Statistical inference by confidence intervals: issues of interpretation and utilization.
(1/1807)
This article examines the role of the confidence interval (CI) in statistical inference and its advantages over conventional hypothesis testing, particularly when data are applied in the context of clinical practice. A CI provides a range of population values with which a sample statistic is consistent at a given level of confidence (usually 95%). Conventional hypothesis testing serves to either reject or retain a null hypothesis. A CI, while also functioning as a hypothesis test, provides additional information on the variability of an observed sample statistic (ie, its precision) and on its probable relationship to the value of this statistic in the population from which the sample was drawn (ie, its accuracy). Thus, the CI focuses attention on the magnitude and the probability of a treatment or other effect. It thereby assists in determining the clinical usefulness and importance of, as well as the statistical significance of, findings. The CI is appropriate for both parametric and nonparametric analyses and for both individual studies and aggregated data in meta-analyses. It is recommended that, when inferential statistical analysis is performed, CIs should accompany point estimates and conventional hypothesis tests wherever possible. (+info)
Variability in meta-analytic results concerning the value of cholesterol reduction in coronary heart disease: a meta-meta-analysis.
(2/1807)
Despite official support for the efficacy of cholesterol reduction, considerable controversy exists, and meta-analyses of this topic have produced conflicting results. The authors assessed the variability of meta-analyses, evaluating the cardiovascular value of cholesterol reduction while attempting to explain the variability. Metaanalyses were identified by electronic search and citation tracking. Included were those conducted prior to 1995 that dealt with cholesterol reduction and total mortality, cardiovascular mortality, or nonfatal cardiovascular disease. In addition to extracting odds ratios for total mortality, cardiovascular mortality, and nonfatal cardiovascular disease, the authors encoded methodological variables, publication variables, and data concerning investigators' backgrounds. Twenty-three meta-analyses were reviewed, and 15 concluded that cholesterol reduction was beneficial. Summary odds ratios for total mortality were heterogeneous, generally failing to support the value of cholesterol reduction. Odds ratios depended on inclusion criteria and investigator variables. Odds ratios for cardiovascular mortality and for nonfatal cardiovascular disease were more homogeneous and supported the value of cholesterol reduction. Methodologically better meta-analyses tended to report more beneficial odds ratios. Although "supportiveness" of the value of cholesterol reduction was associated with inclusion/exclusion criteria and publication variables, the primary outcome variable related to supportiveness was the statistical significance of the odds ratios for cardiovascular mortality. (+info)
Traditional reviews, meta-analyses and pooled analyses in epidemiology.
(3/1807)
BACKGROUND: The use of review articles and meta-analysis has become an important part of epidemiological research, mainly for reconciling previously conducted studies that have inconsistent results. Numerous methodologic issues particularly with respect to biases and the use of meta-analysis are still controversial. METHODS: Four methods summarizing data from epidemiological studies are described. The rationale for meta-analysis and the statistical methods used are outlined. The strengths and limitations of these methods are compared particularly with respect to their ability to investigate heterogeneity between studies and to provide quantitative risk estimation. RESULTS: Meta-analyses from published data are in general insufficient to calculate a pooled estimate since published estimates are based on heterogeneous populations, different study designs and mainly different statistical models. More reliable results can be expected if individual data are available for a pooled analysis, although some heterogeneity still remains. Large prospective planned meta-analysis of multicentre studies would be preferable to investigate small risk factors, however this type of meta-analysis is expensive and time-consuming. CONCLUSION: For a full assessment of risk factors with a high prevalence in the general population, pooling of data will become increasingly important. Future research needs to focus on the deficiencies of review methods, in particular, the errors and biases that can be produced when studies are combined that have used different designs, methods and analytic models. (+info)
Placenta previa: preponderance of male sex at birth.
(4/1807)
To determine the relation between placenta previa and male sex at birth, the authors conducted two types of analysis: 1) a historical cohort analysis of singleton live births in New Jersey hospitals during 1989-1992 (N = 447,963); and 2) a meta-analysis of previously published studies on the subject. For the cohort analysis, subject mother-infant dyads were identified from linked birth certificate and maternal and infant hospital claims data. The infant's sex for mothers with an International Classification of Diseases, Ninth Revision, Clinical Modification, code of 641.0-641.1 for placenta previa (n = 2,685) was compared with infant's sex for mothers without placenta previa (n = 445,270). For the meta-analysis, seven published articles were located and summary effects were calculated using both fixed-effect and random-effects models. In the present cohort study, the male:female ratio at birth was significantly higher in women with placenta previa (1.19) than in those without placenta previa (1.05) (p<0.001). The association of placenta previa with male sex persisted when the analysis was either stratified or adjusted for the effects of maternal age, maternal parity, maternal smoking during the index pregnancy, race/ethnicity, the infant's gestational age, and the infant's birth weight. The meta-analytic results from the fixed-effect and random-effects models showed a 14% excess of placenta previa when women were carrying a viable male fetus as compared with a viable female fetus during pregnancy. The results were the same regardless of whether the present cohort study was included in the meta-analysis. In conclusion, the evidence obtained from these analyses strongly argues for an association between placenta previa and male sex at birth. The mechanism for this association remains to be determined. (+info)
Glucocorticosteroids in the management of rheumatoid arthritis.
(5/1807)
Glucocorticosteroids are used frequently in the management of patients with rheumatoid arthritis. Data supporting their efficacy and safety are still meagre. Glucocorticosteroids may be used systemically with different routes of administration (oral, i.m. and i.v.), in different doses and for different periods of time. The effectiveness of glucocorticosteroids in reducing inflammation in the short term has been shown for oral treatment in a dose of 7.5 mg prednisolone daily or more, for i.m. pulses (120 mg methylprednisolone every 4 weeks) and for i.v. methylprednisolone pulses. For longer periods of treatment, the evidence suggesting effectiveness of low-dose oral glucocorticosteroids is more limited. Some data suggest that different regimens of glucocorticosteroids may retard the development of erosions in patients with rheumatoid arthritis. The toxicity of short-term treatment is relatively low. For long-term treatment, the development of osteoporosis is a serious problem. Concomitant therapy with either calcitriol or bisphosphonates may reduce this risk. (+info)
Evidence-based nephrology.
(6/1807)
Systematic reviews and meta-analyses are the best approaches available for summarizing the available evidence concerning the efficacy of therapies. Although the renal field has been slow to use these techniques, they are being used increasingly. In March 1997, the Cochrane Renal Group was formed, and this group aims to produce and maintain up to date systematic reviews of the evidence on the effectiveness of therapies used to treat patients with renal diseases. This group is part of the Cochrane Collaboration which is an international structure grouping collaborators together, with the aim of preparing, maintaining and disseminating systematic reviews of the effects of health care in all areas of medicine. (+info)
Evaluation of old and new tests of heterogeneity in epidemiologic meta-analysis.
(7/1807)
The identification of heterogeneity in effects between studies is a key issue in meta-analyses of observational studies, since it is critical for determining whether it is appropriate to pool the individual results into one summary measure. The result of a hypothesis test is often used as the decision criterion. In this paper, the authors use a large simulation study patterned from the key features of five published epidemiologic meta-analyses to investigate the type I error and statistical power of five previously proposed asymptotic homogeneity tests, a parametric bootstrap version of each of the tests, and tau2-bootstrap, a test proposed by the authors. The results show that the asymptotic DerSimonian and Laird Q statistic and the bootstrap versions of the other tests give the correct type I error under the null hypothesis but that all of the tests considered have low statistical power, especially when the number of studies included in the meta-analysis is small (<20). From the point of view of validity, power, and computational ease, the Q statistic is clearly the best choice. The authors found that the performance of all of the tests considered did not depend appreciably upon the value of the pooled odds ratio, both for size and for power. Because tests for heterogeneity will often be underpowered, random effects models can be used routinely, and heterogeneity can be quantified by means of R(I), the proportion of the total variance of the pooled effect measure due to between-study variance, and CV(B), the between-study coefficient of variation. (+info)
The role of vitamin D in corticosteroid-induced osteoporosis: a meta-analytic approach.
(8/1807)
OBJECTIVE: To determine if vitamin D is more effective than no therapy or calcium alone in the management of corticosteroid-induced osteoporosis, and to determine how vitamin D compares with other osteoporosis therapies, e.g., bisphosphonates, calcitonin, or fluoride, for this condition. METHODS: We evaluated all formulations of vitamin D, including its active metabolites and analogs. A systematic search for published and unpublished studies was conducted using MEDLINE (1966-December 1997), bibliographic references, abstracts from proceedings of recent national meetings, and contact with pharmaceutical companies and content experts. We included all randomized controlled trials lasting at least 6 months (and reporting extractable results), of patients receiving oral corticosteroids, that compared vitamin D with either 1) no therapy or calcium alone, or 2) bisphosphonates, calcitonin, or fluoride. The primary outcome measure of interest was change in lumbar spine bone mineral density. RESULTS: We found a moderate beneficial effect of vitamin D plus calcium versus no therapy or calcium alone (9 trials) (effect size 0.60; 95% confidence interval [95% CI] 0.34, 0.85; P < 0.0001). In comparisons of vitamin D with other osteoporosis therapies, bisphosphonates were more effective than vitamin D (6 trials) (effect size 0.57; 95% CI 0.09, 1.05). Calcitonin was similar in efficacy to vitamin D (4 trials) (effect size 0.03; 95% CI -0.39, 0.45). Fluoride was more effective than vitamin D, but there were only 2 trials. CONCLUSION: Vitamin D plus calcium is superior to no therapy or calcium alone in the management of corticosteroid-induced osteoporosis. Vitamin D is less effective than some osteoporosis therapies. Therefore, treatment with vitamin D plus calcium, as a minimum, should be recommended to patients receiving long-term corticosteroids. (+info)