhood/adolescence; and time spent outdoors—were not associated with MS. The estimates of the subgroup analysis and the sex-stratified analysis differed only marginally from the results of the final model (eSupplement Tables 3, 4 and 5).

Discussion

This analysis was based on the StERKE study, a case–control study nested within the population-based cohort study NAKO. Our study yielded several novel findings. We found that the cumulative number of childhood infections and SLE are both associated with MS risk. Higher maternal age at first childbirth is known to be associated with various negative consequences for the child (18), and in this study we found an association with MS risk. Recent studies have shown that PA in adulthood is a protective factor against MS (9). The results of our analyses extend this finding to PA during adolescence. Moreover, we confirmed already known risk factors for MS, including a family history of MS (7), an EBV infection (10, 19), and an elevated BMI in childhood/adolescence (9).

Regarding higher maternal age at first childbirth, a smaller study did not observe an association with having MS (20). However, our finding is in line with studies investigating other autoimmune diseases, e.g., T1D (21), although the underlying pathomechanisms need to be further elucidated.

Previous studies yielded conflicting results regarding the cumulative number of childhood infections and the age at which the infectious diseases occurred (22, 23). We observed an association between the cumulative number of childhood infections and the risk of MS, but this was not the case for the individual diseases. The majority of StERKE participants were born and raised before the introduction of standard vaccinations against childhood infections (24) and were accordingly exposed to the risks associated with contracting an infection. Given the increasing prevalence of MS (1) despite vaccination efforts, future studies including younger, predominantly vaccinated persons or even birth cohorts might help to clarify the relevance of this finding. Our results regarding childhood infections contradict the hygiene hypothesis (25). However, the effect of childhood infections may depend on the age at infection or the type of pathogen. With regard to MS, conclusive evidence for the hygiene hypothesis is lacking to date (26).

We observed an association between the cumulative number of SLE and an increased likelihood of MS. Stress is a common risk factor for chronic, non-communicable diseases (NCD) and has been linked with the development of both mental and somatic illnesses (27). SLE have been shown to have a direct effect on the immune system (28). The specific immune-modulating mechanisms underlying this association remain to be explored.

Our results on PA during adolescence corroborate the results of a large multiregional study reporting an inverse association between PA in adolescence and MS risk (29). PA contributes to the regulation of body weight and has an anti-inflammatory effect, which may be relevant in the development of chronic inflammatory diseases such as MS (30). This is also reflected in our results on the association between overweight/obesity at the age of 18 years and MS, confirming elevated BMI in youth as a risk factor for MS (31).

Exposure to tobacco smoke leads to irritation of the lungs with increased pro-inflammatory cell activation and alteration of proteins up to the point of increased autoantigenic activation, which in turn may be associated with MS risk (11, 32). In our study we found a weak association between smoking and MS. While the majority of observational studies suggest an effect of smoking on MS risk (11), a recently published systematic review summarizing mendelian randomization studies found no causal effect (9).

Our study has several strengths. We were able to investigate a large study sample and use high-quality data acquired in the course of extensive data collection in the NAKO baseline survey. Moreover, the stringent recruitment management in the StERKE study, with up to three reminders per participant, led to a high proportion of responses (77% for cases, 60% for controls).

Nevertheless, our study also has some limitations. The majority of cases were prevalent cases, which is commonly regarded as a potential source of survival bias (33). To account for this, we investigated the estimates of our final model by first including only incident cases and their controls in our analyses, and then adding successive subgroups of sets with increasing intervals between diagnosis and recruitment. We generally found no trend in the estimates (eSupplement Table 3). Although this is not a formal proof, we nevertheless believe our results are sufficiently valid. In addition, the survival of persons with MS after diagnosis is high, in contrast to, for example, some kinds of cancer.

Our analyses are based on self-reported physician-based MS diagnoses, which may have led to possible misclassification. However, we assume that a self-report is reliable in view of the psychological impact of such a diagnosis. For a plausibility check, we conducted a subgroup analysis restricted to the participants with RRMS who reported treatment with MS-specific medication or immunotherapy. The results were very similar to the findings in the complete set of cases (eSupplement Table 4). While we acknowledge the limitation of the self-reported diagnosis of MS, we consider additional statements on MS type or specific medication as a clear indication of a valid diagnosis. Furthermore, persons with chronic diseases might have a greater interest in participating in a study that aims to investigate potential causes of diseases. Considering the frequent multimorbidity of persons with MS, there may have been overrepresentation of persons with MS in the NAKO.

A further weakness of our study is the investigation of the association between an exposure to EBV and MS by asking the participants whether they had ever had infectious mononucleosis (34). EBV infection in childhood is most often asymptomatic, so there may have been under-reporting of the frequency of EBV infection. This might explain the relatively low OR of 3.05 compared with other studies (10, 19).

Since most factors investigated in this study concern events that occurred on average 40–50 years ago, recall bias may have differed between cases and controls, in that persons with MS explain their disease by reference to potential risk factors and thus remember past events more clearly. This may apply to smoking, the amount of time spent outdoors, and particularly infectious diseases in childhood. However, we assume, given the clear nature of the factors assessed, e.g., SLE, that participants answered most questions adequately (35). An objective measure of childhood infections would be an antibody test to confirm a past infection. However, even with antibody tests it is not possible to determine the precise time of infection, which could be important for assessing the effect on the maturing immune system.

In the course of our analyses we investigated a variety of potential risk factors for MS that have received little attention to date. We found associations between MS and the following factors: cumulative number of childhood infections, major SLE, higher maternal age at first pregnancy, and low levels of PA during adolescence. Furthermore, we confirmed known risk factors for MS, including a family history of MS, EBV infection, overweight and obesity in childhood/adolescence, and smoking. These results contribute to the evidence base for existing preventive measures for other NCD, such as cardiovascular diseases childhood infections, and suggest that such measures may also be useful in the context of MS. Vaccination programs, smoking cessation programs, and initiatives to encourage PA and healthy eating habits to prevent overweight and obesity might be promising strategies with regard to MS prevention. Furthermore, our results may inform further research, e.g., the investigation of on one hand dose–response relationships between the newly suggested risk factors and the severity of MS severity and on the other hand the MS prodrome, assuming a lag time. The NAKO offers many possibilities to:

• Investigate exclusively incident MS cases
• Conduct sex-specific analyses
• Use secondary data to confirm self-reported MS diagnoses
• Use biosamples, for example to demonstrate the presence of antibodies against infectious diseases.

Remaining authors
Tobias Pischon, Matthias B. Schulze, Wolfgang Ahrens, Klaus Berger, Barbara Bohn, Hermann Brenner, Carina Emmel, Beate Fischer, Karin Halina Greiser, Volker Harth, Bernd Holleczek, Rudolf Kaaks, André Karch, Verena Katzke, Thomas Keil, Lilian Krist, Michael Leitzmann, Claudia Meinke-Franze, Karin B. Michels, Katharina Nimptsch, Annette Peters, Oliver Riedel, Tamara Schikowski, Sabine Schipf, Börge Schmidt, Sigrid Thierry, Kerstin Hellwig, Karin Riemann-Lorenz

Affiliations of the remaining authors
University Medical Center Hamburg-Eppendorf (UKE): Prof. Dr. Volker Harth

Ethics

The Study on Risk Factors for the Occurrence and Progression of Multiple Sclerosis (StERKE) was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the ethics committee of the Hamburg Medical Association (PV7292).

Written informed consent was obtained from all participants in the German National Cohort (NAKO) and the Study on Risk Factors for the Occurrence and Progression of Multiple Sclerosis (StERKE).

Data sharing

The data underlying this article were provided by NAKO e.V. by permission. Data will be shared on request to the corresponding author with permission of NAKO e.V.

Funding

This work was conducted with data from the German National Cohort (NAKO Gesundheitsstudie, NAKO) (www.nako.de) and supported by the German Federal Ministry of Education and Research (BMBF) (grant number 01ER1901A PERGOLA 2). NAKO is funded by the German Federal Ministry of Education and Research (grant numbers 01ER1301A/B/C, 01ER1511D, 01ER1801A/B/C/D) the federal states of Germany, and the Helmholtz Association, together with the participating universities and institutes of the Leibniz Association.

Acknowledgments

We thank all NAKO participants and study staff. We are grateful to Christian Aust, Stefan Janisch, Daniel Kraft, Achim Reineke, Gunthard Stübs, and Robert Wolff for the technical implementation and support in the conduct of the StERKE study.

Conflict of interest statement

CH has received research funding and speaker honoraria from Novartis, Merck, and Roche.

KB is a member of the editorial board of Deutsches Ärzteblatt.

KH has received study support, consulting fees, speaker honoraria, and payment of travel and congress attendance costs from Biogen, Sanofi, Teva, Roche, Novartis, Merck, BMS, Janssen, Hexal, and Almiral.

AK is spokesperson of NAKO’s Infectious Diseases Expert Group and acting spokesperson of the NAKO Use & Access Committee.

TP is a member of the executive committee of NAKO e.V., which coordinates the NAKO study.

The remaining authors declare that no conflict of interest exists.

Manuscript submitted on 17 October 2024, revised version accepted on 4 April 2025.

Corresponding author
Anja Holz, M.Sc.

References

1. Wallin MT, Culpepper WJ, Nichols E, et al.: Global, regional, and national burden of multiple sclerosis 1990–2016: A systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol 2019; 18: 269–85

2. Mitchell AJ, Benito-León J, González JMM, Rivera-Navarro J: Quality of life and its assessment in multiple sclerosis: Integrating physical and psychological components of wellbeing. Lancet Neurol 2005; 4: 556–66

3. Lane J, Ng HS, Poyser C, Lucas RM, Tremlett H: Multiple sclerosis incidence: A systematic review of change over time by geographical region. Mult Scler Relat Disord 2022; 63: 103932

4. Dobson R, Giovannoni G: Multiple sclerosis—a review. Eur J Neurol 2019; 26: 27–40

5. Holz A, Obi N, Ahrens W, et al.: Childhood and adolescence factors and multiple sclerosis: Results from the German National Cohort (NAKO). BMC Neurol 2024; 24: 123

6. Maroufi H, Mortazavi SH, Sahraian MA, Eskandarieh S: Environmental risk factors of multiple sclerosis in the Middle East and North Africa region: A systematic review. Curr J Neurol 2021; 20: 166–84

7. Ramagopalan SV, Dobson R, Meier UC, Giovannoni G: Multiple sclerosis: Risk factors, prodromes, and potential causal pathways. Lancet Neurol 2010; 9: 727–39

8. Olsson T, Barcellos LF, Alfredsson L: Interactions between genetic, lifestyle and environmental risk factors for multiple sclerosis. Nat Rev Neurol 2017; 13: 25–36

9. Fazia T, Baldrighi GN, Nova A, Bernardinelli L: A systematic review of Mendelian randomization studies on multiple sclerosis. Eur J Neurosci 2023; 58: 3172–94

10. Khalesi Z, Tamrchi V, Razizadeh MH, et al.: Association between human herpesviruses and multiple sclerosis: A systematic review and meta-analysis. Microb Pathog 2023; 177: 106031

11. Degelman ML, Herman KM: Smoking and multiple sclerosis: A systematic review and meta-analysis using the Bradford Hill criteria for causation. Mult Scler Relat Disord 2017; 17: 207–16

12. Olin A, Henckel E, Chen Y, et al.: Stereotypic immune system development in newborn children. CELL 2018; 174: 1277–92.e14

13. Grabenhenrich LB, Gough H, Reich A, et al.: Early-life determinants of asthma from birth to age 20 years: A German birth cohort study. J Allergy Clin Immunol 2014; 133: 979–88

14. Hidayat K, Zou SY, Shi BM: The influence of maternal body mass index, maternal diabetes mellitus, and maternal smoking during pregnancy on the risk of childhood-onset type 1 diabetes mellitus in the offspring: Systematic review and meta-analysis of observational studies. Obes Rev 2019; 20: 1106–20

15. Drathen S von, Gold SM, Peper J, et al.: Stress and multiple sclerosis—systematic review and meta-analysis of the association with disease onset, relapse risk and disability progression. Brain Behav Immun 2024; 120: 620–9

16. Peters A, Greiser KH, Göttlicher S, et al.: Framework and baseline examination of the German National Cohort (NAKO). Eur J Epidemiol 2022; 37: 1107–24

17. R Core Team. R: A language and environment for statistical computing. Vienna, Austria. 2023. https://www.R-project.org/ (last accessed on 7 August 2024).

18. Myrskylä M, Fenelon A: Maternal age and offspring adult health: Evidence from the health and retirement study. Demography 2012; 49: 1231–57

19. Bjornevik K, Cortese M, Healy BC, et al.: Longitudinal analysis reveals high prevalence of Epstein-Barr virus associated with multiple sclerosis. Science 2022; 375: 296–301

20. Goldacre A, Pakpoor J, Goldacre M: Maternal and perinatal characteristics of infants who, later in life, developed multiple sclerosis: record-linkage study. Mult Scler Relat Disord 2017; 13: 98–102

21. Haynes A, Bower C, Bulsara MK, Finn J, Jones TW, Davis EA: Perinatal risk factors for childhood Type 1 diabetes in Western Australia—a population-based study (1980–2002). Diabet Med 2007; 24: 564–70

22. Bachmann S, Kesselring J: Multiple sclerosis and infectious childhood diseases. Neuroepidemiology 1998;17: 154–60

23. Bager P, Nielsen NM, Bihrmann K, et al.: Childhood infections and risk of multiple sclerosis. Brain 2004; 127: 2491–7

24. Robert Koch-Institut (RKI): Impfempfehlungen im Überblick. 2018. www.rki.de/DE/Themen/Infektionskrankheiten/Impfen/Staendige-Impfkommission/Empfehlungen-der-STIKO/Archiv/Tab_Einf_Impfempfehlungen_Dtl.pdf?__blob=publicationFile&v=3 (last accessed on 26 September 2023).

25. Bach JF: The hygiene hypothesis in autoimmunity: The role of pathogens and commensals. Nar Rev Immunol 2018; 18: 105–20

26. Wendel-Haga M, Celius EG: Is the hygiene hypothesis relevant for the risk of multiple sclerosis? Acta Neurol Scand 2017; 136 Suppl 201: 26–30

27. Liu YZ, Wang YX, Jiang CL: Inflammation: The common pathway of stress-related diseases. Front Hum Neurosci 2017; 11: 316

28. Dich N, Rozing MP, Kivimäki M, Doan SN: Life events, emotions, and immune function: Evidence from Whitehall II Cohort Study. Behav Med 2020; 46: 153–60

29. Wesnes K, Myhr KM, Riise T, et al.: Physical activity is associated with a decreased multiple sclerosis risk: The EnvIMS study. Mult Scler 2018; 24: 150–7

30. Gleeson M, Bishop NC, Stensel DJ, Lindley MR, Mastana SS, Nimmo MA: The anti-inflammatory effects of exercise: Mechanisms and implications for the prevention and treatment of disease. Nat Rev Immunol 2011; 11: 607–15

31. Liu Z, Zhang TT, Yu J, et al.: Excess body weight during childhood and adolescence is associated with the risk of multiple sclerosis: A meta-analysis. Neuroepidemiology 2016; 47: 103–8

32. Hedström AK: Smoking and its interaction with genetics in MS etiology. Mult Scler 2019; 25: 180–6

33. Mandal S, Qin J, Pfeiffer RM: Incorporating survival data into case-control studies with incident and prevalent cases. Stat Med 2021; 40: 6295–308

34. Thacker EL, Mirzaei F, Ascherio A: Infectious mononucleosis and risk for multiple sclerosis: A meta-analysis. Ann Neurol 2006; 59: 499–503

35. Polick CS, Polick SR, Stoddard SA: Relationships between childhood trauma and multiple sclerosis: A systematic review. J Psychosom Res 2022; 160: 110981

e1. UNESCO Unites Nations Educational, Scientific and Cultural Organization: International standard classification of education, ISCED 1997. In: Hoffmeyer-Zlotnik JHP, Wolf C (eds.): Advances in cross-national comparison. Boston, MA: Springer US 2003: 195–220

e2. Godin G, Shephard RJ: Godin leisure-time exercise questionnaire. Med Sci Sports Exerc 1997; 29: 36–8

e3. Robert Koch-Institut (RKI): DEGS: Studie zur Gesundheit Erwachsener in Deutschland. www.rki.de/DE/Content/Gesundheitsmonitoring/Studien/Degs/degs_node.html (last accessed on 8 August 2019).

e4. Neuhäuser M, Becher H: Improved odds ratio estimation by post hoc stratification of case-control data. Stat Med 1997; 16: 993–1004

e5. Lüdecke D: sjPlot: Data visualization for statistics in social science. 2022.https://CRAN.R-project.org/package=sjPlot (last accessed on 7 August 2025).

e6. Therneau TM: A package for survival analysis in R. 2023. https://CRAN.R-project.org/package=survival (last accessed on 7 August 2025).

e7. Wickham H, Averick M, Bryan J, et al.: Welcome to the Tidyverse. J Open Source Softw 2019; 4: 1686