Pitfalls of statistical analysis and clinical interpretation of the estimates in patients with chronic kidney disease. Part I: Risk assessment

The risk ratio (RR) and odds ratio (OR) are widespread methods of assessment of risk factor and outcome contingency. OR is an indirect evaluation of RR. It gives the representation polarity of absolute risk changes and its statistical significance, but often gives a distorted view of the risk multiplicity under the influence of the analyzed risk factor (as a rule, OR "overestimates" the RR). In a case of the marginal frequencies of outcomes less than 10% of the assessment of OR is very close to RR value. Therefore, often when conducting meta-analyzes of the risk factors for rare diseases, the OR is taken for OR, in other cases, the RR can be estimated from the OR with accurate data on the disease prevalence or the risk in the unexposed group. Although the RR is more obvious, the OR is most often used to assess the relationship of the risk factor and outcome. There are several explanations for this. OR is universal: unlike RR, OR can be assessed not only in a cohort study, but also in a case-control study. OR is much more convenient because it allows one to get inverse estimates not only concerning the risk factor, but also concerning the outcome. Univariate analysis, which does not take into account the influence of various confounders (factors which are not the direct purpose of the analysis), can lead to a biased assessment (an example is the Simpson paradox). To overcome this problem, multivariate analysis is used. One of the most commonly used types of multivariate analysis is logistic regression. As in the case of cohort studies, and case-control studies the relationship between risk factors and outcomes is assessed using the expression OR. The article presents a method for converting adjusted OR to RR (as well as the limits of the confidence interval) and provides examples. It should be remembered that both RR and OR, with a direct empirical assessment, are cumulative, do not imply censored observations, do not take into account the influence of covariates and the time of observation. If the subjects of the study have different observation times, the relationship between the risk factor and the outcome can be analyzed by calculating the intensity of events. This indicator is known as incidence density and represents the ratio of the number of events to the total time of observation of patients. The ratio of such assessments in the two groups can be interpreted as relative risk. The adjusted risk value for various factors can be obtained using the Poisson regression, which analyzes the influence of predictors on the intensity of the occurrence of events. Thus, despite the seeming simplicity of the OR and RR estimates, each of them has features of interpretation. It is necessary to possess basic knowledge about the OR and RR features to adequately and fully understand the information given in scientific publications and use it in clinical practice.

отношение шансов, отношение рисков, распространенность заболевания, инцидентность, плотность инцидентности, логистическая регрессия, оценка риска, статистика, odds ratio, risk ratio, disease prevalence, incidence, incidence density, logistic regression, risk assessment, statistics

1. Knol M.J., Duijnhoven R.G., Grobbee D.E. et al. Potential misinterpretation of treatment effects due to use of odds ratios and logistic regression in randomized controlled trials. PLoS One. 2011;6(6):e21248.

2. Janani L., Mansournia M.A., Nourijeylani K. et al. Statistical Issues in Estimation of Adjusted Risk Ratio in Prospective Studies. Arch Iran Med. 2015 Oct;18(10):713-9. DOI: 0151810/AIM.0012.

3. Nurminen M. To use or not to use the odds ratio in epidemiologic analyses? Eur J Epidemiol. 1995;11(4):365-71.

4. McNutt L.A., Wu C., Xue X. et al. Estimating the relative risk in cohort studies and clinical trials of common outcomes. Am J Epidemiol. 2003;157(10):940-3.

5. Guyatt G.H., Oxman A.D., Santesso N. et al. Grade guidelines 12. Preparing summary of findings tables?binary outcomes. J Clin Epidemiol 2012; 69: 158-172.

6. Deeks J.J., Higgins J.P.T., Altman D.G. Cochrane handbook for systematic reviews of interventions. Chapter 9: Analysing data and undertaking meta-analyses. The Cochrane Collaboration; 2011.

7. Balasubramanian H., Ananthan A., Rao S. et al. Odds ratio vs risk ratio in randomized controlled trials. Postgrad Med. 2015; 127(4):359-67. DOI: 10.1080/00325481.2015.1022494.

8. Shrier I., Steele R. Understanding the relationship between risks and odds ratios. Clin J Sport Med. 2006; 16(2):107-10.

9. Рубанович А.В., Хромов-Борисов Н.Н. Теоретический анализ показателей предсказательной эффективности бинарных генетических тестов. Экологическая генетика. 2013; 11(1): 77-90.

10. Huang P.P., Shu D.H., Su Z. et al. Association between lifestyle, gender and risk for developing end-stage renal failure in IgA nephropathy: a case-control study within 10 years. Ren Fail. 2019;41(1):914-920. DOI: 10.1080/0886022X.2019.1635029.

11. The portal for rare diseases and orphan drugs. [Internet] https://www.orpha.net/

12. IDF Diabetes Atlas - ed 8 (Database on the Internet) 2017. [Internet] http://www.idf.org/

13. Emerging Risk Factors Collaboration, Sarwar N., Gao P. et al. Diabetes mellitus, fasting blood glucose concentration, and risk of vascular disease: a collaborative meta-analysis of 102 prospective studies. Lancet. 2010; 375(9733):2215-22. DOI: 10.1016/S0140-6736(10)60484-9.

14. Зулькарнаев А.Б., Фоминых Н.М., Карданахишвили З.Б. Сосудистый доступ и выживаемость пациентов на гемодиализе: особенности причинно-следственной связи. Вестник трансплантологии и искусственных органов. 2019;21(2):49-58. DOI: 10.15825/1995-1191-2019-2-49-58.

15. Бикбов Б.Т., Томилина Н.А. Заместительная терапия терминальной хронической почечной недостаточности в Российской Федерации в 1998-2013 гг. Отчет по данным общероссийского регистра заместительной почечной терапии Российского диализного общества. Часть первая. Нефрология и диализ. 2015; 17(3, приложение): 5-111. DOI: 10.28996/1680-4422-2017-4suppl-1-95.

16. European Renal Association - European Dialysis and Transplant Association (ERA-EDTA) Registry Annual Report 2015. 2017; [Internet] https://www.era-edta-reg.org/files/annualreports/pdf/AnnRep2015.pdf

17. United States Renal Data System. 2016 USRDS annual data report. Volume 2 - End-stage Renal Disease (ESRD) in the United States: 1 • Incidence, Prevalence, Patient Characteristics, and Treatment Modalities 2016; [Internet] https://www.usrds.org/2016/view/Default.aspx

18. Zhang J., Yu K.F. What's the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes. JAMA. 1998;280(19):1690-1.

19. Buraczynska M., Zukowski P., Wacinski P. et al. Chemotactic cytokine receptor 5 gene polymorphism: relevance to microvascular complications in type 2 diabetes. Cytokine. 2012 May;58(2):213-7. DOI: 10.1016/j.cyto.2012.01.007.

20. Parving H.H., Lewis J.B., Ravid M. et al. Prevalence and risk factors for microalbuminuria in a referred cohort of type II diabetic patients: a global perspective. Kidney Int. 2006; 69(11):2057-63.

21. Grant R.L. Converting an odds ratio to a range of plausible relative risks for better communication of research findings. BMJ. 2014;348:f7450. DOI: 10.1136/bmj.f7450.

22. Sanghavi K., Brundage R.C., Miller M.B. et al. Genotype-guided tacrolimus dosing in African-American kidney transplant recipients. Pharmacogenomics J. 2017; 17(1):61-68. DOI: 10.1038/tpj.2015.87.

23. Jacobson P.A., Schladt D., Oetting W.S. et al. Lower calcineurin inhibitor doses in older compared to younger kidney transplant recipients yield similar troughs. Am J Transplant. 2012;12(12):3326-36. DOI: 10.1111/j.1600-6143.2012.04232.x.

24. Peeters L.E.J., Andrews L.M., Hesselink D.A. et al. Personalized immunosuppression in elderly renal transplant recipients. Pharmacol Res. 2018;130:303-307. DOI: 10.1016/j.phrs.2018.02.031.

25. Kievit R.A., Frankenhuis W.E., Waldorp L.J. et al. Simpson's paradox in psychological science: a practical guide. Front Psychol. 2013;4:513. DOI: 10.3389/fpsyg.2013.00513.

26. Hanley J.A., Th?riault G. Simpson's paradox in meta-analysis. Epidemiology. 2000;11(5):613-4.

27. Altman D.G., Deeks J.J. Meta-analysis, Simpson's paradox, and the number needed to treat. BMC Med Res Methodol. 2002;2:3.

28. Gofrit O.N., Zorn K.C., Steinberg G.D. et al. The Will Rogers phenomenon in urological oncology. J Urol. 2008;179(1):28-33.

29. Albertsen P.C., Hanley J.A., Barrows G.H. et al. Prostate cancer and the Will Rogers phenomenon. J Natl Cancer Inst. 2005;97(17):1248-53.

30. Wacholder S. Binomial regression in GLIM: estimating risk ratios and risk differences. Am J Epidemiol. 1986;123(1):174-84.

31. Sangmuang P., Lucksiri A., Katip W. Factors Associated with Mortality in Immunocompetent Patients with Hospital-acquired Pneumonia. J Glob Infect Dis. 2019;11(1):13-18. DOI: 10.4103/jgid.jgid_33_18.

32. Park S.H., Jeong M.H., Park I.H. et al. Effects of combination therapy of statin and N-acetylcysteine for the prevention of contrast-induced nephropathy in patients with ST-segment elevation myocardial infarction undergoing primary percutaneous coronary intervention. Int J Cardiol. 2016;212:100-6. DOI: 10.1016/j.ijcard.2016.03.009.

33. Shah P.K., Yan P.L., Dauw C.A. et al. Emergency Department Switching and Duplicate Computed Tomography Scans in Patients With Kidney Stones. Urology. 2018;114:41-44. DOI: 10.1016/j.urology.2018.01.013.

34. Muduma G., Odeyemi I., Smith-Palmer J. et al. Budget impact of switching from an immediate-release to a prolonged-release formulation of tacrolimus in renal transplant recipients in the UK based on differences in adherence. Patient Prefer Adherence. 2014;8:391-9. DOI: 10.2147/PPA.S60213.

35. Muduma G., Shaw J., Hart W.M. et al. Cost utility analysis of immunosuppressive regimens in adult renal transplant recipients in England and Wales. Patient Prefer Adherence. 2014;8:1537-46. DOI: 10.2147/PPA.S69461.

36. Yang Y., George K.C., Luo R. et al. Contrast-induced acute kidney injury and adverse clinical outcomes risk in acute coronary syndrome patients undergoing percutaneous coronary intervention: a meta-analysis. BMC Nephrol. 2018;19(1):374. DOI: 10.1186/s12882-018-1161-5.

37. Shang W., Li L., Ren Y. et al. History of kidney stones and risk of chronic kidney disease: a meta-analysis. PeerJ. 2017;5:e2907. DOI: 10.7717/peerj.2907.

38. Crowley M.J., Diamantidis C.J., McDuffie J.R. et al. Clinical Outcomes of Metformin Use in Populations With Chronic Kidney Disease, Congestive Heart Failure, or Chronic Liver Disease: A Systematic Review. Ann Intern Med. 2017;166(3):191-200. DOI: 10.7326/M16-1901.

39. Томилина Н.А., Андрусев А.М., Перегудова Н.Г. и соавт. Заместительная терапия терминальной хронической почечной недостаточности в Российской Федерации В 2010-2015 гг. Отчет по данным общероссийского регистра заместительной почечной терапии Российского диализного общества, часть первая. Нефрология и диализ. 2017; 19(4, приложение):1-95. DOI: 10.28996/1680-4422-2017-4suppl-1-95.