Dataset Information

Geographic pair-matching in large-scale cluster randomized trials.

ABSTRACT: Custer randomized trials are often used to study large-scale public health interventions. In large trials, even small improvements in statistical efficiency can have profound impacts on the required sample size and cost. Pair matched randomization is one strategy with potential to increase trial efficiency, but to our knowledge there have been no empirical evaluations of pair-matching in large-scale, epidemiologic field trials. Location integrates many socio-demographic and environmental characteristics into a single feature. Here, we show that geographic pair-matching leads to substantial gains in statistical efficiency for 14 child health outcomes that span growth, development, and infectious disease through a re-analysis of two large-scale trials of nutritional and environmental interventions in Bangladesh and Kenya. We estimate relative efficiencies ≥1.1 for all outcomes assessed and relative efficiencies regularly exceed 2.0, meaning an unmatched trial would have needed to enroll at least twice as many clusters to achieve the same level of precision as the geographically pair-matched design. We also show that geographically pair-matched designs enable estimation of fine-scale, spatially varying effect heterogeneity under minimal assumptions. Our results demonstrate broad, substantial benefits of geographic pair-matching in large-scale, cluster randomized trials.

SUBMITTER: Arnold BF

PROVIDER: S-EPMC10187339 | biostudies-literature | 2023 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Geographic pair-matching in large-scale cluster randomized trials.

Arnold Benjamin F BF Rerolle Francois F Tedijanto Christine C Njenga Sammy M SM Rahman Mahbubur M Ercumen Ayse A Mertens Andrew A Pickering Amy A Lin Audrie A Arnold Charles D CD Das Kishor K Stewart Christine P CP Null Clair C Luby Stephen P SP Colford John M JM Hubbard Alan E AE Benjamin-Chung Jade J

medRxiv : the preprint server for health sciences 20230523

Custer randomized trials are often used to study large-scale public health interventions. In large trials, even small improvements in statistical efficiency can have profound impacts on the required sample size and cost. Pair matched randomization is one strategy with potential to increase trial efficiency, but to our knowledge there have been no empirical evaluations of pair-matching in large-scale, epidemiologic field trials. Location integrates many socio-demographic and environmental charact ...[more]

PMID: 37205361

Similar Datasets

Project description:IntroductionInformation on benefits and risks of drugs is a key element affecting doctors' prescribing decisions. Outreach visits promoting independent information have proved moderately effective in changing prescribing behaviours.ObjectivesTesting the short and long-term effectiveness on general practitioners' prescribing of small groups meetings led by pharmacists.MethodsTwo cluster open randomised controlled trials (RCTs) were carried out in a large scale NHS setting. Ad hoc prepared evidence based material were used considering a therapeutic area approach--TEA, with information materials on osteoporosis or prostatic hyperplasia--and a single drug oriented approach--SIDRO, with information materials on me-too drugs of 2 different classes: barnidipine or prulifloxacin. In each study, all 115 Primary Care Groups in a Northern Italy area (2.2 million inhabitants, 1737 general practitioners) were randomised to educational small groups meetings, in which available evidence was provided together with drug utilization data and clinical scenarios. Main outcomes were changes in the six-months prescription of targeted drugs. Longer term results (24 and 48 months) were also evaluated.ResultsIn the TEA trial, one of the four primary outcomes showed a reduction (prescription of alfuzosin compared to tamsulosin and terazosin in benign prostatic hyperplasia: prescribing ratio -8.5%, p = 0.03). Another primary outcome (prescription of risedronate) showed a reduction at 24 and 48 months (-7.6%, p = 0.02; and -9,8%, p = 0.03), but not at six months (-5.1%, p = 0.36). In the SIDRO trial both primary outcomes showed a statistically significant reduction (prescription of barnidipine -9.8%, p = 0.02; prescription of prulifloxacin -11.1%, p = 0.04), which persisted or increased over time.InterpretationThese two cluster RCTs showed the large scale feasibility of a complex educational program in a NHS setting, and its potentially relevant long-term impact on prescribing habits, in particular when focusing on a single drug. National Health systems should invest in independent drug information programs.Trial registrationControlled-Trials.com ISRCTN05866587.

Dataset Information

Geographic pair-matching in large-scale cluster randomized trials.

Publications

Geographic pair-matching in large-scale cluster randomized trials.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets