Meta-Analysis in Psychiatry Research

Fundamental and Advanced Methods

Mallikarjun B. Hanji

• Publisher: Apple Academic Press
• Year: 2017
• ISBN:
  9781315342009
• Edition: 1
• Topic: Matematica
• Subtopic: Probabilità e statistica

This book introduces the latest meta-analytical methods and discusses their applications in the field of psychiatry. A comprehensive list of methods used in meta-analysis has been described in simple language and demonstrated with real-time examples. This informative volume explains the importance of meta-analysis and describes how it differs from narrative and systematic reviews. It also relates the historical development of meta-analysis and explains methods used for locating and selecting the required studies in a given domain. Suitable software is examined in detail as well.

CHAPTER 1

INTRODUCTION

CONTENTS

Abstract

1.1 Features of Meta-Analysis

1.2 Scope and Benefits of Meta-Analysis

1.3 Some Examples

1.4 Historical Background

1.5 Limitations of Meta-Analysis

Keywords

References

ABSTRACT

Meta-analysis can be defined as a systematic statistical method for analyzing and synthesizing results from independent studies, taking into account all pertinent information. Readers of narrative studies face problems such as lack of detailed description, the process that led to the review, and hence the readers cannot replicate and verify the results and conclusions of the review. Most effective mechanism for systematic review is to reduce bias and increase precision, by including maximum possible number of relevant individual studies and providing a detailed description of their strengths and limitations. Vote counting is clearly unsound, since it ignores sample size, effect size, and research design. Meta-analysis is trying to answer four basic questions, namely, (1) are the results of the different studies similar and to the extent that they are similar, (2) what is the best overall estimate, (3) how precise and robust is the estimate, and (4) can dissimilarities be explained. Exploratory analysis, such as regarding subgroups of patients who are likely to respond particularly well to a treatment, may generate promising new research questions. Meta-analysis identifies areas where further studies are needed. Meta-analysis provides robust evidence and may utilize a less biased sample of evidence. Physicians can now make decisions regarding the use of therapies or diagnostic procedures on the basis of a single article that synthesizes the findings of tens or hundreds of clinical studies. The Cochrane Collaboration which is an international organization involved in preparing meta-analysis of the effects of interventions in all aspects of health care. The science of meta-analysis is relevant to clinical and community psychiatry to evaluate the potential errors and sources of bias and offer guidelines for evaluation. The statistical basis of meta-analysis reached back to the 17th century wherein astronomy and geodesy intuition and experience suggested that combinations of data might be better than attempts to choose amongst them. Meta-analysis has had critics and criticisms over the years. Most prominent of which is publication bias, which refers to the tendency for journals and authors not to publish articles on research that has no significant findings. There is a danger that meta-analysis of observational data produce very precise but spurious results. The complex methods used in meta-analysis should always be complemented by clinical acumen and common sense in designing the protocol of a systematic review, deciding what data can be combined, and determining whether data should be combined. Meta-analysis provides an opportunity for shared subjectivity in reviews rather than true objectivity. Meta-analyses are most easily performed with the assistance of computer databases and statistical software.

1.1 FEATURES OF META-ANALYSIS

1.1.1 META-ANALYSIS

Meta-analysis can be defined as a systematic statistical method for analyzing and synthesizing results from independent studies, taking into account all pertinent information. By synthesizing, scrutinizing, tabulating, and perhaps integrating all relevant studies, meta-analysis allows a more objective appraisal, which can help to resolve uncertainties when the original research, classical reviews, and editorial comments disagree. Meta-analysis is a scientific activity that borrows from both the expert review and the methodology of multicenter studies (Fisher et al., 1993). There are varieties of synonyms for meta-analysis used in the literature: overviews, aggregates, syntheses, integration, amalgamation, pooling, and combining. Quantitative is the heart of the meta-analysis and combining results is an essential ingredient in meta-analysis.

1.1.2 NARRATIVE STUDIES

Traditionally, individuals often considered experts in the field who have conducted narrative reviews of the literature, associated with a particular field using informal and subjective methods to collect and interpret information. Readers of narrative studies face problems such as lack of detailed description, the process that led to the review, and hence the readers cannot replicate and verify the results and conclusions of the review.

1.1.3 SYSTEMATIC REVIEWS

Reviews being the product of a scientific process to reduce bias, to increase precision and by providing detailed information to allow replication by others. Most effective mechanism for systematic review is to reduce bias and increase precision, by including maximum possible number of relevant individual studies and providing a detailed description of their strengths and limitations.

1.1.4 VOTE COUNTING METHODS

Once a set of studies have been assembled, a common way to review the results is to count the number of studies reporting various sides of an issue and to choose the view receiving the most votes. This procedure is clearly unsound, since it ignores sample size, effect size, and research design.

1.2 SCOPE AND BENEFITS OF META-ANALYSIS

1.2.1 COMBINE RESULTS

A quantitative systematic review or meta-analysis use statistical methods to combine the results of multiple studies.

1.2.2 HETEROGENEITY

They are trying to answer four basic questions, namely: (1) Are the results of the different studies similar and to the extent that they are similar? (2) What is the best overall estimate? (3) How precise and robust is the estimate? and (4) Can dissimilarities be explained (Lau et al., 1997)?

1.2.3 EXPLORATORY ANALYSIS

Exploratory analysis, such as regarding subgroups of patients who are likely to respond particularly well to a treatment, may generate promising new research questions to be addressed in future studies. Meta-analysis can help us to investigate the relationship between study features and study outcomes. One can code the study features according to the objectives of the review and transform the study outcomes to a common metric so that comparison of the outcome is possible.

1.2.4 IDENTIFICATION OF RESEARCH AREAS

Meta-analysis may demonstrate the level of adequate evidence and this identifies areas where further studies are needed.

1.2.5 PROVIDING EVIDENCE

Meta-analysis can examine questions, provide formal standard of rigorous for accumulating evidence from different studies, formulize the process of policy making, increase statistical power, provide robust evidence, and may utilize a less biased sample of evidence.

Meta-analysis, if appropriate, will enhance the precision of estimates of treatment effects, leading to reduced probability of false negative results, and potentially timely introduction of effective treatments.

1.2.6 BENEFITS OF META-ANALYSIS

Physicians can now make decisions regarding the use of therapies or diagnostic procedures on the basis of a single article that synthesizes the findings of tens or hundreds of clinical studies. Scientists in every field can similarly gain a coherent view of the central reality behind the multifarious and often discordant findings of research in their areas. Meta-analysis of a series of small clinical trials of a new therapy often yields a finding on the basis of which physicians can confidently begin using it without waiting long years for a massive trial to be conducted.

1.3 SOME EXAMPLES

Sharma et al. (2003) has successfully employed meta-analytical procedures to determine the effect of inhaled steroids on bone mineral density. Shann (1997) has employed meta-analysis to obtain evidence of trials of prophylactic antibiotics for children with measles for adequate evidence. The meta-analysis (Gupta and Gupta, 1996; Gupta, 1997) was performed to determine the time trend in the prevalence of coronary heart diseases in India and age and gender specific changes.

The Cochrane Collaboration which is an international organization involved in preparing maintaining and disseminating highly structured, frequently updated, and good quality systematic reviews and meta-analysis of the effects of interventions in all aspects of health care (Cochrane Injuries Group Albumin Reviewer, 1998; Kennedy et al., 2002; Olsen and Gotzsche, 2001).

The national library of medicine defines meta-analysis as a quantitative method of combining the results of independent studies and synthesizing summaries and conclusions, which may be used to evaluate therapeutic effectiveness, plan new studies, etc. with application chiefly in the areas of research and medicine.

Meta-analyses are based on trials of parallel group design, but some trials assessing the treatment of interest may use other designs. This is particularly the case in certain chronic diseases whose treatment is often evaluated by cross over-trials; typical examples include hypertension, asthma, or rheumatic diseases. Parallel and cross-over trials both provide estimates of the same treatment effect (Curtin et al., 2002a,b).

Laird and Ware (1982) have discussed the random effects model for longitudinal data on health effects of air pollution. Malhotra et al. (2001) have conducted a meta-analysis of controlled clinical trials comprising low-molecular-weight heparins with unfractionated heparin in unstable angina. Pavia et al. (2003) have carried out a meta-analysis of residential exposure to radon gas and lung cancer. Ezzat et al. (2004) have carried out a systematic review on the prevalence of pituitary adenomas. Gisbert et al. (2003) have carried out a systematic review and meta-analysis to determine prevalence of hepatitis C virus infection in porphyria cutaneatarda. Devereaux et al. (2002) have carried out meta-analysis of studies comprising mortality rates of private for-profit and private for nonprofit hospitals.

1.3.1 PSYCHIATRIC RESEARCH

The science of meta-analysis is relevant to clinical and community psychiatry to evaluate the potential errors and sources of bias and offer guidelines for evaluation. Meta-analysis is a specific technique that was developed in social sciences, but was soon adapted as a fundamental tool in psychiatric research with a number of aims.

The relevance of meta-analysis to psychiatry stems from one of the earliest meta-analyses ever un...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• About the Author
• Table of Contents
• List of Abbreviations
• Preface
• Acknowledgments
• 1. Introduction
• 2. Protocol Writing for Meta-Analysis Study
• 3. Formulation of Research Problem
• 4. Location and Selection of Studies
• 5. Quality Assessment of Selected Studies
• 6. Effect Sizes of Primary Studies
• 7. Preparation of Meta-Analysis Master Sheet
• 8. Meta-Analysis Plots
• 9. Meta-Analysis of Two Studies
• 10. Methods for Pooling Estimates: Fixed Effects Model
• 11. Method for Pooling Estimates: Random Effects Model
• 12. Individual Patient Data Meta-Analysis
• 13. Meta-Analysis of Observational Studies
• 14. Additional Meta-Analysis Techniques
• 15. Reporting Meta-Analysis Results
• 16. Implications of Results of Meta-Analysis
• 17. Meta-Analysis Software
• 18. Running Meta-Analysis Using Stata
• Appendix I: Numerical Demonstration: Meta-Analytical Approach to Estimate Prevalence of Schizophrenia in India
• Index