Statistical Thinking for Non-Statisticians in Drug Regulation (2nd ed, 2014)

+ b9 d$ K, U3 [+ W0 L

  q/ W, q5 i& K* `  Q( H! ~. i
& }' H. C5 R" H  z% ^; O% r5 VTitle: Statistical Thinking for Non-Statisticians in Drug Regulation
Author(s): Richard Kay
8 @5 Q. m/ H9 L1 yPublisher: Wiley
Year: 2014       
/ t& c7 B" ~6 J6 O( {9 C; SEdition: 2
- ~) w0 {0 p  K4 X: eLanguage: English       
" D3 u% n3 `* a$ t4 k# H8 ?Pages: 368
ISBN: 111847094X, 9781118470947
Size: 2 MB (2224739 bytes)       
Extension: pdf


" U* F$ x- J) ~: u3 Z+ V, ]6 bDescription
" K$ v/ I1 |, F3 Y$ d$ U
Statistical Thinking for Non-Statisticians in Drug Regulation, Second Edition, is a need-to-know guide to understanding statistical methodology, statistical data and results within drug development and clinical trials.
0 Z. w% }8 ~" M4 q  n) a$ c
  |" O/ J$ q  w+ v4 c/ HIt provides non-statisticians working in the pharmaceutical and medical device industries with an accessible introduction to the knowledge they need when working with statistical information and communicating with statisticians. It covers the statistical aspects of design, conduct, analysis and presentation of data from clinical trials in drug regulation and improves the ability to read, understand and critically appraise statistical methodology in papers and reports. As such, it is directly concerned with the day-to-day practice and the regulatory requirements of drug development and clinical trials.
( }9 n5 D! U  k8 N+ }) ~
& B, ?4 R" t2 G" yFully conversant with current regulatory requirements, this second edition includes five new chapters covering Bayesian statistics, adaptive designs, observational studies, methods for safety analysis and monitoring and statistics for diagnosis.
8 G. ?; C2 o% {. |! g( |
- l, m+ G4 y' A# W/ {Authored by a respected lecturer and consultant to the pharmaceutical industry, Statistical Thinking for Non-Statisticians in Drug Regulation is an ideal guide for physicians, clinical research scientists, managers and associates, data managers, medical writers, regulatory personnel and for all non-statisticians working and learning within the pharmaceutical industry.


- X; Q, K# s( [+ K) s+ zTable of Contents
( Y; g! v. S6 `
Preface to the second edition xv
8 m5 c  C) B# E8 Z0 ~
Preface to the first edition xvii

+ `6 g, Z& e. V: y$ W+ b% ~Abbreviations xxi

1 Basic ideas in clinical trial design 1
3 t( m2 y+ X1 T" H+ t
1.1 Historical perspective 1
/ l+ ~/ j# o' ^) n( |
0 K* ?% ~% ?* W1.2 Control groups 2
/ B9 C2 h: Y2 `2 s5 @
1.3 Placebos and blinding 3
+ M  K4 G& A7 U% l6 `0 I, v
1.4 Randomisation 3

1.5 Bias and precision 9

, r9 ~9 z% x/ x( l2 }1 x1.6 Between- and within-patient designs 11

! f: s; M2 [8 R) t2 V1 k5 ?1.7 Crossover trials 12

1.8 Signal noise and evidence 13
8 Z  Y2 t( n5 X( g
1.9 Confirmatory and exploratory trials 15

1.10 Superiority equivalence and non-inferiority trials 16

. k; Q# @' ^* t- o1.11 Data and endpoint types 17

1.12 Choice of endpoint 18

2 Sampling and inferential statistics 23
% Y% P: W/ E: l6 R5 B4 I/ S
4 }, m, h/ E# V. z( Q1 j/ q2.1 Sample and population 23

; g+ x/ H) s( d' s1 I: I: l2.2 Sample statistics and population parameters 24
0 [( I& M5 h/ k& q) Z
2.3 The normal distribution 28
- W" h8 B) }. P/ A, }5 _
2.4 Sampling and the standard error of the mean 31

  y- Q" W8 F) l2.5 Standard errors more generally 34
" A# b& U2 R9 N
3 Confidence intervals and p-values 38

$ U" C! i3 t! ^% R  C3.1 Confidence intervals for a single mean 38
$ D* r. b) B- b" m) ^! {
3.2 Confidence interval for other parameters 42

3.3 Hypothesis testing 45

4 Tests for simple treatment comparisons 56
$ L3 T* f2 k2 X
4.1 The unpaired t-test 56

9 p& i' I& j$ ]/ `: b6 d5 D4.2 The paired t-test 57

7 `$ _/ I9 M# y/ L" A5 ]# s' o. o7 X4.3 Interpreting the t-tests 60

, V5 j0 L% `" J! d' Y' p  [, B4.4 The chi-square test for binary data 61
, f( g4 V! h+ O, H( i# ~2 w7 k/ s" Z
4.5 Measures of treatment benefit 64

' \( B5 k; a9 J! }7 V% Y4.6 Fisher’s exact test 69

! `, u# I% ?& x. S, L% L; o4.7 Tests for categorical and ordinal data 71

4.8 Extensions for multiple treatment groups 75

7 J# W- y. J7 ?" g9 Z0 G" G2 c# R5 Adjusting the analysis 78
' u1 ?5 {- p, L5 o) G9 \4 T7 F4 ]
  F8 d& m$ D5 }( B. m: z2 o5.1 Objectives for adjusted analysis 78

5.2 Comparing treatments for continuous data 78

5.3 Least squares means 82
4 I( ~" I" H2 z
/ m4 `5 }7 X% o3 s* U5.4 Evaluating the homogeneity of the treatment effect 83
) ^* g  O) s3 {( Z1 z4 F4 l; p& a
5.5 Methods for binary categorical and ordinal data 86

5.6 Multi-centre trials 87

( R$ r& \: B! Z) o6 Regression and analysis of covariance 89
: k7 p6 f: H+ z$ ~, b- L: {; e
6.1 Adjusting for baseline factors 89
5 u9 A8 X+ u7 ]* u
6.2 Simple linear regression 89

. @! \& E- n& H- i, T3 `6.3 Multiple regression 91

6.4 Logistic regression 94
! x; f( Z1 }# S" K8 A  q0 m6 n4 b
6.5 Analysis of covariance for continuous data 94
7 Q0 d8 c0 w- W/ a
4 w5 Y& s# d) H# l8 ~1 f" ?6.6 Binary categorical and ordinal data 101

% d8 s) G1 }' E4 u" Q6.7 Regulatory aspects of the use of covariates 103

6.8 Baseline testing 105

1 S* i) d% |# X7 l0 j% h; e& M3 g* G' h7 Intention-to-treat and analysis sets 107

! l0 }: l4 [3 A7 k7.1 The principle of intention-to-treat 107

8 {" H+ q5 F7 D& J  [% A" E7.2 The practice of intention-to-treat 110

! |: |, K% e  u6 q7.3 Missing data 113
) W* |; ]; n, {2 I, V2 h3 h
7.4 Intention-to-treat and time-to-event data 118

7.5 General questions and considerations 120
, n' V5 V  h3 d# B
, V! N& I3 j1 t0 q# f' O8 Power and sample size 123

8.1 Type I and type II errors 123
1 x6 @: o" j* R1 P7 V; }# a
1 @; Q, I6 z( {+ D4 J/ k8.2 Power 124

8.3 Calculating sample size 127

8.4 Impact of changing the parameters 130

8.5 Regulatory aspects 132
, b% j( ?5 C$ u9 ~8 ~
8.6 Reporting the sample size calculation 134
5 K9 \8 M: J( Y# W' w/ e! |  M
9 Statistical significance and clinical importance 136

) _. d; B# A, F$ ^5 U9.1 Link between p-values and Confidence intervals 136

9.2 Confidence intervals for clinical importance 137
& y  |5 ~8 D: J' _
9.3 Misinterpretation of the p-value 139

5 K# L2 H# p. R( I5 K+ V% |  u, |9.4 Single pivotal trial and 0.05 140
1 K: f  x  Q1 c5 b( n
10 Multiple testing 142

" U! C! L1 r0 c% a10.1 Inflation of the type I error 142

) Y( y" }2 ~) ?7 h& w- B10.2 How does multiplicity arise? 143
5 {* p0 S3 \( G- G
10.3 Regulatory view 144

7 N# r3 r5 \/ R, z10.4 Multiple primary endpoints 145
) \" v0 H5 W; N0 }
10.5 Methods for adjustment 149
6 o' ^& E) |9 M2 d. a. r& R
; ]* ?; W) S' k6 a) K10.6 Multiple comparisons 152

7 }; [; C* H2 `, d10.7 Repeated evaluation over time 153

10.8 Subgroup testing 154

10.9 Other areas for multiplicity 156
4 B+ Z5 R, B; @0 }* N: K6 g
11 Non-parametric and related methods 158
0 x7 B5 x; E) D7 `9 r7 M
) V; ?  k. E# Q  W% F11.1 Assumptions underlying the t-tests and their extensions 158
1 K7 U) B. _6 l; [* N
11.2 Homogeneity of variance 158
/ Q' ?/ z/ r/ E% _* e& C" E
11.3 The assumption of normality 159

11.4 Non-normality and transformations 161

$ ^7 q& w" b3 p' \) A  Y1 R( q% _11.5 Non-parametric tests 164
' }' S4 C/ V# E& G! v8 u7 g
& R2 l6 r' t- b" e11.6 Advantages and disadvantages of non-parametric methods 168

11.7 Outliers 169

12 Equivalence and non-inferiority 170

12.1 Demonstrating similarity 170
# i. V. y. j% Y! q: G- D7 Q+ F0 b
12.2 Confidence intervals for equivalence 172
2 L- p* \! @4 f# }# S
  ]' j9 r) d8 ]  F# G$ j' H12.3 Confidence intervals for non-inferiority 173

' \' S! X! \& f+ U  a12.4 A p-value approach 174

3 R) y; ]# R+ Q1 b: s& W, A8 D, q; c12.5 Assay sensitivity 176
$ z3 N- P. P* C( ?$ ]& c
: c* @' {5 x& ]7 Y12.6 Analysis sets 178

12.7 The choice of Δ 179

8 |* _( g: b9 r$ |12.8 Biocreep and constancy 184

12.9 Sample size calculations 184

12.10 Switching between non-inferiority and superiority 186
, |0 [' b6 a. ^
13 The analysis of survival data 189

13.1 Time-to-event data and censoring 189
$ U* A! o8 Z" u8 a% a) Q+ f# i
13.2 Kaplan-Meier curves 190
% g- a+ v( C5 g+ c
8 T4 w$ b( G, _8 X7 W0 I13.3 Treatment comparisons 193
0 d1 Y1 G0 W% N, c* V$ y
; g* N* a: L+ I' f13.4 The hazard ratio 196

13.5 Adjusted analyses 199
7 O9 Y8 ^; c) _0 L6 B/ W
13.6 Independent censoring 202
/ _1 j* _4 j  Q- ^, m' {
. V: v/ L- T. R0 l: H1 H; O13.7 Sample size calculations 203
9 g! \. l7 J, _' [8 r
2 Y' ^% `3 E& c" L6 ~# t14 Interim analysis and data monitoring committees 205

# d& w, T. n# C) D4 `$ ^/ }0 }14.1 Stopping rules for interim analysis 205
) E0 o; D3 z4 z  K
/ P& U- ^+ |2 {; I  e8 s5 }14.2 Stopping for efficacy and futility 206

14.3 Monitoring safety 210
4 {& b1 \' y! ]$ e3 r
' P1 h; u1 f. g14.4 Data monitoring committees 211

1 E9 \' n" T; C3 T2 N! w+ H2 d7 s15 Bayesian statistics 215
! @. }* E( B' g
& X6 ]3 W  I/ V. l/ G9 n# l/ L15.1 Introduction 215

15.2 Prior and posterior distributions 215

! W9 {4 l7 W) ?15.3 Bayesian inference 219
0 ?' E; J- d7 p, b3 }+ l
  ]- Q0 |" J0 S5 O+ B15.4 Case study 221

15.5 History and regulatory acceptance 222
* v' H9 s( P" q4 a) Y! ^" n5 S4 ]
; k9 Z$ l  y/ w6 A: f15.6 Discussion 224

# E# L0 y/ p4 B16 Adaptive designs 225
1 `; x+ @% ]; E- e
16.1 What are adaptive designs? 225

1 w0 s5 k; B# r+ y; S+ t$ X" ?16.2 Minimising bias 228
/ Z6 Y6 A% `$ _0 h7 T. G
1 H: }: w6 G7 O% O16.3 Unblinded sample size re-estimation 232

16.4 Seamless phase II/III studies 234

16.5 Other types of adaptation 236

16.6 Further regulatory considerations 238
1 t! N& {, W* e6 b& h7 n
17 Observational studies 241
  P5 z% S# K! ?- Y2 s6 J- d
17.1 Introduction 241
  N# I7 H1 r7 P) H; a# O! F
! t3 Y8 T5 g; H/ M) L! c% _17.2 Guidance on design conduct and analysis 247
$ u+ N! }0 m/ r  Q" w* N! J' A
' g2 g' X7 Z* j17.3 Evaluating and adjusting for selection bias 249

17.4 Case–control studies 257
4 g( K# {# K; c3 }, j5 V- M
18 Meta-analysis 261

2 W' Q1 d* R. k$ ]# |18.1 Definition 261
) A$ n( U2 L; o
7 N& U! j& k8 h" \- ~) e18.2 Objectives 263
7 i: R& n" y! }8 U% \, H$ s' e
18.3 Statistical methodology 264
. T6 j" \3 D; ~0 L* C1 {! J
1 B% P3 O- L: Y" W7 h, G18.4 Case study 270

  F. a5 s( [! Q18.5 Ensuring scientific validity 271

5 H3 ?6 f7 }$ l& o  C18.6 Further regulatory aspects 275

8 R; e' K) r1 ?- K) q19 Methods for the safety analysis and safety monitoring 277

19.1 Introduction 277
1 |% Y3 N4 n4 E
19.2 Routine evaluation in clinical studies 279
" C# ~5 m, g8 |# S. B9 ?
19.3 Data monitoring committees 289

+ r- O! X7 x! t# N4 ~19.4 Assessing benefit–risk 290
* ?: |1 z3 l9 d' x/ h9 B- J
) _* F9 ?$ }# K1 o1 A  E7 K7 g19.5 Pharmacovigilance 299
: e1 I" h+ A+ M7 j
20 Diagnosis 304
( Y$ A/ b. w7 ?2 {2 |6 i& {
& R; n- ^$ X! R9 A& H, J20.1 Introduction 304
% {) l6 [5 m, x+ Y/ S0 ~
20.2 Measures of diagnostic performance 304

20.3 Receiver operating characteristic curves 308

% `+ i+ Z( e1 A! {& ]* x6 U/ m) S* G20.4 Diagnostic performance using regression models 310
' c! B: _! x2 r/ e" n1 ^
9 j6 O- Y9 ^& f( b20.5 Aspects of trial design for diagnostic agents 312

+ z9 W  K9 C1 M* }. J20.6 Assessing agreement 313

8 F" W& E4 C" C21 The role of statistics and statisticians 316
5 m2 s. ]2 w5 z0 d- I
21.1 The importance of statistical thinking at the design stage 316
- U) s+ }9 P2 B% k
5 z0 b2 X7 a/ j/ L! @3 R& D21.2 Regulatory guidelines 317

6 |) g* p2 W8 ^+ c! e. a21.3 The statistics process 321

2 k) b* i( Z2 f5 _! G; R21.4 The regulatory submission 327

  u& u2 F  F- ^+ N& p5 ]5 ]9 Y1 C21.5 Publications and presentations 328

References 331
6 a; m" }8 B# E1 o. z+ B1 k
Index 339

/ \% s2 M( j, m; ]7 W

Author Information

Richard Kay, Consultant in Statistics for the Pharmaceutical Industry, Great Longstone, Derbyshire, UK
. x) p& N$ e" S7 P; R2 c! Z1 C* Z
- n/ R5 _+ P/ V+ O. N( K+ d
% `% m3 p( w  n: T" }9 h  f
Statistical Thinking for Non-Statisticians in Drug Regulation (2nd ed, 2014).pdf (2.12 MB, 下载次数: 44)

2015-5-27 18:45 上传

点击文件名下载附件
阅读权限: 10

; W0 C0 s6 ?/ \* @5 c- J
* |, j5 ^; q# l% d0 O: d3 r
来自群组: RGCE