This concept has been a difficult one for me to grasp but this is my take on the subject.
I currently work on a post orthopedic surgery/trauma unit. On our unit, it is very important that all patients get out of bed at least once on the day of surgery (activity level allowing). According to Zou, et all: It was found that adding a 15-minute walk on the day of surgery did not increase pain in patients, it reduced the length of stay, and reduced revision rates.  It is believed that this one task provides patients with shorter hospital stays and faster recovery times. To determine if this statement could be true, I would ask the question: If all post-surgery patients take one 15-minute walk on the day of surgery will they be more likely to discharge from the hospital within 3 days? Finding the confidence interval of the sample group will ensure that 95% of the time the true population mean will fall within the number range provided for the confidence interval (Chamberlain University, 2019). Using a sample size of 101 post-operative patients the mean length of stay was 5.41 days with a standard deviation of 7.42. I chose to use the mean because I wanted to look at the average length of stay for all the patients in the sample.
Confidence level 95% (3.95, 6.87)
Confidence level 90% (4.18, 6.63)
Confidence level 99% (3.47, 7.35)
I believe the 95% CI is an appropriate interval to use the range is still small enough to provide an adequate view of the results of the study. Though this study does not show that the average patient will stay less than 3 days post-operatively the numbers are close enough that I believe this information could be helpful for surgeons when determining their post-op standard of care.
References
Zou X, N.R., Cong L, N.R., Yu Z, M.D., Wang J, N.R. (March 2019) Fast-track rehabilitation and nursing care in post-anesthesia care unit on orthopedic patients. Australian Journal of Advanced Nursing  (3):39-44.
Chamberlain University. (2019). Week 6 lesson: From samples to populations. Week 6 Lesson: From Samples to Populations: Statistical Reasoning for the Health Sciences-Riggs (instructure.com)