PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Effect Modifier, Epidemiologic”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2Linked to original sources
PubMed1992

How much 'better' is good enough? The magnitude of treatment effect in clinical trials.

OBJECTIVES: Among the various factors required to calculate sample size for clinical trials, the magnitude of treatment effect anticipated is an important component. The objective of this report is to present some of the complexities involved in selection of treatment effect size in clinical trials. As a framework for discussion, an analysis of published reports related to surfactant therapy was carried out. DESIGN: Twenty-one consecutive exogenous surfactant trials for neonatal respiratory distress syndrome were analyzed. The "Methods" sections were reviewed for evaluating various components of sample size calculation, including the anticipated treatment effect size. RESULTS: Sixteen (76%) of the 21 reports provided a description of sample size calculations, and 12 of these gave some reasons for the choice of the anticipated treatment effect size. Expressed as percent change, the median treatment effect from intervention anticipated by the investigators was 50% (range, 15% to 90%), with a positively skewed distribution. The actual median percent reduction in adverse events from treatment (as compared with baseline) was 36% (range, 75% reduction to 5% excess). When the treatment effect was expressed as difference in adverse event rate, in the 14 (of 16) trials that could be analyzed, the median observed reduction in adverse events (death, bronchopulmonary dysplasia, or occurrence of respiratory distress syndrome) was 14.5% (range, 52% reduction to 2% excess). All trials except one concluded, however, that the intervention was effective, mostly based on additional subgroup calculations. CONCLUSIONS: Researchers often select sample sizes capable of detecting only large treatment effects, thus risking type II error, although sometimes a much smaller effect could be clinically important. While pragmatic considerations must be considered during the design of randomized clinical trials, researchers ought to present a rationale for anticipating a given magnitude of treatment effect in their sample size calculations. It may be possible to consider innovative trial designs that help determine the most appropriate treatment choice with the least possible sample size.

Confidence Intervals↗
PubMed1992

Sensitivity to change of the health assessment questionnaire (HAQ) and other clinical and health status measures in rheumatoid arthritis: results of short-term clinical trials and observational studies versus long-term observational studies.

To obtain evidence concerning short-term and long-term efficacy of clinical and health status measures in rheumatoid arthritis (RA), we conducted two observational studies--a 6-month study of 233 patients receiving methotrexate and a 10-year study of 157 patients receiving multiple treatments in a rheumatic disease clinic. Results of the 6-month study yielded effect sizes for treatment similar to the meta-analyses reported by Felson [corrected] et al. (Arthritis Rheum 33:1449-1461, 1990) and the controlled trials of methotrexate reported by Weinblatt et al. (Arthritis Rheum 33:330-338, 1990), suggesting that observational studies provide valid measurements of treatment effect. The effect size for the Health Assessment Questionnaire (HAQ) was 0.5. By contrast, the 10-year study suggested that standard clinical variables changed little and were not useful in assessing RA outcome, while the effect size of the HAQ was -2.39. These data continue to underscore the differences between short-term trials and the long-term outcome of RA, and suggest an important place for the HAQ or similar instruments in all phases of RA evaluation and assessment.

Arthritis, Rheumatoid↗
PubMed1992

Meta-analysis of diabetes patient education research: variations in intervention effects across studies.

Data from a previously reported meta-analysis of diabetes patient education literature were reanalyzed to determine the influence of study/subject characteristics, such as study quality and age of subjects, on patient outcomes. Patient knowledge and self-management skills, weight loss, glycosylated hemoglobin levels, and psychological outcomes were analyzed as outcome variables. Seventy-three relevant published and unpublished studies were located. Patient education appeared to be more effective in younger patients, particularly for the knowledge outcome. For all patients, glycosylated hemoglobin levels improved between 1 and 6 months postintervention, but decreased to 1-month levels after 6 months. Length of the educational intervention did not appear to influence outcomes. More rigorous, experimental research designs tended to produce more conservative effect size estimates. Implications for diabetes patient education are discussed.

Adult↗
PubMed1992

Adjustments to the Mantel-Haenszel test for data from stratified multistage surveys.

The usual form of the Mantel-Haenszel test statistic assumes independent observations. This is inappropriate for data from a stratified multistage survey. Two alternative adjustments to the test statistic are developed to deal with this: (a) a modification of the effective sample size for each row of each table, using the design effects, extending a method proposed by Donald and Donner for familial aggregation studies; (b) a Taylor series approximation to the variance of the square root of the numerator of the Mantel-Haenszel statistic. Both methods are evaluated by application to both simulated and real data. The two methods perform equally well, and offer a considerable improvement over the unadjusted test statistic when observations from the same cluster are highly correlated. A simplified adjustment is also considered, as is the need for correction to the variance of the odds ratio estimator.

Chi-Square Distribution↗
PubMed1992

Analysis of aberrations in public health surveillance data: estimating variances on correlated samples.

The detection of unusual patterns in health data presents an important challenge to health workers interested in early identification of epidemics or important risk factors. A useful procedure for detection of aberrations is the ratio of a current report to some historic baseline. This work addresses the problem of finding the variance of such a ratio when the surveillance reports are correlated. Results show that, when estimating this variance or the variance of the sample mean from a series of observations with an estimated correlation structure, bootstrap and jackknife estimates may be overly optimistic. The delta method or a classical method may be more useful when such model dependence is inappropriate.

Analysis of Variance↗
PubMed1992

Effects of mid-point imputation on the analysis of doubly censored data.

Doubly censored data arise in some cohort studies of the AIDS incubation period because the time of infection may be known only up to an interval defined by two successive screening tests for HIV antibody. A simple analytic approach is to impute the infection time by the mid-point of the interval and then apply standard survival techniques for right censored data. The objective of this paper is to investigate the statistical properties of such a mid-point imputation approach. We investigated the asymptotic bias of the Kaplan-Meier estimate, coverage probabilities of associated confidence intervals, bias in hazard ratio, and the size of the logrank test. We show that the statistical properties of mid-point imputation depend strongly on the underlying distributions of infection times and the incubation periods, and the width of the interval between screening tests. In the absence of treatment, the median incubation period of HIV infection is approximately 10 years, and we conclude that, for this situation, mid-point imputation is a reasonable procedure for interval widths of 2 years or less.

Bias↗
PubMed1990

Hospital stay length as an effect modifier of other risk factors for nosocomial infection.

This paper addresses the problem of hospital stay length as a risk factor for nosocomial infection and as a modifier of the effect of other risk factors for hospital infection. Patients were selected form two cross-sectional studies done in two different seasons of 1986. Risk of infection rose fairly steadily as hospital stay length increased (correlation coefficient: 0.83, p less than 0.01). Several risk factors (operation, underlying disease, and age) were analyzed on the basis of 1) raw data and 2) data stratified by length of stay. The results showed that hospital stay length is a strong modifier of the remaining risk factors, generally reducing, their effect on the development of hospital infection as length of stay increases.

Adolescent↗
PubMed1990

Intra and inter-person sources of variability in fat intake in a feeding trial of 14 men.

An impediment to analyzing the effect of nutritional factors on biologic processes or health status in human populations arises from the relatively small dietary differences that exist between individuals in relation to large periodic fluctuations in dietary intake and the imprecision with which diet is normally assessed. We report here on characteristics of dietary variability in a group of 14 young men who successfully completed an intervention study specifically designed to create large differences in fat intake between baseline and two dietary intervention-periods each lasting two months (during which safflower and coconut oil supplements were given). We found that in the second supplemental phase of the intervention inter-person sources of variability were greatly increased over the low-fat baseline values. For proportion of calories as fat it increased to 64.2% of total variance from 21.6% without supplementation; for saturated fatty acids, 47.3% from 17.7%; for polyunsaturated fatty acids, 62.4% from 22.8%; and for the P:S ratio, 71.5% from 21.6%. During the first intervention phase we observed only moderate changes. Reasons for the intervention phase differences in effect, implications for feeding trials designed to look at dietary fat effects, and the need for future studies aimed at clarifying these results are discussed.

Adult↗
PubMed1991

Theory-based screening for prevention: focusing on mediating processes in children of divorce.

Prevention programs in mental health theoretically can benefit from selecting participants who have a greater likelihood of developing psychological problems because of their exposure to the putative mediators targeted for change in an intervention. Screening on mediators may increase statistical power to detect program effects, enhance the cost-effectiveness of intervention trials, and decrease the possibility of iatrogenic effects. The circumstances that optimize the strategy of screening on the basis of mediating variables are discussed, and data are presented to illustrate the development of a mediational selection strategy to identify families who might best benefit from a preventive intervention for children of divorce. In addition, we present evidence that adjustment problems for children experiencing a divorce, as with most mental health problems, are not the result of one specific factor, but are jointly determined by several mediating processes that occur subsequent to the divorce. The mediational selection strategy developed illustrates the utility of measuring a set of mediational processes central to conferring risk for mental health problems to children of divorce.

Adolescent↗
PubMed1991

Planning the analysis of a breast cancer prevention trial.

The analysis of cancer prevention trials necessitates specialized procedures. Differences in the analysis plan between cancer prevention and therapy trials are emphasized. Discussion is focused upon (a) how trial conduct influences the analysis, (b) how noncompliance affects the analysis and the considerations this raises during the design and conduct of the trial, (c) how delays in the effect of the intervention impinges upon the analysis, and (d) statistical methods available to carry out the analysis. The analysis of a cancer prevention trial represents a final step in the evaluation of the study intervention and requires an adaptive planning procedure.

Breast Neoplasms↗
PubMed1989

Practical problems in interim analyses, with particular regard to estimation.

This article considers some of the practical problems inherent in interim analyses and stopping rules for randomized clinical trials. Topics covered include group sequential designs, trials with unplanned interim analyses, estimation problems in clinical trials with planned interim analyses, and the balance between individual and collective ethics. Particular attention is paid to the fact that clinical trials that stop early are prone to exaggerate the magnitude of treatment effect. Accordingly, a Bayesian "shrinkage" method of analysis is proposed to help quantify the extent to which surprisingly large point and interval estimates of treatment difference in clinical trials that stop early should be moderated.

Bayes Theorem↗
PubMed1991

Out of sight, out of mind--2. Social and historical contexts of self-harmful behaviour among aborigines of remote Australia.

Statements regarding the absence of suicidal and other self-harmful behaviour among Australian Aborigines are now untenable. The preceding paper presented the wider context of suicides in custody by examining suicide and other self-harmful behaviour in the Aboriginal population of the Kimbereley region of Western Australia. The emergence of suicide in that population was further located within an increase in deaths from non-natural causes over the last two decades. Following a brief history of the region, this paper develops an even wider context--suicidal behaviour among indigenous groups elsewhere. The relevance of these findings to the trends among Kimberely Aborigines is explored, leading to the central theme of the paper, the importance of an understanding of the socio-historical context in evaluating emerging patterns of harmful behaviour, and their relationship to alcohol use. Results obtained during a survey of the patterns and consequences of alcohol use from a stratified random sample of 516 Kimbereley Aborigines are presented. The author concludes by using the behaviours described in this and the preceding paper to examine the way in which Aboriginal 'problems' are constructed by the wider society, and 'solutions' defined.

Alcoholism↗
PubMed1992

Reporting the size of effects in research studies to facilitate assessment of practical or clinical significance.

"Statistically significant" results may lack practical or clinical significance, and the description of results in reports of research studies is frequently incomplete or misleading in presentation of data necessary to assess such significance (effect sizes). Various common presentations are discussed, and one method of presentation of results is proposed that not only encompasses all the others, but enhances the scope of useful information conveyed in a research report. The results of the Infant Health and Development Program (IHDP) are used to illustrate the method.

Child Behavior↗
PubMed1990

Sensitivity of effect variables in rheumatoid arthritis: a meta-analysis of 130 placebo controlled NSAID trials.

In a meta-analysis of placebo controlled NSAID trials, the sensitivity of the effect variables was calculated as the correlation coefficient and as the difference between drug and placebo, divided by the placebo group standard deviation. The patient's global evaluation was the most sensitive variable overall. Pain was more sensitive than Ritchie's index. Several variables may be omitted from clinical trials, especially if two active drugs are being compared. For example, the best maximum estimate for the difference in ESR between NSAIDs and placebo was 1.0 mm/hr (95% confidence interval -1.5 to 3.4 mm/hr), and for joint size 0.44% (-1.0 to 1.9%), corresponding to a quarter of a millimeter for each of the 10 joints usually measured. It is suggested to record only the patient's global evaluation, pain, and morning stiffness.

Anthropometry↗
PubMed1992

Cholesterol and coronary artery disease: age as an effect modifier.

An elevation of serum cholesterol has been one of the more frequently cited risk factors for coronary heart disease, found in both case-control and cohort studies. As a result, this country has undertaken massive screening of adults older than 20 years of age in an attempt to identify those persons with cholesterol levels greater than 200 mg/dl, and follow up with an active approach for intervention. The suggested cutpoints for borderline (200-240 mg/dl), and definite (> or = 240 mg/dl) hypercholesterolemia have been applied to all age groups despite suggestions of a diminution of risk conferred by cholesterol in the elderly. This study of 2544 white men undergoing coronary angiography shows that for all men, aged 25-84 years, plasma cholesterol levels were associated with an increase in coronary artery occlusion (rs = 0.15, p < 0.01). However, when stratified by age, this association held only for the younger men, the association diminishing to near zero in the oldest age group. The negative interaction between cholesterol levels and age in predicting coronary artery disease proved highly significant (p < 0.001) in multivariable linear regression analysis, suggesting that cholesterol levels are much less predictive of coronary artery disease in the elderly as compared to the young. These results point to the need for a more finely tuned set of criteria for the evaluation of hypercholesterolemia, one that takes into account the age of the screenee.

Adult↗