Power calculator for instrumental variable analysis in pharmacoepidemiology

Abstract

General rightsThis document is made available in accordance with publisher policies. Please cite only the published version using the reference above. Abstract Background: Instrumental variable analysis, for example with physicians' prescribing preferences as an instrument for medications issued in primary care, is an increasingly popular method in the field of pharmacoepidemiology. Existing power calculators for studies using instrumental variable analysis, such as Mendelian randomization power calculators, do not allow for the structure of research questions in this field. This is because the analysis in pharmacoepidemiology will typically have stronger instruments and detect larger causal effects than in other fields. Consequently, there is a need for dedicated power calculators for pharmacoepidemiological research. Methods and Results: The formula for calculating the power of a study using instrumental variable analysis in the context of pharmacoepidemiology is derived before being validated by a simulation study. The formula is applicable for studies using a single binary instrument to analyse the causal effect of a binary exposure on a continuous outcome. An online calculator, as well as packages in both R and Stata, are provided for the implementation of the formula by others. Conclusions: The statistical power of instrumental variable analysis in pharmacoepidemiological studies to detect a clinically meaningful treatment effect is an important consideration. Research questions in this field have distinct structures that must be accounted for when calculating power. The formula presented differs from existing instrumental variable power formulae due to its parametrization, which is designed specifically for ease of use by pharmacoepidemiologists.

Substance P and Calcitonin gene‐related peptide expression in human periodontal ligament after root canal preparation with Reciproc Blue, WaveOne Gold, XP EndoShaper and hand files

“…The number of files needed to -Bucheli et al SP and CGRP in periodontal ligament after root canal preparation International Endodontic Journal, 51, 1358-1366, 2018 prepare the canal and the pressure applied to the instruments during the up-down movement could also influence the results obtained due to a greater possibility of debris and irrigant solutions extrusion to the periodontal tissues (Ferraz et al 2001, Caviedes-Bucheli et al 2010). Significant differences in SP and CGRP expression between both control groups validate the study design, where the power of the sample was estimated to be 100% (Walker et al 2017). Moreover, similar tendencies in neuropeptide expression behaviour on the other groups also corroborate the experimental design providing data with good levels of confidence, as both of the neuropeptides measured are originated from the same trigeminal nerve fibres.…”

“…Significant differences in SP and CGRP expression between both control groups validate the study design, where the power of the sample was estimated to be 100% (Walker et al . ). Moreover, similar tendencies in neuropeptide expression behaviour on the other groups also corroborate the experimental design providing data with good levels of confidence, as both of the neuropeptides measured are originated from the same trigeminal nerve fibres.…”

Mendelian randomization

“…Power calculators are available for simple MR studies to determine whether a particular sample size is sufficient for estimation to give reasonably precise results. [33][34][35][36] Simulation studies to determine power are also commonly used to accommodate unique data features. 37 The association of the proposed genetic instrument with the exposure can be estimated in a sample other than that used to estimate the effect of the proposed genetic instrument on the outcome.…”

Effects of Education Savings Accounts on Student Engagement: Instrumental Variable Analysis

“…However, given the recommendation to emphasize estimation and confidence interval with a lower threshold (e.g., 75% or 85%) for pilot studies (Lee et al 2014), we also report the 90% confidence interval that corresponds to a .10 significance level. A post-hoc power calculation for instrumental variable analysis using the PharmIV Stata command (Walker et al 2017) and the recommended significance threshold for pilot studies (Lee et al 2014) yielded statistical power between 72 and 86% to detect a medium to strong effect, as expected for a small pilot study. Also, per the recommended N < 70 threshold for pilot and feasibility studies (Teare et al 2014) the current study's analytic sample size (132) exceeds the lower threshold.…”