Assessment Of Treatment Effect Evidence

1251 Words Jul 11th, 2016 6 Pages
3.4. Imprecision
Similar to the imprecision assessment of treatment effect evidence, the imprecision assessment for relative importance of outcomes scrutinise the confidence interval and the sample size. What is challenging is because diversified study designs, result presentations or methodologies are used, usually there is no confidence interval constructed, or there is not a simple way to calculate the minimum sample size with sufficient power to produce a narrow estimate for the relative importance outcomes.[53]
3.4.1. Confidence interval
3.4.1.1. Is the confidence interval narrow (for systematic reviews)? (Answer options: Yes; probably yes; probably no; No)
The answers of “yes” or “probably yes” suggest a precise estimate, while “no” or “probably no” suggest imprecision estimate.
Rationale: To consider imprecision, systematic review authors need to decide whether confidence interval around the estimate is sufficient narrow. Unfortunately, we do not have an empirical narrow confidence interval for relative importance of outcomes evidence.
Due to diversity in study designs, methodology and presentation of results, there could be situations without confidence intervals, for example, the result reported could be the proportion of willing to accept the prophylaxis treatment if the risk of adverse events decreased by 5%.
3.4.1.2. Does the confidence interval exclude the clinical decision threshold between recommending and not recommending a treatment (for clinical practice…
Open Document