Critical Appraisal of Epidemiological Studies and Clinical Trials

by Elwood, J. Mark

• ISBN13: 9780192627452
• ISBN10: 0192627457
• Edition: 2nd
• Format: Hardcover
• Copyright: 1998-02-26
• Publisher: Oxford University Press

Summary

Critical appraisal is now the central focus of evidence-based medicine andknowledge-based healthcare, and is currently a major influence in mostcountries. This second edition has been thoroughly revised and new sections onmeta-analysis, evidence based medicine and misclassification have been added.This book presents a system of critical appraisal to help the reader evaluatestudies in the health and medical fields.

Table of Contents

Preface and acknowledgements	v
Introduction	1	(2)
1. The importance of causal relationships in medicine and health care	3	(11)
What is causation?	4	(3)
A direct test of causation	7	(2)
Methods of counting events	9	(3)
Summary	12	(1)
Self-test questions	12	(2)
2. Study designs which can demonstrate and test causation	14	(23)
Relationship of study design to the definition of causation	14	(1)
A classification of comparative studies: classification by the groups of people compared	15	(3)
Classification by time relationships	18	(2)
More complex designs	20	(2)
Characteristics of the different study designs: intervention trials	22	(5)
Observational cohort studies	27	(2)
Case-control studies	29	(3)
Cross-sectional surveys	32	(1)
Summary of comparisons	33	(1)
Summary	33	(3)
Self-test questions	36	(1)
3. The results obtained from studies of causation	37	(18)
Cohort and intervention studies	37	(2)
Relative risk and relative odds	39	(1)
Risk difference (attributable risk)	39	(2)
Preventive factors, and application to intervention studies; number needed to treat	41	(1)
Person-time as the denominator of rates	42	(1)
More than one causal agent	43	(2)
Case-control studies	45	(2)
Use of the odds ratio	47	(2)
Different sampling schemes in case-control studies: why a case can also be a control	49	(2)
Surveys	51	(1)
General applicability of these analytical approaches	51	(1)
Summary	52	(1)
Self-test questions	53	(2)
4. Selection of subjects for study	55	(39)
Target, source, eligible, and participant populations	55	(4)
The distinction between internal and external validity	59	(2)
Selection bias and its effects	61	(2)
Methods of reducing selection problems	63	(3)
Selection of subjects for comparative studies	66	(1)
Selection of the comparison subjects	67	(1)
Selection of subjects in randomized intervention trials	68	(7)
Selection of the subjects for a cohort study	75	(7)
Selection of subjects for a case-control study	82	(7)
Assessment of selection issues in a completed study	89	(2)
Summary	91	(2)
Self-test questions	93	(1)
5. Error and bias in observations	94	(22)
Sources of error and of bias	94	(6)
Practical issues in reducing bias and error	100	(1)
Examples of bias	101	(2)
Assessment of bias and error	103	(1)
Assessment of the consistency of information	104	(5)
Effects of misclassification on study results	109	(4)
Summary	113	(1)
Self-test questions	114	(2)
6. Confounding	116	(45)
Confounding--definition and non-mathematical examples	116	(4)
Confounding in cohort and intervention studies	120	(4)
Confounding in case-control studies	124	(2)
Methods for the control of confounding	126	(1)
Control of confounding: restriction	126	(1)
Randomization	127	(8)
Stratification	135	(10)
Matching	145	(5)
Multivariate methods	150	(5)
Application of the definition of confounding to study design	155	(1)
Application of the definition of confounding to analysis	156	(1)
Summary	157	(1)
Self-test questions	158	(3)
7. Chance variation	161	(37)
The concept of significance	162	(1)
Simple statistical tests for a two-by-two table	163	(4)
The Mantel-Haenszel test	167	(5)
The concept of precision; confidence limits	172	(5)
Statistical power and the size of a study	177	(6)
Repeated testing in trials: stopping rules and data monitoring groups	183	(2)
Ordered exposure variables and tests of trend	185	(1)
Assumptions in stratification: interaction	186	(3)
Statistical tests in matched analysis	189	(1)
Statistical tests in multivariate analysis	190	(3)
Issues in the interpretation of results	193	(1)
Summary	194	(2)
Self-test questions	196	(2)
8. Combining results from several studies: meta-analysis	198	(20)
The process of meta-analysis	199	(1)
Publication bias	200	(2)
Finding and selecting studies for inclusion	202	(3)
Analyses based on original data, or on published results	205	(1)
Limitations of meta-analysis	206	(1)
Exploring differences between studies	206	(1)
Statistical methods in meta-analysis	207	(8)
The Cochrane collaboration	215	(1)
Summary	215	(1)
Self-test questions	216	(2)
9. The diagnosis of causation	218	(27)
Questions 1 to 5: what evidence do we have?	218	(1)
A. Description of the evidence	219	(2)
B. Internal validity--consideration of non-causal explanations	221	(1)
C. Internal validity--consideration of positive features of causation	222	(6)
D. External validity--generalization of the results	228	(1)
E. Comparison of the results with other evidence	229	(7)
Assessing causal relationships and making decisions	236	(1)
Further application of these concepts	237	(2)
Application of the scheme in designing a study	239	(1)
The application of critical appraisal to health care decisions: evidence based medicine	239	(4)
Summary	243	(1)
Self-test questions	244	(1)
Introduction to Chapters 10-15	245	(2)
10. Critical appraisal of a randomized clinical trial	247	(16)
11. Critical appraisal of a randomized trial of a preventive agent	263	(20)
12. Critical appraisal of a prospective cohort study	283	(17)
13. Critical appraisal of a retrospective cohort study	300	(17)
14. Critical appraisal of a matched case-control study	317	(14)
15. Critical appraisal of a large, population based, case-control study	331	(17)
Answers to self-test questions	348	(11)
References	359	(14)
Appendices	373	(70)
Appendix 1: Statistical analysis methods: formulae and worked examples: list of contents	373	(46)
Appendix 2: Lifetable methods: list of contents	419	(11)
Appendix 3: Statistical tables: list of tables	430	(13)
Index	443