EVALUATION OF STATISTICAL METHODS APPLIED IN OBSERVATIONAL STUDIES IN NEUROLOGY
Abstract
Introduction/Objective: Statistical methods play a key role in the planning, analysis, and interpretation of results in medical research. Due to the complexity of neurological diseases and the heterogeneity of study populations, the proper application of statistical methods in neurology is particularly important for drawing reliable clinical conclusions. This study aims to evaluate the application of statistical methods in observational studies in neurology, with special emphasis on the frequency of the use of different statistical techniques, the adequacy of their application, and the interpretation of results.
Materials and methods: A literature search of the MEDLINE/PubMed database was conducted for all publications available up to December 1, 2024, using the term “neurology” with the application of the filter “observational study.” After applying the inclusion and exclusion criteria, 95 observational studies published between 2013 and 2024 were randomly selected for analysis. Two researchers independently extracted the data using a previously designed data collection form.
Results: A retrospective design was identified in 53.7% of studies, while 46.3% were prospective. Ethical approval was reported in 78.9% of studies. Assessment of the normality of distribution was presented in 29.5% of studies. The most commonly used statistical tests were the chi-square test (46.3%), T-test (36.8%), Fisher’s exact test (36.8%), and the Mann–Whitney test (33.7%). Regression analysis was applied in 33.7% of studies. Sample size planning was clearly reported in only 10.5% of studies. The application of the T-test was statistically significantly more frequent in studies that reported a previous assessment of the normality of distribution (p < 0.001).
Conclusion: The results indicate that statistical methods in observational neurological studies are generally applied adequately and in accordance with data characteristics. However, greater attention should be devoted to testing statistical assumptions and proper sample size planning in order to further improve the quality and reliability of future research.
References
Altman DG. Statistics in medical journals: developments in the 1980s. Stat Med. 1991 Dec;10(12):1897-913. doi: 10.1002/sim.4780101206.
Dawson B, Trapp RG. Basic & clinical biostatistics. 4th ed. New York: Lange Medical Books-McGraw-Hill, Medical Pub. Division; 2004. 438 p.
Campese N, Fanciulli A, Stefanova N, Haybaeck J, Kiechl S, Wenning GK. Neuropathology of multiple system atrophy: Kurt Jellinger`s legacy. J Neural Transm (Vienna). 2021 Oct;128(10):1481-94. doi: 10.1007/s00702-021-02383-3. Erratum in: J Neural Transm (Vienna). 2021 Oct;128(10):1495. doi: 10.1007/s00702-021-02412-1.
Jellinger KA, Lantos PL. Papp-Lantos inclusions and the pathogenesis of multiple system atrophy: an update. Acta Neuropathol. 2010 Jun;119(6):657-67. doi: 10.1007/s00401-010-0672-3.
Hobart J, Cano S. Improving the evaluation of therapeutic interventions in multiple sclerosis: the role of new psychometric methods. Health Technol Assess. 2009 Feb;13(12):iii, ix-x, 1-177. doi: 10.3310/hta13120.
Giussani G, Westenberg E, Garcia-Azorin D, Bianchi E, Yusof Khan AHK, Allegri RF, et al. Global COVID-19 Neuro Research Coalition. Prevalence and Trajectories of Post-COVID-19 Neurological Manifestations: A Systematic Review and Meta-Analysis. Neuroepidemiology. 2024;58(2):120-33. doi: 10.1159/000536352.
Biggin F, Emsley HCA, Knight J. Routinely collected patient data in neurology research: a systematic mapping review. BMC Neurol. 2020 Nov 27;20(1):431. doi: 10.1186/s12883-020-01993-w.
Ferrarello F, Viligiardi M, Di Bari M. Ethics reporting practices in randomized controlled trials of physical therapy interventions after stroke. Arch Physiother. 2018 Jun 5;8:8. doi: 10.1186/s40945-018-0049-9.
Ng LC, Hanlon C, Yimer G, Henderson DC, Fekadu A. Ethics in global health research: the need for balance. Lancet Glob Health. 2015 Sep;3(9):e516-7. doi: 10.1016/S2214-109X(15)00095-9.
Miranda JS, Abbade LPF, Abbade JF, Thabane L, Mbuagbaw L, Pascon GC, et al. Deficiencies in reporting inclusion/exclusion criteria and characteristics of patients in randomized controlled trials of therapeutic interventions in pressure injuries: a systematic methodological review. Int Wound J. 2023 Oct 30;21(2):e14351. doi: 10.1111/iwj.14351.
von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Bull World Health Organ. 2007 Nov;85(11):867-72. doi: 10.2471/blt.07.045120.
Button KS, Ioannidis JP, Mokrysz C, Nosek BA, Flint J, Robinson ES, et al. Power failure: why small sample size undermines the reliability of neuroscience. Nat Rev Neurosci. 2013 May;14(5):365-76. doi: 10.1038/nrn3475. Erratum in: Nat Rev Neurosci. 2013 Jun;14(6):451.
Lee PH, Tse ACY. The quality of the reported sample size calculations in randomized controlled trials indexed in PubMed. Eur J Intern Med. 2017 May;40:16-21. doi: 10.1016/j.ejim.2016.10.008.
Kilkenny C, Parsons N, Kadyszewski E, Festing MF, Cuthill IC, Fry D, et al. Survey of the quality of experimental design, statistical analysis and reporting of research using animals. PLoS One. 2009 Nov 30;4(11):e7824. doi: 10.1371/journal.pone.0007824.
Blakley BW, Janzen B. Statistical Considerations in Otolaryngology Journals. Otolaryngol Head Neck Surg. 2017 Nov;157(5):745-7. doi: 10.1177/0194599817704395.
Hoekstra R, Kiers HA, Johnson A. Are assumptions of well-known statistical techniques checked, and why (not)? Front Psychol. 2012 May 14;3:137. doi: 10.3389/fpsyg.2012.00137.
Parsons NR, Price CL, Hiskens R, Achten J, Costa ML. An evaluation of the quality of statistical design and analysis of published medical research: results from a systematic survey of general orthopaedic journals. BMC Med Res Methodol. 2012 Apr 25;12:60. doi: 10.1186/1471-2288-12-60.
Ghasemi A, Zahediasl S. Normality tests for statistical analysis: a guide for non-statisticians. Int J Endocrinol Metab. 2012 Spring;10(2):486-9. doi: 10.5812/ijem.3505.
To MS, Jukes A. Reporting trends of p values in the neurosurgical literature. J Neurosurg. 2019 Feb 8;132(2):662-70. doi: 10.3171/2018.8.JNS172897.
Ginsel B, Aggarwal A, Xuan W, Harris I. The distribution of probability values in medical abstracts: an observational study. BMC Res Notes. 2015 Nov 26;8:721. doi: 10.1186/s13104-015-1691-x.
Kumar A, Kishun J, Singh U, Gaur D, Mishra P, Pandey CM. Use of appropriate statistical tools in biomedical research: Current trend & status. Indian J Med Res. 2023 Apr;157(4):353-7. doi: 10.4103/ijmr.IJMR_809_20.
Jin Z, Yu D, Zhang L, Meng H, Lu J, Gao Q, et al. A retrospective survey of research design and statistical analyses in selected Chinese medical journals in 1998 and 2008. PLoS One. 2010 May 25;5(5):e10822. doi: 10.1371/journal.pone.0010822.
Campbell I. Chi-squared and Fisher-Irwin tests of two-by-two tables with small sample recommendations. Stat Med. 2007 Aug 30;26(19):3661-75. doi: 10.1002/sim.2832.
Oliveira NL, Pereira CAB, Diniz MA, Polpo A. A discussion on significance indices for contingency tables under small sample sizes. PLoS One. 2018 Aug 2;13(8):e0199102. doi: 10.1371/journal.pone.0199102.
Serra N. A significant minimization of Pearson’s chi square statistics in 2x2 contingency tables: preliminary results for small samples. Epidemiol Biostat Public Health. 2018;15(3):e12949. doi: 10.2427/12949.
