PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 8981683

Borrowing strength from external trials in a meta-analysis.

J P Higgins·A Whitehead
View original source ↗Export metadata ↓

Abstract

There exists a variety of situations in which a random effects meta-analysis might be undertaken using a small number of clinical trials. A problem associated with small meta-analyses is estimating the heterogeneity between trials. To overcome this problem, information from other related studies may be incorporated into the meta-analysis. A Bayesian approach to this problem is presented using data from previous meta-analyses in the same therapeutic area to formulate a prior distribution for the heterogeneity. The treatment difference parameters are given non-informative priors. Further, related trials which compare one or other of the treatments of interest with a common third treatment are included in the model to improve inference on both the heterogeneity and the treatment difference. Two approaches to estimating relative efficacy are considered, namely a general parametric approach and a method explicit to binary data. The methodology is illustrated using data from 26 clinical trials which investigate the prevention of cirrhosis using beta-blockers and sclerotherapy. Both sources of external information lead to more precise posterior distributions for all parameters, in particular that representing heterogeneity.

Explore related subjects

Adrenergic beta-AntagonistsBayes TheoremData Interpretation, StatisticalGastrointestinal HemorrhageHumansLikelihood FunctionsLiver CirrhosisMeta-Analysis as TopicModels, StatisticalMultivariate AnalysisOdds RatioRandomized Controlled Trials as TopicSclerotherapyComparative StudyJournal ArticleResearch Support, Non-U.S. Gov't

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

J P Higgins, A Whitehead. 1996-12-30. Borrowing strength from external trials in a meta-analysis.. https://doi.org/10.1002/(sici)1097-0258(19961230)15%3A24%3C2733%3A%3Aaid-sim562%3E3.0.co%3B2-0

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2007

Bayesian sensitivity analysis for unmeasured confounding in observational studies.

We consider Bayesian sensitivity analysis for unmeasured confounding in observational studies where the association between a binary exposure, binary response, measured confounders and a single binary unmeasured confounder can be formulated using logistic regression models. A model for unmeasured confounding is presented along with a family of prior distributions that model beliefs about a possible unknown unmeasured confounder. Simulation from the posterior distribution is accomplished using Markov chain Monte Carlo. Because the model for unmeasured confounding is not identifiable, standard large-sample theory for Bayesian analysis is not applicable. Consequently, the impact of different choices of prior distributions on the coverage probability of credible intervals is unknown. Using simulations, we investigate the coverage probability when averaged with respect to various distributions over the parameter space. The results indicate that credible intervals will have approximately nominal coverage probability, on average, when the prior distribution used for sensitivity analysis approximates the sampling distribution of model parameters in a hypothetical sequence of observational studies. We motivate the method in a study of the effectiveness of beta blocker therapy for treatment of heart failure.

Adrenergic beta-Antagonists↗
PubMed2007

[Polypharmacy, compliance and non-prescription medication in patients with cardiovascular disease in Germany].

BACKGROUND AND OBJECTIVE: The expenses of health insurances are continuously raising. With implementation of evidence-based medicine resulting in polypharmacy, compliance is decreasing and patients as well as physicians are facing unintentional drug interactions. Furthermore the arbitrary use of additional drugs apart from prescribed medication has to be considered. The objective of this study was to analyse polypharmacy, compliance and utilisation patterns for non-prescription medications (OTC) in patients with cardiovascular diseases. METHODS: 100 patients with cardiovascular diseases (45 women, 55 men, 58 - 87 years) were interviewed using a questionnaire. The compliance was determined by the Morisky-Score. RESULTS: 78 % of the patients received more than four pills every day (median 8.34). Most common products were beta-blockers (89 %), ACE-Inhibitors/Sartans (69 %) and aspirin (65 %). Only 52 % of the patients knew the indications of their medication. Although 83 % of the patients claimed to be absolutely compliant concerning their medication, the Morisky-Score revealed a high compliance only in 52 %. The compliance decreased significantly if the number of prescribed medication increased to more than four pills a day. 48 % of the patients took regularly non-prescription products, 35 % more than three additional products daily. Most commonly multivitamins, minerals, herbals and non-steroidal anti-inflammatory drugs were used. This non-prescription medication did not affect the compliance of the patients. CONCLUSIONS: Polypharmacy with more than four pills daily leads to a lower compliance and can therefore influence the implementation of guideline-medicine. Non-prescription medication is widely used and should be considered because of their potential side-effects and drug interactions.

Adrenergic beta-Antagonists↗
PubMed2007

Propranolol blocks chronic risperidone treatment-induced enhancement of spatial working memory performance of rats in a delayed matching-to-place water maze task.

RATIONALE: Atypical antipsychotics improve cognitive function, including working memory, in schizophrenia. Some atypical antipsychotics have been reported to activate the locus coeruleus and induce beta-adrenoceptor antagonist sensitive c-Fos-like immunoreactivity in the prefrontal cortex. MATERIALS AND METHODS: The present study investigated the effects of chronic treatment of rats with risperidone (1 mg kg(-1) day(-1) s.c.), clozapine (10 mg kg(-1) day(-1) s.c.), or acidified saline vehicle control for 2, 4, or 8 weeks on spatial working memory performance in a delayed matching-to-place water maze task with a 60-s inter-trial retention interval with and without acute challenge with propranolol (10 mg/kg i.p.). RESULTS: Treatment with risperidone for 8 weeks, but not 2 or 4 weeks, significantly improved working memory performance. In contrast, treatment with clozapine for up to 8 weeks did not improve working memory. Acute challenge with propranolol blocked the improvement in working memory produced by chronic treatment with risperidone, but had no significant effect on performance in saline- or clozapine-treated animals. CONCLUSIONS: The delayed matching-to-place water maze task may prove valuable in the investigation of the behavioural pharmacology of the cognitive effects of antipsychotic drugs. These data suggest that beta adrenoceptors may contribute to the cognitive effects of chronic treatment with atypical antipsychotics.

Adrenergic beta-Antagonists↗