Currently, in the health area, several experimental studies (investigating an intervention versus placebo/control) attempt to demonstrate the relevance of outcomes through statistically significant results¹1 Krause MS. Statistical significance testing and clinical trials. Psychotherapy (Chic). 2011;48(3):217-22. https://doi.org/10.1037/a0022088
https://doi.org/10.1037/a0022088... . However, statistical significance (i.e., p<0.05) does not indicate clinical relevance (Figure 1)²2 Andrei AC. Statistical significance: is there a way out of it? J Thorac Cardiovasc Surg. 2021;161(4):1377-8. https://doi.org/10.1016/j.jtcvs.2020.04.138
https://doi.org/10.1016/j.jtcvs.2020.04.... ,³3 Vishnu VY, Vinny PW. Statistical significance and clinical importance. Neurol India. 2021;69(5):1509. https://doi.org/10.4103/0028-3886.329535
https://doi.org/10.4103/0028-3886.329535... . In fact, it is possible to find a statistically significant result with no clinical relevance, just as it is possible to find a statistically significant result with clinical relevance⁴4 Sharma H. Statistical significance or clinical significance? A researcher’s dilemma for appropriate interpretation of research results. Saudi J Anaesth. 2021;15(4):431-4. https://doi.org/10.4103/sja.sja_158_21
https://doi.org/10.4103/sja.sja_158_21... .

Figure 1
Statistical significance (i.e., p≤0.05) does not show clinical relevance.

A challenge in longitudinal studies (e.g., clinical trials) is the difficulty in translating numbers (outcome) into something applicable to the clinical context (real world) because the p-value (<0.05 or >0.05) only indicates statistical significance⁵5 Amrhein V, Greenland S, McShane B. Scientists rise up against statistical significance. Nature. 2019;567(7748):305-7. https://doi.org/10.1038/d41586-019-00857-9
https://doi.org/10.1038/d41586-019-00857... ,⁶6 Chander NG. Beyond statistical significance. J Indian Prosthodont Soc. 2019;19(3):201-2. https://doi.org/10.4103/jips.jips_207_19
https://doi.org/10.4103/jips.jips_207_19... , in which interpretation only translates a hypothesis test governed by a previously defined probability of error alpha (H0 versus H1)⁷7 Andrade C. The P value and statistical significance: misunderstandings, explanations, challenges, and alternatives. Indian J Psychol Med. 2019;41(3):210-5. https://doi.org/10.4103/IJPSYM.IJPSYM_193_19
https://doi.org/10.4103/IJPSYM.IJPSYM_19... . The language of health is biostatistics⁸8 Villeneuve PJ, Paradis G, Muhajarine N. Always better together: the Canadian journal of public health and the Canadian society for epidemiology and biostatistics. Can J Public Health. 2020;111(3):305-7. https://doi.org/10.17269/s41997-020-00362-x
https://doi.org/10.17269/s41997-020-0036... ,⁹9 Shim S, Yoon BH, Shin IS, Bae JM. Network meta-analysis: application and practice using stata. Epidemiol Health. 2017;39:e2017047. https://doi.org/10.4178/epih.e2017047
https://doi.org/10.4178/epih.e2017047... , but patients are not numbers¹⁰10 Pontes-Silva A. Statistical significance does not show clinical relevance: we need to go beyond the P-value. J Clin Exp Hepatol. 2022;12(5):1402. https://doi.org/10.1016/j.jceh.2022.04.017
https://doi.org/10.1016/j.jceh.2022.04.0... . Therefore, numerical conclusions should be translated into applicability to routine clinical care¹¹11 McClellan J. Clinical relevance versus statistical significance. J Am Acad Child Adolesc Psychiatry. 2017;56(12):1008-9. https://doi.org/10.1016/j.jaac.2017.09.431
https://doi.org/10.1016/j.jaac.2017.09.4... . As such, science should be combined with clinical context so that patients could receive optimal treatment¹²12 Greenhalgh T, Howick J, Maskrey N, Evidence Based Medicine Renaissance Group. Evidence based medicine: a movement in crisis? BMJ. 2014;348:g3725. https://doi.org/10.1136/bmj.g3725
https://doi.org/10.1136/bmj.g3725... . How to solve it? It could be done simply by evaluating the clinical relevance of the results¹⁰10 Pontes-Silva A. Statistical significance does not show clinical relevance: we need to go beyond the P-value. J Clin Exp Hepatol. 2022;12(5):1402. https://doi.org/10.1016/j.jceh.2022.04.017
https://doi.org/10.1016/j.jceh.2022.04.0... .

One way to verify the clinical relevance of results is through health economic evaluations¹³13 Jiang S, Wang Y, Si L, Zang X, Gu YY, Jiang Y, et al. Incorporating productivity loss in health economic evaluations: a review of guidelines and practices worldwide for research agenda in China. BMJ Glob Health. 2022;7(8):e009777. https://doi.org/10.1136/bmjgh-2022-009777
https://doi.org/10.1136/bmjgh-2022-00977... , effect size assessments¹⁴14 Pandis N. The effect size. Am J Orthod Dentofacial Orthop. 2012;142(5):739-40. https://doi.org/10.1016/j.ajodo.2012.06.011
https://doi.org/10.1016/j.ajodo.2012.06.... , or estimates of minimal clinically important differences and minimal detectable change¹⁵15 Terwee CB. Estimating minimal clinically important differences and minimal detectable change. J Hand Surg Am. 2019;44(12):e1. https://doi.org/10.1016/j.jhsa.2019.10.001
https://doi.org/10.1016/j.jhsa.2019.10.0... . I suggest that new studies describe Cohen’s effect size¹⁴14 Pandis N. The effect size. Am J Orthod Dentofacial Orthop. 2012;142(5):739-40. https://doi.org/10.1016/j.ajodo.2012.06.011
https://doi.org/10.1016/j.ajodo.2012.06.... ,¹⁶16 Kelley K, Preacher KJ. On effect size. Psychol Methods. 2012;17(2):137-52. https://doi.org/10.1037/a0028086
https://doi.org/10.1037/a0028086... (e.g., d-value or w-value). Cohen’s d can be used to assess effect sizes when comparing two means (0.2=small effect, 0.5=moderate effect, and 0.8=large effect)¹⁷17 Cohen J. Statistical power analysis for the behavioral sciences. Cambridge, MA: Academic Press; 2013. and the Cohen’s w can be used to assess effect sizes using a chi-squared test (0.1=small effect, 0.3=moderate effect, and 0.5=large effect)¹⁸18 Cohen J. Quantitative methods in psychology: a power primer. Psychol Bull. 1992;112:1155-9. https://doi.org/10.1037//0033-2909.112.1.155
https://doi.org/10.1037//0033-2909.112.1... .

ACKNOWLEDGMENTS

I would like to thank the Coordination for the Improvement of Higher Education Personnel (CAPES), National Council for Scientific and Technological Development (CNPq), São Paulo Research Foundation (FAPESP), Federal University of Maranhão (UFMA), Federal University of São Carlos (UFSCar), and Maria de Fátima Pontes-Silva.

REFERENCES

Publication Dates

History