Search PubMed⌕ Search

PubMed · 10474144

Meta-analysis by random effect modelling in generalized linear models.

Abstract

The meta-analysis of multi-centre trials can be based on either fixed or random effect models. This paper argues for the general use of random effect models, and illustrates the value of non-parametric maximum likelihood (NPML) analysis of such trials. The same general approach unifies administrative 'league table' analyses in epidemiological and other studies. Several examples of the NPML analysis are given, including a 70-centre trial.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

M Aitkin. Meta-analysis by random effect modelling in generalized linear models.. https://doi.org/10.1002/(sici)1097-0258(19990915%2F30)18%3A17%2F18%3C2343%3A%3Aaid-sim260%3E3.0.co%3B2-3

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

KEEP EXPLORING

Related citations

Implementing guidelines in primary care: can population impact measures help?

BACKGROUND: Primary care organisations are faced with implementing a large number of guideline recommendations. We present methods by which the number of eligible patients requiring treatment, and the relative benefits to the whole population served by a general practice or Primary Care Trust, can be calculated to help prioritise between different guideline recommendations. METHODS: We have developed measures of population impact, "Number to be Treated in your Population (NTP)" and "Number of Events Prevented in your Population (NEPP)". Using literature-based estimates, we have applied these measures to guidelines for pharmacological methods of secondary prevention of myocardial infarction (MI) for a hypothetical general practice population of 10,000. RESULTS: Implementation of the NICE guidelines for the secondary prevention of MI will require 176 patients to be treated with aspirin, 147 patients with beta-blockers and with ACE-Inhibitors and 157 patients with statins (NTP). The benefit expressed as NEPP will range from 1.91 to 2.96 deaths prevented per year for aspirin and statins respectively. The drug cost per year varies from euro 1940 for aspirin to euro 60,525 for statins. Assuming incremental changes only (for those not already on treatment), aspirin post MI will be added for 37 patients and produce 0.40 of a death prevented per year at a drug cost of euro 410 and statins will be added for 120 patients and prevent 2.26 deaths per year at a drug cost of euro 46,150. An appropriate policy might be to reserve the use of statins until eligible patients have been established on aspirin, ACE-Inhibitors and beta blockers. CONCLUSIONS: The use of population impact measures could help the Primary Care Organisation to prioritise resource allocation, although the results will vary according to local conditions which should be taken into account before the measures are used in practice.

Adrenergic beta-Antagonists↗

C-reactive protein and ischemia in users and nonusers of beta-blockers and statins: data from the Heart and Soul Study.

BACKGROUND: Elevated levels of C-reactive protein (CRP) are associated with an increased risk of coronary events, but whether inflammation is associated with inducible ischemia in patients with stable coronary disease is unknown. METHODS AND RESULTS: We recruited patients with known coronary disease from 2 VA Medical Centers and 1 University-based medical center for the Heart and Soul Study. We measured CRP levels in 118 participants who had exercise-induced ischemia and in 111 who did not have inducible ischemia, as determined by stress echocardiography. We used logistic regression to examine the risk of exercise-induced ischemia associated with elevated CRP. We found that 75% (39/52) of participants in the highest CRP category (>0.38 mg/dL) had inducible ischemia, compared with 45% (79/177) in the lower 4 categories combined (adjusted odds ratio 4.2; 95% confidence interval 1.6 to 11; P=0.004). However, this association differed in users and nonusers of beta-blockers and statins. Among 89 participants who did not use beta-blockers, 93% in the highest CRP category had exercise-induced ischemia, compared with 42% in the lower 4 categories (P=0.03). Among 67 participants who did not use statins, 94% in the highest CRP category had exercise-induced ischemia, compared with 44% in the lower 4 categories (P=0.009). We did not observe a significant association between CRP and ischemia among participants who were treated with either of these medications. CONCLUSION: Elevated CRP levels are associated with inducible ischemia in patients with stable coronary disease, particularly among those not treated with beta-blockers or statins.

Adrenergic beta-Antagonists↗