Sample size calculations for intervention trials in primary care randomizing by primary care group: an empirical illustration from one proposed intervention trial

Eldridge, Sandra; Cryer, Colin; Feder, Gene; Underwood, Martin

doi:10.1002/1097-0258(20010215)20:3<367::aid-sim798>3.0.co;2-r

Cited by 27 publications

(18 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on these assumptions, we calculated that 17.4 clinics with 172,680 patients were required for this study. Intracluster correlation coefficients (ICCs) were calculated using an analysis of variance method proposed by Eldridge, Cryer, Feder, and Underwood (2001) and programmed using S-PLUS software. Conservative (unadjusted) ICCs estimated from pilot data for ask, advise, and assist were 0.22, 0.15, and 0.22, respectively.…”

Section: Sample Size and Randomizationmentioning

confidence: 99%

Provider feedback to improve 5A's tobacco cessation in primary care: A cluster randomized clinical trial

Bentz

Bayley

Bonin

et al. 2007

CNTR

104

View full text Add to dashboard Cite

The electronic health record (EHR) may be an effective tool to help clinicians address tobacco use more consistently. To evaluate the impact of EHR-generated practice feedback on rates of referral to a state-level tobacco quitline, we conducted a cluster randomized clinical trial (feedback versus no feedback) within 19 primary care clinics in Oregon. Intervention clinics received provider-specific monthly feedback reports generated from EHR data. The reports rated provider performance in asking, advising, assessing, and assisting with tobacco cessation compared with a clinic average and an achievable benchmark of care. During 12 months of follow-up, EHR-documented rates of advising, assessing, and assisting were significantly improved in the intervention clinics compared with the control clinics (p<.001). A higher case-mix index and presence of a clinic champion were associated with higher rates of referral to a state-level quitline. EHR-generated provider feedback improved documentation of assistance with tobacco cessation. Connecting physician offices to a state-level quitline was feasible and well accepted.

show abstract

Section: Sample Size and Randomizationmentioning

confidence: 99%

Provider feedback to improve 5A's tobacco cessation in primary care: A cluster randomized clinical trial

Bentz

Bayley

Bonin

et al. 2007

CNTR

104

View full text Add to dashboard Cite

show abstract

“…The minimum sample size to detect the expected difference of proportions assuming individual randomization with a level of significance set at 95% (2-sided) and at 80% power was first calculated and adjustment was made by inflating the individually randomized sample size by a factor given by where is the average cluster size (10) and ρ the internal variability that represents how strongly individuals within clusters are related to each other (intracluster correlation (ICC)) [33]. Inflating the individually randomized sample size by this factor counteracts the loss of power due to a clustered design [34].…”

Section: Methodsmentioning

confidence: 99%

Continuous quality improvement interventions to improve long-term outcomes of antiretroviral therapy in women who initiated therapy during pregnancy or breastfeeding in the Democratic Republic of Congo: design of an open-label, parallel, group randomized trial

Yotebieng

Behets

Kawende

et al. 2017

BMC Health Serv Res

View full text Add to dashboard Cite

BackgroundDespite the rapid adoption of the World Health Organization’s 2013 guidelines, children continue to be infected with HIV perinatally because of sub-optimal adherence to the continuum of HIV care in maternal and child health (MCH) clinics. To achieve the UNAIDS goal of eliminating mother-to-child HIV transmission, multiple, adaptive interventions need to be implemented to improve adherence to the HIV continuum.MethodsThe aim of this open label, parallel, group randomized trial is to evaluate the effectiveness of Continuous Quality Improvement (CQI) interventions implemented at facility and health district levels to improve retention in care and virological suppression through 24 months postpartum among pregnant and breastfeeding women receiving ART in MCH clinics in Kinshasa, Democratic Republic of Congo. Prior to randomization, the current monitoring and evaluation system will be strengthened to enable collection of high quality individual patient-level data necessary for timely indicators production and program outcomes monitoring to inform CQI interventions. Following randomization, in health districts randomized to CQI, quality improvement (QI) teams will be established at the district level and at MCH clinics level. For 18 months, QI teams will be brought together quarterly to identify key bottlenecks in the care delivery system using data from the monitoring system, develop an action plan to address those bottlenecks, and implement the action plan at the level of their district or clinics.DiscussionIf proven to be effective, CQI as designed here, could be scaled up rapidly in resource-scarce settings to accelerate progress towards the goal of an AIDS free generation.Trial registrationThe protocol was retrospectively registered on February 7, 2017. ClinicalTrials.gov Identifier: NCT03048669.

show abstract

“…In trials with naturally small clusters, the extent of variation in the outcome between clusters, measured using the ICC, tends to be higher than for naturally large clusters. [21][22][23] In addition, there are usually a large number of potential clusters available, and generally, all eligible cluster members would be invited to take part in both the pilot and main trial. By contrast, for naturally larger clusters, sometimes due to costs or administrative burden only a subset of eligible members may be invited to participate and this may differ between the main trial and the pilot.…”

Section: Overview Of Pilot Studies For Cluster Randomised Trialsmentioning

confidence: 99%

How big should the pilot study for my cluster randomised trial be?

Eldridge

Costelloe

Kahan

et al. 2015

Stat Methods Med Res

116

View full text Add to dashboard Cite

There is currently a lot of interest in pilot studies conducted in preparation for randomised controlled trials. This paper focuses on sample size requirements for external pilot studies for cluster randomised trials. We consider how large an external pilot study needs to be to assess key parameters for input to the main trial sample size calculation when the primary outcome is continuous, and to estimate rates, for example recruitment rates, with reasonable precision. We used simulation to provide the distribution of the expected number of clusters for the main trial under different assumptions about the natural cluster size, intra-cluster correlation, eventual cluster size in the main trial, and various decisions made at the piloting stage. We chose intra-cluster correlation values and pilot study size to reflect those commonly reported in the literature. Our results show that estimates of sample size required for the main trial are likely to be biased downwards and very imprecise unless the pilot study includes large numbers of clusters and individual participants. We conclude that pilot studies will usually be too small to estimate parameters required for estimating a sample size for a main cluster randomised trial (e.g. the intra-cluster correlation coefficient) with sufficient precision and too small to provide reliable estimates of rates for process measures such as recruitment or follow-up rates.

show abstract

Sample size calculations for intervention trials in primary care randomizing by primary care group: an empirical illustration from one proposed intervention trial

Cited by 27 publications

References 13 publications

Provider feedback to improve 5A's tobacco cessation in primary care: A cluster randomized clinical trial

Provider feedback to improve 5A's tobacco cessation in primary care: A cluster randomized clinical trial

Continuous quality improvement interventions to improve long-term outcomes of antiretroviral therapy in women who initiated therapy during pregnancy or breastfeeding in the Democratic Republic of Congo: design of an open-label, parallel, group randomized trial

How big should the pilot study for my cluster randomised trial be?

Contact Info

Product

Resources

About