Interpretation and Uses of Medical Statistics (eBook)

Leslie Daly is the author of Interpretation and Uses of Medical Statistics, 5th Edition, published by Wiley. Geoffrey J Bourke is the author of Interpretation and Uses of Medical Statistics, 5th Edition, published by Wiley.

Interpretation and Uses of Medical Statistics 5
Contents 7
Preface 11
Structure of the Book 14
1 Describing Data —A Single Variable 17
1.1 Introduction 17
1.2 Types of data 17
1.3 Qualitative data —simple tables and bar charts 21
1.4 Picturing quantitative data 23
1.5 Shapes of distributions 35
1.6 Measures of central value 36
1.7 Other measures of location —quantiles 47
1.8 Measures of dispersion 53
1.9 Summary 57
2 Probability, Populations and Samples 59
2.1 Introduction 59
2.2 Probability 59
2.3 Populations and samples 64
2.4 Sample surveys 68
2.5 Summary 72
3 Associations: Chance, Confounded or Causal? 73
3.1 Introduction 73
3.2 Examining associations 73
3.3 Interpreting associations 81
3.4 Associations due to chance 81
3.5 Associations due to bias or confounding 93
3.6 Causal associations 97
3.7 Summary 101
4 Confidence Intervals: General Principles Proportions, Means, Medians, Counts and Rates
4.1 Introduction 102
4.2 The normal distribution 103
4.3 Sampling variation — proportions 107
4.4 Confidence intervals for a proportion 110
4.5 Sampling variation — means 115
4.6 Confidence intervals for a mean 117
4.7 Confidence intervals for a geometric mean [!] 123
4.8 Confidence intervals for a median [!] 125
4.9 Confidence intervals for a count or rate [!] 127
4.10 Summary 130
5 Hypothesis Testing: General Principles and One-sample Tests for Means, Proportions, Counts and Rates 132
5.1 Introduction 132
5.2 The null and alternative hypotheses 133
5.3 The significance test 134
5.4 Relationship with confidence intervals 138
5.5 One-sided and two-sided tests 139
5.6 General structure of a significance test: the one sample z test for a mean 141
5.7 Non-significant results and power: type I errors, type II errors and sample size 145
5.8 The one-sample t test for a mean 151
5.9 The one-sample z test for a proportion 152
5.10 The one-sample x2 test for many proportions [!] 152
5.11 The one-sample z test for counts or rates 155
5.12 Small sample sizes and the validity of assumptions [!] 156
5.13 Summary 158
6 Epidemiological and Clinical Research Methods 159
6.1 Introduction 159
6.2 Observational and experimental studies 160
6.3 The cross-sectional study or survey 161
6.4 The cohort study 162
6.5 Measures of association in cohort studies 165
6.6 Risk with variable-time follow-up [!] 174
6.7 The case–control study 177
6.8 Measures of association in case–control studies – the odds ratio 181
6.9 The analysis of cohort and case–control studies [!] 184
6.10 Comparisons of cohort and case–control studies 185
6.11 The randomized controlled trial 189
6.12 Applicability versus validity of trial results 202
6.13 Alternative trial designs 208
6.14 Ethical considerations for trials 211
6.15 Summary 216
7 Confidence Intervals and Hypothesis Tests: Two-group Comparisons 218
7.1 Introduction 219
7.2 Independent and paired comparisons 219
7.3 Parametric and non-parametric significance tests 221
7.4 Comparison of two independent means 223
7.5 Inferring significance from confidence intervals 228
7.6 Comparison of two independent geometric means [!] 231
7.7 Comparison of two independent medians 234
7.8 Comparison of paired means 237
7.9 Comparison of paired medians 239
7.10 Comparison of two independent proportions or risks (2 X 2 tables) 241
7.11 Comparison of two independent sets of proportions 255
7.12 Comparison of paired proportions 258
7.13 Comparison of two counts 263
7.14 Comparison of two rates 264
7.15 Comparison of two life tables [!] 268
7.16 Significance testing versus confidence intervals 278
7.17 Summary 280
8 Sample Size Determination 285
8.1 Introduction 285
8.2 Factors affecting the approach to sample size estimation 286
8.3 Sample sizes for single-group studies 288
8.4 Sample size specifications for two-group comparisons 292
8.5 Derivation of a formula for two-group comparisons [!] 295
8.6 Sample sizes for the comparison of two independent groups 297
8.7 Sample sizes for the comparison of paired means 306
8.8 Confidence intervals and sample sizes for comparative studies 308
8.9 Summary 311
9 Comparison of More than Two Independent Groups [!] 312
9.1 Introduction 312
9.2 Comparison of independent means in more than two groups: one-way analysis of variance 313
9.3 Comparison of independent proportions in more than two groups: the x2 test 325
9.4 Comparison of other statistics in more than two groups 329
9.5 Summary 329
10 Associations between Two Quantitative Variables: Regression and Correlation 331
10.1 Introduction 331
10.2 Describing relationships between two quantitative variables 332
10.3 Correlation 338
10.4 Calculation of regression and correlation coefficients [!] 341
10.5 Statistical inference in regression and correlation 343
10.6 When the assumptions do not hold [!] 348
10.7 Biases and misinterpretations in regression 351
10.8 Summary 354
11 Multivariate Analysis and the Control of Confounding[!] 355
11.1 Introduction 355
11.2 Quantitative dependent variable 356
11.3 Qualitative dependent variable: (i) case–control and cohort studies 364
11.4 Confounding versus effect modification or interaction [!] 377
11.5 Qualitative dependent variable: (ii) vital statistics approaches 380
11.6 Multivariate analysis 395
11.7 Summary 396
12 Bias and Measurement Error 397
12.1 Introduction 397
12.2 Biases in study design 398
12.3 Biases in selecting the sample 398
12.4 Accuracy: measurement error for a single test 400
12.5 Agreement: comparing two different tests or observers 418
12.6 Validity: comparing a test with a gold standard 427
12.7 Bias in statistical analysis and interpretation 431
12.8 Critical reading of the literature 433
12.9 A note on research procedures 435
12.10 Summary 437
Appendix A: Computational Shortcuts 438
A.1 Introduction 438
A.2 The standard deviation 438
A.3 The x2 test for independent 2 X 2 tables 440
A.4 Regression and correlation 442
A.5 Sums of squares in ANOVA 442
Appendix B: Statistical Tables 446
B.1 Introduction 446
B.2 Tables 447
Appendix C: A ‘Cookbook’ for Hypothesis Tests and Confidence Intervals 536
C.1 Introduction 536
C.2 The z test for a single proportion. Confidence intervals for a proportion 539
C.3 The z and t tests for a single mean. Confidence intervals for a mean or geometric mean 541
C.4 Confidence intervals for a single median 542
C.5 The z test for a single count or rate. Confidence intervals for a count or rate 543
C.6 The x2 test for many proportions (in one group) 544
C.7 The t test for two independent means. Confidence intervals for the difference between two means and for a ratio of geometric means 545
C.8 The Wilcoxon rank sum test for two independent medians 547
C.9 The t test for paired means. Confidence intervals for a mean difference 548
C.10 The sign test for paired medians 549
C.11 The Wilcoxon signed rank test for paired medians 550
C.12 The z test for two independent proportions. Confidence intervals for a difference between proportions 551
C.13 The x2 test for a 2 X 2 table. Confidence intervals for a relative risk (ratio of proportions) and for an odds ratio 552
C.14 Fisher’s test for a 2 X 2 table 554
C.15 The x2 test for a general I X J table 555
C.16 The McNemar and exact tests for paired proportions. Confidence intervals for a difference between proportions and for an odds ratio 556
C.17 The z test for two independent counts or rates. Confidence intervals for a difference between counts or rates, and for a ratio of counts or rates 558
C.18 One-way analysis of variance for means in more than two groups. Con.dence intervals for means and mean differences 560
C.19 The x2 test for a trend in proportions in a 2 X k table 562
C.20 Tests for regression and correlation 563
C.21 Spearman’s rank order correlation coefficient 564
Appendix D: World Medical Association Declaration of Helsinki 566
Bibliography and References 570
Index 575

"Medical Uses of Statistics, 3rd Edition" presents the concepts of
medical statistics across a broad range of topics with a practical
perspective, a moderate level of detail, and a minimal number of
formulae . . . The text is clearly written in a consistent style,
as illustrated by these excerpts." (Journal of Clinical Research
Best Practices, 8 August 2011)

...the book can be warmly recommended as a textbook and
reference for undergraduate medical students and health
professionals." Statistical Methods in Medical Research (on
the fourth edition)

"It should prove useful for both undergraduate medical and
paramedical students, and for postgraduate researchers and
clinicians who wish to have a greater understanding of medical
statistics or to analyse their own data." Modern Medicine of
Australia (on the fourth edition)

"...the book can be warmly recommended as a textbook and reference
for undergraduate medical students and health professionals."

Statistical Methods in Medical Research (on the fourth edition)

"It should prove useful for both undergraduate medical and
paramedical students, and for postgraduate researchers and
clinicians who wish to have a greater understanding of medical
statistics or to analyse their own data."

Modern Medicine of Australia (on the fourth edition)