DQ 2 : Define clinical significance, and explain the difference between clinical and statistical significance.

Statistical significance is a statement about probability under a null hypothesis: if there were truly no effect, how often would we see a difference at least this large? Clinical significance is a statement about whether the difference matters to a patient or a service. The two come apart in both directions and the reason is sample size. A very large study can produce a highly significant reduction in length of stay of two hours, which no one would reorganise a unit for. A small capstone project on one thirty-bed unit can produce a fall rate that halves and still returns a p-value above 0.05, because with that few events the study never had the power to detect anything. The second case is yours, and knowing why it happens is the difference between reporting a failure and reporting an underpowered study with a promising effect. Both directions matter, because knowing only the small-study case is half the idea.

So report the things that carry clinical significance rather than arguing about the p-value. Give the effect size and the absolute difference, not just the percentage change, because a fall rate moving from four to two per thousand patient days is both a halving and a small absolute number and a reader deserves both. Give the confidence interval, which shows the range of effects your data are compatible with and often makes clear that a clinically important benefit has not been excluded. Where one exists, cite the minimal clinically important difference for your outcome and say whether you reached it. Add the practical measures that never appear in a p-value at all: staff adoption, patient-reported experience, cost avoided, harm avoided. Then state honestly what the result does and does not support — usually continued use with continued measurement rather than a claim of proven effectiveness, which is what a careful marker is looking for.