Search PubMed⌕ Search

Biomedical subjects

F Mosteller

Publications and source records attributed to F Mosteller.

At least 19 recordsLinked to original sources
PubMed1998

Meta-analysis of multiple outcomes by regression with random effects.

Earlier work showed how to perform fixed-effects meta-analysis of studies or trials when each provides results on more than one outcome per patient and these multiple outcomes are correlated. That fixed-effects generalized-least-squares approach analyzes the multiple outcomes jointly within a single model, and it can include covariates, such as duration of therapy or quality of trial, that may explain observed heterogeneity of results among the trials. Sometimes the covariates explain all the heterogeneity, and the fixed-effects regression model is appropriate. However, unexplained heterogeneity may often remain, even after taking into account known or suspected covariates. Because fixed-effects models do not make allowance for this remaining unexplained heterogeneity, the potential exists for bias in estimated coefficients, standard errors and p-values. We propose two random-effects approaches for the regression meta-analysis of multiple correlated outcomes. We compare their use with fixed-effects models and with separate-outcomes models in a meta-analysis of periodontal clinical trials. A simulation study shows the advantages of the random-effects approach. These methods also facilitate meta-analysis of trials that compare more than two treatments.

Clinical Trials as Topic↗
PubMed1998

The comparative effects of postoperative analgesic therapies on pulmonary outcome: cumulative meta-analyses of randomized, controlled trials.

UNLABELLED: We performed meta-analyses of randomized, control trials to assess the effects of seven analgesic therapies on postoperative pulmonary function after a variety of procedures: epidural opioid, epidural local anesthetic, epidural opioid with local anesthetic, thoracic versus lumbar epidural opioid, intercostal nerve block, wound infiltration with local anesthetic, and intrapleural local anesthetic. Measures of forced expiratory volume in 1 s (FEV1), forced vital capacity (FVC), vital capacity (VC), peak expiratory flow rate (PEFR), PaO2, and incidence of atelectasis, pulmonary infection, and pulmonary complications overall were analyzed. Compared with systemic opioids, epidural opioids decreased the incidence of atelectasis (risk ratio [RR] 0.53, 95% confidence interval [CI] 0.33-0.85) and had a weak tendency to reduce the incidence of pulmonary infections (RR 0.53, 95% CI 0.18-1.53) and pulmonary complications overall (RR 0.51, 95% CI 0.20-1.33). Epidural local anesthetics increased PaO2 (difference 4.56 mm Hg, 95% CI 0.058-9.075) and decreased the incidence of pulmonary infections (RR 0.36, 95% CI 0.21-0.65) and pulmonary complications overall (RR 0.58, 95% CI 0.42-0.80) compared with systemic opioids. Intercostal nerve blockade tends to improve pulmonary outcome measures (incidence of atelectasis: RR 0.65, 95% CI 0.27-1.57, incidence of pulmonary complications overall: RR 0.47, 95% CI 0.18-1.22), but these differences did not achieve statistical significance. There were no clinically or statistically significant differences in the surrogate measures of pulmonary function (FEV1, FVC, and PEFR). These analyses support the utility of epidural analgesia for reducing postoperative pulmonary morbidity but do not support the use of surrogate measures of pulmonary outcome as predictors or determinants of pulmonary morbidity in postoperative patients. IMPLICATIONS: When individual trials are unable to produce significant results, it is often because of insufficient patient numbers. It may be impossible for a single institution to study enough patients. Meta-analysis is a useful tool for combining the data from multiple trials to increase the patient numbers. These meta-analyses confirm that postoperative epidural pain control can significantly decrease the incidence of pulmonary morbidity.

Administration, Oral↗
PubMed1996

Simple robust procedures for combining risk differences in sets of 2 x 2 tables.

Meta-analyses often use a random-effects model to incorporate unexplained heterogeneity of study results. Trimmed versions of meta-analytic estimators for the risk difference, adapted from procedures designed for a random-effects analysis, can resist the impact of a few anomalous studies. A simulation study compared untrimmed and trimmed versions of four meta-analytic procedures that give weighted averages of risk differences. An adaptation of Winsorized estimates of components of variance gains some resistance to anomalous studies when estimating variability. The simulations found that a modified version of the DerSimonian-Laird estimator is attractive when risk differences reveal the added variability described by a random-effects model, and that a 20 per cent trimmed, weighted version of this procedure offers resistance against the impact of highly anomalous results. Among four trimmed procedures considered, the trimmed version of the modified DerSimonian-Laird estimator offers the best performance over a wide range of simulation designs and sample sizes. None of the methods, whether trimmed or untrimmed, is uniformly preferable. A published meta-analysis of a vaccination against TB provides data that serve to illustrate differences among the eight procedures.

BCG Vaccine↗
PubMed1996

Uncertainty of the time of first significance in random effects cumulative meta-analysis.

We propose a process for evaluating quantities of clinical and statistical interest in cumulative meta-analysis. We use Monte Carlo simulation studies to assess the error variance of the time to significance in cumulative meta-analysis. For the specific cumulative meta-analyses that we simulated, the 95% confidence interval of the treatment effect, estimated by the random effects method at the time of earliest significance, appears to be approximately appropriate except when the hypothesized treatment effect is near the null. The Monte Carlo approach that we used can also estimate the power of a cumulative meta-analysis when two treatments differ in efficacy. We illustrate these issues in the context of five cumulative meta-analyses, each comparing two treatments for preventing mortality from myocardial infarction. These simulated meta-analyses demonstrate our main point, which is that the time of first significance, however parameterized, is itself a random variable with error variance.

Bias↗
PubMed1996

Understanding research synthesis (meta-analysis).

Synthesis of research findings has long been a part of reviewing and summarizing a field of study. Public health decisions are made on the available evidence. We summarize the approaches to research synthesis that draw on the best available evidence and the use of quantitative summaries through meta-analysis. We focus on observational studies. Heterogeneity offers the potential to observe a relation across study populations and circumstances. We emphasize the benefits of heterogeneity in overviews and the need to explore and describe the sources of heterogeneity. Random effects approaches to combining data are recommended, and the use of regression approaches is emphasized. Excluding studies with extreme results may bias a research synthesis and underestimate the true variance of the results, thus contributing to misleading inference. Thorough searching is the best guard against publication bias. We conclude with guidelines for combining epidemiological studies.

Analysis of Variance↗
PubMed1995

A random-effects regression model for meta-analysis.

Many meta-analyses use a random-effects model to account for heterogeneity among study results, beyond the variation associated with fixed effects. A random-effects regression approach for the synthesis of 2 x 2 tables allows the inclusion of covariates that may explain heterogeneity. A simulation study found that the random-effects regression method performs well in the context of a meta-analysis of the efficacy of a vaccine for the prevention of tuberculosis, where certain factors are thought to modify vaccine efficacy. A smoothed estimator of the within-study variances produced less bias in the estimated regression coefficients. The method provided very good power for detecting a non-zero intercept term (representing overall treatment efficacy) but low power for detecting a weak covariate in a meta-analysis of 10 studies. We illustrate the model by exploring the relationship between vaccine efficacy and one factor thought to modify efficacy. The model also applies to the meta-analysis of continuous outcomes when covariates are present.

BCG Vaccine↗
PubMed1995

Multiple-outcomes meta-analysis of treatments for periodontal disease.

The results of periodontal therapy vary by disease severity, outcome measure, and method of data analysis. Several clinical trials and a subsequent meta-analysis have demonstrated that, for teeth with severe disease, surgery decreases probing depth (PD) and increases attachment level (AL) more than non-surgical treatment. For other disease levels, the choice of therapy depends on the outcome measure. When clinical trials use two or more outcome measures (such as PD and AL), investigators ordinarily analyze each outcome separately. When the correlations are incorporated among the outcomes, a meta-analysis can use generalized-least-squares (GLS) regression to analyze multiple outcomes jointly. We applied the GLS multiple-outcomes model in a meta-analysis of 5 trials comparing surgical and non-surgical periodontal treatments, each assessing the outcomes PD and AL one year after treatment. The clinical conclusions are similar to those reported earlier, but our estimates of the relative benefits of surgical and non-surgical treatment should be more accurate, because the GLS method takes into account correlation between AL and PD. When correlations between the two outcomes rise, as they do with increasing severity of disease, the GLS estimates depart from those derived from separate analyses of PD and AL.

Dental Scaling↗
PubMed1995

The efficacy of bacillus Calmette-Guérin vaccination of newborns and infants in the prevention of tuberculosis: meta-analyses of the published literature.

OBJECTIVE: To quantify the efficacy of vaccination of infants with bacillus Calmette-Guérin (BCG) against tuberculosis. DATA SOURCES: MEDLINE with index terms BCG vaccine, tuberculosis, and human; lists of all known studies provided by experts at the Centers for Disease Control and Prevention, the World Health Organization, and other organizations. STUDY SELECTION: A total of 1264 articles and abstracts were reviewed for details on BCG vaccination, the availability of concurrent vaccinated and unvaccinated groups, and a tuberculosis outcome. Seventy articles were reviewed in depth for method of vaccine allocation used to create comparable groups, age at vaccination of study participants, comparability of surveillance and follow-up of recipient and concurrent control groups in trials, an appropriately defined control group in case-control studies, and outcome measures (tuberculosis cases and/or deaths). Five prospective trials and eleven case-control studies of vaccination during infancy were included in the present analyses. DATA EXTRACTION: We recorded study design, age range of study population, number of patients enrolled, efficacy of vaccine, location of the study, and a series of items to assess the potential for bias in study design, follow-up, and diagnosis. We extracted or computed vaccine efficacy by years since vaccination wherever possible. At least two readers independently extracted data and evaluated data validity. DATA SYNTHESIS: The relative risk (RR) or odds ratio (OR) for tuberculosis in vaccinated versus unvaccinated infants was the measure of vaccine efficacy analyzed. A random-effects method estimated a weighted average RR or OR from data extracted from the trials and case-control studies. The protective effect was then computed by 1-RR or 1-OR. Overall, the protective effect of vaccination against cases of tuberculosis was 0.74 (95% confidence interval [95% CI], 0.62 to 0.83) when estimated from four randomized controlled trials, and 0.52 (95% CI, 0.38 to 0.64) when estimated from nine case-control studies. Five trials reporting deaths from tuberculosis showed a BCG protective effect of 0.65 (95% CI, 0.12 to 0.86), five studies reporting on meningitis showed a protective effect of 0.64 (95% CI, 0.30 to 0.82), and three studies of disseminated tuberculosis showed a protective effect of 0.78 (95% CI, 0.58 to 0.88). Three case-control studies included separate results for laboratory-confirmed cases of tuberculosis. These studies documented a protective effect of 0.83 (95% CI, 0.58 to 0.93). In a random-effects regression model of the nine case-control studies, study validity score explained 15% of the heterogeneity among study-estimated protective effects, suggesting that better studies reported greater efficacy. Three trials and six case-control studies provided some age-specific information that allowed us to examine the duration of BCG efficacy. Most of this evidence suggested that BCG efficacy may persist through 10 years after infant vaccination. CONCLUSION: BCG vaccination of newborns and infants significantly reduces the risk of tuberculosis--by over 50%, on average. Protection has been observed across many populations, study designs, and forms of tuberculosis. Rates of protection against cases that are confirmed by laboratory tests, reflecting reduced error in disease classification and consequently more accurate estimates of BCG efficacy, are highest at 83%.

BCG Vaccine↗
PubMed1995

A survey of current problems in meta-analysis. Discussion from the Agency for Health Care Policy and Research inter-PORT Work Group on Literature Review/Meta-Analysis.

The United States Agency for Health Care Policy and Research brought together representatives from each of the Patient Outcomes Research Teams (PORTs) as the Inter-PORT Meta-Analysis Work Group. The purpose of the meeting was to discuss a wide range of problems the panel had encountered in the area of meta-analysis. Several actual problems were presented, and tentative solutions to those problems were given. The PORTs had to consider issues such as quality assessment, survival analysis, rare events, sensitivity and specificity, and random effects models. The solutions presented represent the methods used by several researchers who are active in the area of meta-analysis.

Evaluation Studies as Topic↗
PubMed1994

Guidelines for meta-analyses evaluating diagnostic tests.

OBJECTIVES: To introduce guidelines for the conduct, reporting, and critical appraisal of meta-analyses evaluating diagnostic tests and to apply these guidelines to recently published meta-analyses of diagnostic tests. DATA SOURCES: Based on current concepts of how to assess diagnostic tests and conduct meta-analyses. They are applied to all meta-analyses evaluating diagnostic tests published in English-language journals from January 1990 through December 1991, identified through MEDLINE searching and by experts in the field. STUDY SELECTION: Meta-analyses were included if at least two of three independent readers regarded their main purpose as the evaluation of diagnostic tests against a concurrent reference standard. DATA EXTRACTION: By three independent readers on the extent to which meta-analyses fulfilled each guideline, with consensus defined as agreement by at least two readers. DATA SYNTHESIS: The guidelines are concerned with determining the objective of the meta-analysis, identifying the relevant literature and extracting the data, estimating diagnostic accuracy, and identifying the extent to which variability is explained by study design characteristics and characteristics of the patients and diagnostic test. In general, the guidelines were only partially fulfilled. CONCLUSION: Meta-analysis is potentially important in the assessment of diagnostic tests. Those reading meta-analyses evaluating diagnostic tests should critically appraise them; those doing meta-analyses should apply recently developed methods. The conduct and reporting of primary studies on which meta-analyses are based require improvement.

Bias↗
PubMed1994

Efficacy of BCG vaccine in the prevention of tuberculosis. Meta-analysis of the published literature.

OBJECTIVE: To quantify the efficacy of BCG vaccine against tuberculosis (TB). DATA SOURCES: MEDLINE with index terms BCG vaccine, tuberculosis, and human. Experts from the Centers for Disease Control and Prevention and the World Health Organization, among others, provided lists of all known studies. STUDY SELECTION: A total of 1264 articles or abstracts were reviewed for details on BCG vaccination, concurrent vaccinated and unvaccinated groups, and TB outcome; 70 articles were reviewed in depth for method of vaccine allocation used to create comparable groups, equal surveillance and follow-up for recipient and concurrent control groups, and outcome measures of TB cases and/or deaths. Fourteen prospective trials and 12 case-control studies were included in the analysis. DATA EXTRACTION: We recorded study design, age range of study population, number of patients enrolled, efficacy of vaccine, and items to assess the potential for bias in study design and diagnosis. At least two readers independently extracted data and evaluated validity. DATA SYNTHESIS: The relative risk (RR) or odds ratio (OR) of TB provided the measure of vaccine efficacy that we analyzed. The protective effect was then computed by 1-RR or 1-OR. A random-effects model estimated a weighted average RR or OR from those provided by the trials or case-control studies. In the trials, the RR of TB was 0.49 (95% confidence interval [CI], 0.34 to 0.70) for vaccine recipients compared with nonrecipients (protective effect of 51%). In the case-control studies, the OR for TB was 0.50 (95% CI, 0.39 to 0.64), or a 50% protective effect. Seven trials reporting tuberculous deaths showed a protective effect from BCG vaccine of 71% (RR, 0.29; 95% CI, 0.16 to 0.53), and five studies reporting on meningitis showed a protective effect from BCG vaccine of 64% (OR, 0.36; 95% CI, 0.18 to 0.70). Geographic latitude of the study site and study validity score explained 66% of the heterogeneity among trials in a random-effects regression model. CONCLUSION: On average, BCG vaccine significantly reduces the risk of TB by 50%. Protection is observed across many populations, study designs, and forms of TB. Age at vaccination did not enhance predictiveness of BCG efficacy. Protection against tuberculous death, meningitis, and disseminated disease is higher than for total TB cases, although this result may reflect reduced error in disease classification rather than greater BCG efficacy.

BCG Vaccine↗
PubMed1994

Efficacy and safety of nonsteroidal antiinflammatory drugs for cancer pain: a meta-analysis.

PURPOSE: To assess the efficacy and safety of nonsteroidal antiinflammatory drugs (NSAIDs) in the treatment of cancer pain by meta-analyses of the published randomized control trials (RCTs). PATIENTS AND METHODS: Twenty-five studies met inclusion criteria for analysis. Of these, 13 tested a single-dose effect, nine multiple-dose effects, and three both single- and multiple-dose effects of 16 different NSAIDs in a total of 1,545 patients. Baseline pain intensity (when provided) of moderate or higher was indicated in 81% of patients. RESULTS: Single-dose NSAID studies found greater analgesic efficacy than placebo, with rough equivalence to 5 to 10 mg of intramuscular morphine. Pain scores differed insignificantly for aspirin versus three other NSAIDs. Analgesic responses to low- and high-dose NSAIDs suggested a dose-response relationship, but this was not statistically significant. Recommended and supramaximal single doses of three NSAIDs produced comparable changes in pain scores, which indicates a ceiling analgesic effect. Common side effects included upper gastrointestinal symptoms, dizziness, and drowsiness. The incidence of side effects showed a trend to increase with dose, without a ceiling effect, and to increase with multiple doses. Single or multiple doses of weak opioids (WO) alone or in combination (WO/C) with nonopioid analgesics did not produce greater analgesia than NSAIDs alone. Single doses of WO/C analgesics produced more side effects than NSAIDs alone, although both side effect incidence and patient dropout rates were equal when multiple doses were administered. CONCLUSION: These findings question whether the traditional World Health Organization (WHO) second analgesic step (addition of a weak opioid when pain is inadequately treated by a nonopioid analgesic alone) is warranted. A lack of comparable studies precluded testing the hypothesis that NSAIDs are particularly effective for malignant bone pain.

Anti-Inflammatory Agents, Non-Steroidal↗
PubMed1994

Improving health: measuring effects of medical care.

The impact of medical care on the quality and length of life of the population has been poorly documented. The rapid growth of evidence of efficacy of therapy for individual medical conditions now offers the opportunity to create an inventory of benefits. A method for creating such an inventory is described, as is its application to a selection of condition-treatment pairs, chosen for their high incidence of prevalence, their serious outcomes, and the demonstrated efficacy of their treatment. An aggregate effect of medical care on life expectancy is found to be roughly five years during this century, with a further potential of two years. Although there is no overall index of quality of life analogous to life expectancy, our inventory demonstrates the enormous burden of pain, suffering, and dysfunction that afflicts the population for which medical care can provide a large measure of relief.

Adolescent↗
PubMed1993

Vitamin A supplementation and child mortality. A meta-analysis.

OBJECTIVE: A two-part meta-analysis of studies examining the relationship of vitamin A supplementation and child mortality. DATA SOURCES: We identified studies by searching the MEDLARS database from 1966 through 1992 and by scanning Current Contents and bibliographies of pertinent articles. STUDY SELECTION: All 12 vitamin A controlled trials with data on mortality identified in the search were used in the analysis. DATA EXTRACTION: Data were independently extracted by two investigators who also assessed the quality of each study using a previously described method. DATA SYNTHESIS: We formally tested for heterogeneity across studies. We pooled studies using the Mantel-Haenszel and the DerSimonian and Laird methods and adjusted for the effect of cluster assignment of treatment groups in community-based studies. Vitamin A supplementation to hospitalized measles patients was highly protective against mortality (DerSimonian and Laird odds ratio, 0.39; 95% confidence interval, 0.22 to 0.66; P = .0004) (part 1 of the meta-analysis). Supplementation was also protective against overall mortality in community-based studies (DerSimonian and Laird odds ratio, 0.70; clustering-adjusted 95% confidence interval, 0.56 to 0.87; P = .001) (part 2 of the meta-analysis). CONCLUSIONS: Vitamin A supplements are associated with a significant reduction in mortality when given periodically to children at the community level. Factors that affect the bioavailability of large doses of Vitamin A need to be studied further. Vitamin A supplements should be given to all measles patients in developing countries whether or not they have symptoms of vitamin A deficiency.

Child↗
PubMed1993

Assessing the effectiveness of ambulatory cardiac monitoring for specific clinical indications. Introduction.

This introduction and the three essays that follow examine ambulatory cardiac monitoring for specific clinical indications. They also examine the ways in which evidence from the literature may be synthesized through the framework of decision analysis to guide its appropriate use and identify areas in which more research is needed. The essays discuss ambulatory cardiac monitoring for evaluation of syncope in the elderly; detection of silent ischemia after a myocardial infarction; and selection of antiarrhythmic drugs for malignant ventricular arrhythmias.

Aged↗