Evaluating A Quantitative Research Study Discussion

Work through the framework your course supplies, but let the study's own features drive what you say at each step, because a template filled in generically reads as a template. Start with the question and the design: this is an intervention study of a brief behaviour-change intervention delivered in general practice, which raises the design questions that always attend that situation. Was allocation at the level of the practice or the patient, and if practices were allocated, does the analysis account for the clustering, since patients within a practice are more alike than patients across practices and ignoring that inflates significance. Was blinding possible at all, which for a lifestyle intervention it usually is not for participants or providers, and what that does to the risk of bias. Say which of those the paper reports and which it does not, because an appraisal that only describes what was done misses half of what an appraisal is for.

Then interrogate the outcomes, which is where a behaviour-change trial is most vulnerable. The SNAP risk factors are smoking, nutrition, alcohol and physical activity, and several of those are measured by self-report, which is subject to social desirability bias in a direction that favours the intervention. Ask what was measured objectively and what was not, whether the follow-up was long enough for a behaviour change to persist rather than merely occur, and how much loss to follow-up there was and whether the analysis was by intention to treat. Then read the results with the same care you would want applied to your own work: was the effect statistically significant, was it clinically meaningful, and does the confidence interval exclude an effect too small to matter. Finish with applicability, since a trial in one country's primary care system does not automatically transfer to another's, and saying which features would have to hold is more useful than a general caution.