Study Designs and Applied Statistics Methods in the Clinical Research

1
Cheng-Hsin General Hospital
Study Designs and Applied Statistics Methods in
the Clinical Research ????????? ???
???? 2008.2.26
2
Cheng-Hsin General Hospital
Part I Basic Study Designs
3
????????
Cheng-Hsin General Hospital

• ?????????????????
• ?????????????????

4
Cheng-Hsin General Hospital
???????

• ?????
• 1.??????????(Case-series or Descriptive)
• 2.????(Cross-sectional studies)
• 3.????(Cohort or Prospective studies)
• 4.??????(Case-control or Retrospective studies)
• 5.??????? (Historical cohort studies)

5
Cheng-Hsin General Hospital

• ?????(Experimental studies)
• 1. ?????(Controlled trials)
• A. Parallel or concurrent controls
• a. Randomized
• b. Not randomized
• B. Sequential controls
• a. Self-controlled
• b. Crossover
• C. External controls (including
  historical)
• 2.????
• ?????(Meta-analysis)

6
?????
Cheng-Hsin General Hospital

• ????
• 1. ???????????????
• 2. ????(control group)
• 3. ????? (research hypotheses)

7
Cheng-Hsin General Hospital
8
Cheng-Hsin General Hospital

• ????(Cross-sectional studies)

??
????
??
Time
????
No direction of inquiry
Question What is happening?
9
???????
Cheng-Hsin General Hospital

• ??
  
• 1.?????? 2.??????????
  
• ????
• ?????????????
• ??????
• 1.???? 2.?? 3.????? 4.
  ????
• ??
• 1. ???? 2. ???????? 3. ?????
• 4. ????????? 5. ????????
• ??
• 1.????????(causal relationship)
  
• 2.???????

10
Cheng-Hsin General Hospital
The Study Procedure
11
Cheng-Hsin General Hospital
12
Cheng-Hsin General Hospital

• ????(Cohort study)

??
??? (exposed)
????
??
Direction of inquiry
??
Question What will happen?
???? (unexposed)
??
Time
?????
13
???????
Cheng-Hsin General Hospital

• ??
• ???????????????(incidence)????
• ????
• 1.????????????
• 2.?????????????
• ??
• 1.??????
• 2.???????
• 3.???????????????
• 4.???????????
• 5.???????(multiple diseases)
• ??
• 1.?????? 2.????
• 3.????????????(bias)

14
Cheng-Hsin General Hospital

• ??????
• ?????(relative risk, RR)
• a/(ab)/ c/(cd)
• ?????(attributable risk, AR)
• a/(ab)-c/(cd)

?? ??
??? ???
a b ab c
d cd
ac bd abcd
15
Cheng-Hsin General Hospital
16
Cheng-Hsin General Hospital
17
Cheng-Hsin General Hospital

• ????(Outcome assessment)
• 1.????(Functional status)
• 2.????(Quality of life)
• 3.???(Patient satisfaction)
• 4.????(survival time)
• 5.????(Economic evaluation)
• A.Cost-effectiveness analysis CEA
• B.Cost-utility analysis CUA
• C.Cost-benefit analysis CBA

18
Cheng-Hsin General Hospital
19
Cheng-Hsin General Hospital
20
Cheng-Hsin General Hospital

• ??????(Case-control studies)

??(exposed)
???(cases)
???(unexposed)
??(exposed)
???(controls)
???(unexposed)
Time
??????
Direction of inquiry
Question What happened?
21
??????????
Cheng-Hsin General Hospital

• ??
  
• ??????????????????
• 
  
• ????
  
• ???????????, ???????, ??????
• ???????????
• ??????????
• 1.?????????????
• 2.?????????????

22
Cheng-Hsin General Hospital

• ??
• 1.????
  
• 2.???????
• 3.??????????
• 4.???????
• 5.????????????????
• ??
• 1.?????????????
• 2.??????????
• 3.?????????????
• 4.???????
• 5.?????????(selection bias)?
• ?????(recall bias)

23
Cheng-Hsin General Hospital
?? ???
??? a b ab ???
c d cd
ac bd
1.????case?control??exposed?unexposed???
(ratio),???a/b?c/d???exposure?disease???? 2.?case
?????control?? (a/bgtc/d),??????? ???? 3.case
??exposure oddsa/b control ??exposure
oddsc/d ?odds ratio(OR)(a/b)/(c/d)ad/bc
??OR???,????????????.
24
Cheng-Hsin General Hospital
25
Cheng-Hsin General Hospital

• ???????(Historical cohort studies)

??? (exposed)
??
????
??
??
Direction of inquiry
??? (unexposed)
??
Time
??????
26
??????????
Cheng-Hsin General Hospital

• ?? ?? ??
  ??
• ????
  Subjects selected,
• 
  data gathered on
• 
  exposure status and
• 
  outcome
• ?????? Information Case
  chosen
• obtained on
  with outcome
• past
  exposure Controls chosen
• 
  without outcome
• ????
  Subjects selected outcome
• 
  and classified as
  measured
• 
  to exposure
• ???????
  Subjects identified
• Subjects
  who were exposed
• classified as
  to in the past
• exposure from

27
?????(Experimental studies)
Cheng-Hsin General Hospital

• Experimental studies that involve humans are
  called clinical trials
• Controlled trials are studies in which the
  experimental drug or procedure is compared with
  another drug or procedure, sometimes a placebo
  and sometimes the previously accept treatment
• Uncontrolled trials are studies in which the
  investigators experience with the experimental
  drug or procedure is described, but the treatment
  is not compared with another treatment

28
Cheng-Hsin General Hospital

• ?????????

???
???
????
???
???
???
???
Time
??????
??(intervention)
29
Cheng-Hsin General Hospital

• ???????
• 1.????????? ???????????,???? ??????????
• 2.????????????????????????
• ????????????(placebo)

30
Cheng-Hsin General Hospital

• ????(Random Allocation)
• 1.???????????,??random allocation???,
• ?????????
• 2.RandomizationRandom Allocation
• ?????????????????????
• 3.????????????????????????
• ?????(comparability),???????(self-
• selection)???, ????????????(validity)
• 4.??????
• (1)??????(simple randomization)
• (2)??????(Block randomization)
• (3)??????(Stratified randomization)

31
Cheng-Hsin General Hospital

• ????( Blinding procedure)
• 1.????????,????????????bias,????bias,
• ???????double blind
• 2.????(single blind)?????,?????????????
• ????(double blind)??????????????????
• ????(triple blind)??????????????????
• 3.?????????????
• (1)???(side effect)
• (2)??trial?home VS hospital
• (3)??????patient??????,?????????

32
Cheng-Hsin General Hospital

• Randomized clinical trials
• Provided the strongest evidence for
  concluding causation
• Nonrandomized trials
• Do nothing to prevent bias in patient
  assignment
• Trials with self-controls
• 1.Only one group in which patients are
  assessed before
• and after the intervention
• 2.Hawthorne effect

33
Cheng-Hsin General Hospital

• ????(Crossover study)

with outcome
experimental subjects
experimental subjects
with outcome
subjects meeting entry criteria
without outcome
without outcome
with outcome
with outcome
controls
without outcome
controls
without outcome
Time
washout period
intervention
onset of study
intervention
34
Cheng-Hsin General Hospital

• Trials with external controls

subjects
with outcome
without outcome
with outcome
Results from previous study
without outcome
Time
onset of study
intervention in subjects only
35
Cheng-Hsin General Hospital
36
????? (Quasi-experimental study)
Cheng-Hsin General Hospital
37
?????(Meta-analysis)
Cheng-Hsin General Hospital

• Meta-analysis uses published information from
  other studies and combines the results so as to
  permit an overall conclusion
• Meta-analysis is similar to review articles, but
  additionally
• includes a quantitative assessment and
  summary of the
• findings
• Meta-analysis is especially appropriate when the
  studies that have been reported have small
  numbers of subjects or come to different
  conclusions

38
??????
Cheng-Hsin General Hospital

• How to evaluating the study design the medical
• literatures?
• 1.In a clinical trial
• (1) How are subjects recruited?
• (2) Are subjects randomly assigned to the
  study group?
• (3) Is there a control group?
• (4) Are appropriate therapies included?
• (5) Is the study blind? Double blind?
• (6) How is compliance evaluated?
• (7) If some cases are censored, is a
  survival method
• such as Kaplan-Meier or the Cox model
  used?

39
Cheng-Hsin General Hospital

• 2. In a cohort study
• (1) How are subjects recruited?
• (2) Are subjects randomly selected from an
  eligible pool?
• (3) How rigorously are subjects followed? How
  many
• dropouts does the study have and who are
  they?
• (4) If some cases are censored, is a survival
  method such
• as Kaplan-Meier or the Cox model used?
• 3. In a case-control study
• (1) Are subjects randomly selected from an
  eligible pool?
• (2) Is the control group a good one?
• (3) Are records reviewed independently by more
  than one
• person?

40
Cheng-Hsin General Hospital

• 4. In a cross-sectional study
• (1) Are the questions unbiased?
• (2) Are subjects randomly selected from an
  eligible?
• (3) What is the response rate?
• 5. In a meta-analysis
• (1) How is the literature search conducted?
• (2) Are the criteria for inclusion and
  exclusion of studies
• clearly stated?
• (3) Is an effort made to reduce publication
  bias?
• (4) Is there information on how many studies
  are need to
• change the conclusion?

41
Cheng-Hsin General Hospital
Part II Basic Statistics Methods
42
Types of clinical data analysis
Cheng-Hsin General Hospital

• Time-unrelated analysis
• Time-related analysis

43
Time-unrelated analysis
Cheng-Hsin General Hospital

• Categorical scale
• 1. Proportion Z test ?2 test
• 2. Logistic regression
• Interval scale
• 1. Independent T-test (Two-sample)
• 2. ANOVA
• 3. Linear Regression

44
Time-related analysis
Cheng-Hsin General Hospital

• Balanced repeated measurement
• 1.Categorical scale
• (1) Generalized Estimation Equation
  (GEE)
• (2) Mixed model
• 2.Interval scale
• (1) Pair T-test
• (2) Mixed model
• Survival analysis

45
Survival analysis
Cheng-Hsin General Hospital

• Non-parametric method
• ?Kaplan-Meier (Life-Table) method
• Semi-parametric method
• ? Cox regression model
• Parametric method

46
Two-Sample Problems
Cheng-Hsin General Hospital

• Continuous variable
• 1.Independent T-test
• 2.Paired T- test
• Binary variable
• 1.Proportion Z Test or Chi-square test
• 2.McNemars Test

47
Independent T-test Paired T-test
Cheng-Hsin General Hospital

• Example 1 Evaluation of effectiveness of
• community rehabilitation
  care (n30)
• Pre
  Post
• 83
  77
• 95
  80
• 75
  70
• 62 59

48
Statistical evaluation
Cheng-Hsin General Hospital

• Mean
• Pre
  Post
• 78.20
  70.26
• (SD14.31) (SD
  18.04)
• Paired T T-3.09 P0.0079
• Independent T T-1.33 P0.19

49
Correlation Repeated measurement
Cheng-Hsin General Hospital

• Paired T test enhance statistical
• efficiency by considering ...... the
• correlation between Pre- and Post-
• (the same individual)

50
?2 test Proportion Z test
Cheng-Hsin General Hospital

• Example 2
• The study on the assessment of mental
• health program
• Intervention Experimental Control
• Outcome Favorable (F) Unfavorable (U)

51
Two Data Sets
Cheng-Hsin General Hospital

• 1. Data 1
• F
  U
• Test 40 20
  
• Control 16 48
• 2. Data 2
• F
  U
• Test 10 2
• Control 2 4

52
Statistical evaluation
Cheng-Hsin General Hospital

• Data 1
• ?2 (O-E)2/E ? ?2(1) 21.79 P0.001
• Proportion test DiffP1-P2
• 
  0.667-0.250.417
• (95CI
  0.26-0.58)
• The intervention group has a more significant
• favorable response than the control group

53
Fisher Exact Test
Cheng-Hsin General Hospital

• Data 2
• ?2(1) 4.50 P0.034 (right??)
• Fishers Exact Test
• One-Tail? P0.057
• Two-Tail? P0.107

54
Correlated Data(McNemars Test)
Cheng-Hsin General Hospital

• Example 3
• A match study on the efficacy of attending
  CRC screening with respect to health promotion
  (matched by sex and age)
• Outcome whether to attend CRC screening

55
General ?2 test
Cheng-Hsin General Hospital
Health Promotion

• -
• 43 33
• Attend
• - 16 26
• ?2(1)3.70 P0.055 (right??)

56
McNemars test
Cheng-Hsin General Hospital

• Health Promotion
• Yes
• -
• 27 16 43
• No
• - 6 10 16
• 33 26 59
• ?2(1) 4.54 P0.033

57
Measuring agreement between two people
Cheng-Hsin General Hospital

• Example 4
• Two reviewers on mammography
• Second
• Yes No
• Yes 20 5
• First
• No 3 200

58
Kappa statistics
Cheng-Hsin General Hospital

• Kappa (Observed-Expected agreement)
• /(1-expected agreement)
• (O-E)/(1-E)
• Kappa0.70 (95 CI 0.50-0.89)

59
How to interpret kappa?
Cheng-Hsin General Hospital

• Byrt (1996)
• 0.93-1.00 ? Excellent agreement
• 0.81-0.92 ? Very good agreement
• 0.61-0.80 ? Good agreement
• 0.41-0.60 ? Fair agreement
• 0.21-0.40 ? Slight agreement
• 0.01-0.20 ? Poor agreement
• ?0.00 ? No agreement

60
Mutivariate analysis
Cheng-Hsin General Hospital

• Continuous variable
• 1. Linear regression
• 2. Mixed model (repeated measurement)
• Binary variable
• 1. Logistic regression
• 2. GEE model (repeated measurement)

61
Linear regression
Cheng-Hsin General Hospital

• Example 5
• Is the effect of diet and exercise program on
  the SBP effective?
• Independent variable diet, exercise
• Outcome SBP

62
Multiple Linear Regression
Cheng-Hsin General Hospital

• Y ab1dietb2exerb3dietexer ?
• (Interaction
  term)
• bs SE T
  P-value
• diet 1.03 2.22 0.47
  0.64
• exer 0.78 0.13 6.21
  0.0001
• dietexer 0.30 0.20 1.54
  0.13

63
Logistic regression
Cheng-Hsin General Hospital

• Example 6
• A study was performed on 53 patients
• with prostate cancer to collect data on
• several variables considered predictive of
• nodal involvement

64
Definitions of predictors
Cheng-Hsin General Hospital

• Age
• Acid level of serum acid phosphate in
• King-Armstrong units (0.46-1.26)
• X-ray examination (0negative 1positive)
• Grade Pathological grade
• (0less serious 1more serious)
• Size size of tumor (0small, 1large)
• Nodalinv Laprectomy results
• (0No involvement1Involveme
  nt)

65
Model selection
Cheng-Hsin General Hospital

• X-ray, Size and Acid are
• three significant factors for
• predicting nodal involvement

66
Model Prediction
Cheng-Hsin General Hospital
Odds Ratio (OR)exp(?)
67
Probability prediction of nodal involvement
(risk classification)
Cheng-Hsin General Hospital

• High-risk
• Acid 1.26 , X-ray with positive finding,
• and large tumor 0.96
• Low-risk
• Acid0.48 , X-ray with negative finding,
• and small tumor0.05

68
Generalized Estimation Equation (GEE method)
Cheng-Hsin General Hospital

• Indications The regression model for
• correlated binary outcome
• Example 7
• A clinical trial comparing two treatments
• for a respiratory disease

69
Definitions of variables
Cheng-Hsin General Hospital

• Outcomes Respiratory status
• (0poor 1good)?four visits
• The main variable Treatment
• (1Active
  0placebo)
• Other explanatory variables
• center, age, sex, and
  baseline
• respiratory status

70
Characteristics of GEE model
Cheng-Hsin General Hospital

• Working correlation matrix
• to accommodate correlation
• between outcomes

71
Survival Analysis
Cheng-Hsin General Hospital

• Outcome variable survival time
• Starting points?endpoints Survival time
• Censoring problem
• i.e Right censoring study ends or lost to
• follow-up

72
Kaplan-Meier method
Cheng-Hsin General Hospital

• Example 8
• Prognosis for women with breast cancer
• A histochemical marker called Helix pomatia
  agglutinin (HPA) is used to assess whether tumor
  have already been metastasized

73
Definition of Variables
Cheng-Hsin General Hospital

• 32 Patients
• Survival time from treatment to death
• Censoring 1dead 0censoring
• Prognostic factor HPA (1positive
• 
  0negative)

74
Cox regression model
Cheng-Hsin General Hospital

• Proportional hazard model
• One of common multiple regression
• models for survival analysis.
• No baseline hazard estimation
• h(t,x1,x2,,xn)h0(t)exp(b1x1b2x2bnxn)

75
Importance of the Cox Model
Cheng-Hsin General Hospital

• Provided the only valid method of predicting a
  time-dependent outcome
• Producing survival curves that are adjusted for
  confounding factors.
• The Cox model can be extended to the case of
  multiple events for a subject
• Cox model could estimate RR?exp(?)

76
Parametric regression modelsAccelerated failure
time (AFT) model
Cheng-Hsin General Hospital

• To accommodates left censoring and interval
  censoring
• Distribution
• 1.exponential distribution
• 2.weibull distribution
• 3.log-normal distribution
• 4.log-logistic distribution
• 5.Gamma distribution

77
Other methods for multiple variables
Cheng-Hsin General Hospital

• Discriminant analysis
• It assumes that the independent variables
  follow a multivariate normal distribution, so it
  must be used with caution if some X variables are
  nominal
• Log-Linear analysis
• All the variables, both independent and
  dependent, are measured on a nominal scale
• Factor analysis
• All variables are considered to be
  independent variables

78
Cheng-Hsin General Hospital

• Cluster analysis
• The object is to determine a classification
  or taxonomic
• scheme that accounts for variance among the
  subjects
• Multivariate analysis of variance (MANOVA)
• Involved multiple dependent variables as well
  as multiple
• independent variables
• Canonical correlation analysis
• When both the independent variables and the
  outcomes
• are numerical and the research question
  focuses on the
• relationship between the set of independent
  dependent
• variables

79
Cheng-Hsin General Hospital
Summary of conceptual framework for
questions involving two variables
Independent variable
Dependent variable
Method
nominal nominal
chi-square
binary numerical
t-test
nominal numerical
one-way ANOVA (more than two values)
numerical numerical
regression correlation
80
Cheng-Hsin General Hospital
Summary of conceptual framework for questions
involving two or more independent variables
Independent variable
Dependent variable
Method
nominal nominal
Log-linear
nominal and numerical dichotomous
Logistic regression
nominal and numerical nominal
Discriminant analysis
(two or more
values)
nominal numerical
ANOVA
numerical numerical
Multiple regression
nominal and numerical censored
Cox regression
nominal with confounding numerical
ANCOVA
nominal
Mantel-Haenszel
numerical only ---
Factor analysis Cluster analysis
81
Cheng-Hsin General Hospital
Thank you for your attention!