Gabriel Macharie: Healthcare professionals depend primarily on evidence-based healthcare to inform clinical decision-making. A research hypothesis is frequently tested, and findings are presented with p values, confidence intervals, or both. Furthermore, the scientists estimate or determine statistical or scientific significance. Unfortunately, healthcare practitioners’ comfort levels in interpreting these findings may vary, impairing the data’s appropriate use (Rumil, 2021). Both confidence intervals and hypothesis tests are inferential procedures that depend on an estimated sample pattern. Confidence intervals determine a population parameter using data from a sample. To test a hypothesis, hypothesis tests employ data from a sample. A confidence interval is the mean of the estimation plus and minus its variance. Inside a specific degree of confidence, this is the value spectrum within which a researcher anticipates their estimate to settle if they repeat the test. Confidence denotes probability. Hypothesis testing is the principle of evaluating the quality of evidence gathered from a specimen and providing a foundation for creating population-related conclusions. It offers a technique for acknowledging how dependably identified research results in a sample can be extrapolated to the larger population from which the sample was drawn. The investigator develops a particular hypothesis, examines data from the sample, and utilizes the data to determine if the specific hypothesis is supported (Rumil, 2021). For instance, my local organization’s patients who were given drug 23 had fewer symptoms than those who were given Drug 22, and the difference was statistically significant: If the bar had been placed at 0.05, the null hypothesis would have been invalidated, and we would have concluded that there were significant differences. Notably, some studies will publish findings with a p-value of less than 0.000001, while others will offer an exact p-value (0.000001) but never zero. Readers should understand how p values are provided while reviewing studies. It is recommended practice to provide all p values for all variables within a research design rather than only those with significant findings. Including all p values offer evidence for research validity and reduce the possibility of selective reporting or data mining. 95% of CIs cannot account for mistakes made by research, such as improper data analysis). Researchers should look for journal preferences when deciding whether to provide p-values or CIs. When in doubt, it can be helpful to report both with such an illustration as After three days, participants who were administered Drug 23 had no symptoms, much quicker than those prescribed Drug 22 (p = 0.009). The mean difference in recovery days between the two groups was 4.2 days (95% CI: 1.9 – 7.8). References Fleischmann, M., & Vaughan, B. (2019). Commentary: Statistical significance and clinical significance – A call to consider patient-reported outcome measures, effect size, confidence interval, and minimal clinically important difference (MCID). Journal of bodywork and movement therapies, 23(4), 690–694. https://doi.org/10.1016/j.jbmt.2019.02.009 Rumil Legaspi (2021). The Relationship between Hypothesis Testing and Confidence intervals. https://towardsdatascience.com/the-relationship-between-hypothesis-testing-and-confidence-intervals-43196f1b44bf
