Professional Documents
Culture Documents
Basic Clinical Study Designs
Basic Clinical Study Designs
Basic Clinical Study Designs
!
(Basic clinical study designs)
.
(Drug Information Service) 01 2
345
36
7685
9 :
7;
70
2<0 76345 2= >2:?76
@2AB4
85
H@4
J46
H
7
:08
@K H
46
H;B24= <04= J
B4
:0 ;B24= <00L <?46 J54
4H M
J
46 H 22 34
K 3@M
!
;!5<6 !
J7601 2 K B4
1. 345 6 !
! (retrospective study) 01
507
@K 2
B40
B4:646 2H=
@K 7=
?
B48;P
2?3@MH
4?
?
502 1
@K
507M85 H4 H;0R> 2
A34524 H=
A345 46 <6
J54 ? ?B24 <65
6 4
345 46 2 44J H 850S9?
85HH44J 3454?34850S 9?
J 850S 0RHH:2=
@K G@24 H0R>
B24 ?B244B24 H
H= 34854= J
@K 7A3454 =
@K 2<6
JA345
B48;P
7<0
3 <? 46 6 4
@K 4
<6;@0
H
(adverse effect) 24 H?3@M
4
?2= (rare incidence)
B44 H?3@M764
? 8 (long term
adverse effect) 01
Past
Present
Time period
1
(retrospective study)
2. 345 6 !>
(prospective study) 01
@K 2
A3452H2H
?3@M4 ?
502 2
@K
507M34? B4
A345<03 H
J447
A3459?P2
J0V44 (bias) 2
J?3@M<?
0V44 6
00X(blinding)
:6(randomization) 01 7
A345<03 H= A
345<?
J 6
A34574 34= H?=
@K
507MB4 0R>
B24
<6
J? 850S
J:
<?(loss to follow-up) 7 6 H 6 5
@K 46
@K 7A345<03 6
@K ? 0
P874
<6;@0
M
0
randomized controlled trial
Present
Future
Time period
2
%&
(prospective study)
!
;!5< ?@
76
507
@K M;H
K34
:
@K 3485H
@K 2
85H3 <0:=
@K H?01
@K ?4 (experimental study) 76
@K 285H<6<?
3 <0:= 2;76 7H?@3456 MH?01
@K (observational
study) ?
502 3
'())
(clinical study)
-)1
2
(observational study)
-)45
(descriptive study)
- cross-sectional study
- longtitudinal study
- retrospective study
-)'.
(experimental study)
@K
(analytical study)
- cross-sectional study
- cohort study
- case-control study
randomization
A 3 K
@K
non-randomization
K
B4
:60
2=
@K
@K
507
@K
J76<? 2 0
B4
1. 345D<
(descriptive study)
@K H ;
6>6H01
: (prevalence) 74:
(incidence) 349
0RHH 0
6 N 201K2 234
?9
6 4 : ; BM4
8
HH= 4
5034
;
9?<6
0
345
6 :60
2
@K 7664<?01 3 K
@K H:? ? @2 (cross-sectional study)
M
01 case report
B401
@K 2 H= 850S;?
B4 case-series B4
@K 2H = 850S
6 1 3@M<0 H7?345?J;
201
64B2434:6?@201
@2 N
(longtitudinal study)
46
?54
?9
7
G434850S2 019
(incidence)
@K ;
9?A3454
46
@K :34850S2019
<3B4?444:
1
9
;
0
K
60a 2545
2. 345
E (analytical study)
@K
J:0
;B24
0
8
@K
6 :66 N G@276
4401 3 K ?M
Cross-sectional study
01
@K ;P
346 6 N H:?@234
B4
2 = ?
B401
@K ?2
2H 0
:6>6;:6? H:?@234
B4
2 = ?
@K ?M =
@K 2
H _9
7
@K BM40
?A4B2N 6
@K ;P
6 pressure sore risk score
?78?
Cohort study
@K 7 cohort study 01
@K
(Analytical study) 201
@K 6
8;P
(outcome) ;P
0RHH2H@K (exposure)
B4<6
@K 7 cohort HfV
0
285H<?
?B4< 9?:60
2HM01:6 0
2 <6?9
B4
:
285HH (event-free) H M? ?5 6 :60
?:
285H H
B4<6
9?2<0
= cohort J:0
;B24
1. 3 ?340R>
B44:
?9
(incidence)
B4 4:
(mortality)
B4 2 (risk)
2. ;P
6
2
B40
P89?H0
6 4
B4
0
4:
34
?8;P
34:6:6@2 2H 6
@K H :6850SB4
Framingham
C Massachusetts ;B24 4:
?9
BM4H (The Framingham
Heart Study)
A 4
@K 7 Cohort ?6 N
345
J cohort study
@K H 7 cohort 4 H7644<?01 2 7 B4
1. Prospective cohort
Prospective cohort
B4 concurrent cohort J@ 852=
@K H
2A3452
2
Dyspepsia and ulcers
NSAIDs
No dyspepsia and ulcers
<6
NSAIDs
No dyspepsia and ulcers
Present
Future
Time period
5
@K 7 prospective cohort
2. Retrospective cohort
Retrospective cohort J@ 852=
@K HA3454?H 76345 6 7@
7;
B44 H<?345H 0
01 2
0HR H2 (exposure)
B4<6
0RHH2 (not exposure) 34:60
2@K 4?642850SH?9
B48;P
7?8;P
2?3@M
B4<66 J? HJ@0RHH: 34?34
@K 7M B4 H6?34H= ??
502 6
2
Osteoporotic Fracture
<?
warfarin
NO osteoporotic Fracture
Osteoporotic Fracture
<6<?
warfarin
NO osteoporotic Fracture
Past
Present
Time period
A 6
@K 7 retrospective cohort
<
345 cohort
@K 7 cohort 01
@K 9?
2H :602 <6019
? H
B4<6
0RHH2 ;B24
?9
6 H@
J44:
34
?9
:627 <60RHH
2<? 4H M
J0
4:
349
6 :620HR H294 ?9
5012
6 34:62<60HR H2 6 2<?H016 ;P
2
6 2;P
(relative risk: RR) G@2
<?H 4
?9
6 :62<?
0RHH2:6<2 6<?
0RHH2 0
9?
= H 82<?
2x2 ?
2 1
2 1
2x2
@K 7 cohort
Exposure
Outcome/Disease
4:
?9
a+b
a/(a+b)
c+d
c/(c+d)
?9
<6?9
:620RHH
:62<60RHH
4:
349
:6 20RHH2
H= 2019
:620HR H2
H= 2<?
0RHH2M ?
4:
349
:6 2<60 RHH2
a / (a+b)
H= 2019
:620HR H2
H= 2<?
0RHH2M ?
2;P (RR)
c / (c+d)
4:
349
:6 20RHH2
4:
349
:62<60RHH2
a / (a+b)
c / (c+d)
46
@K 7 cohort ;B24 ;P
6
5:
27
?9
4?B4?H
(coronary heart disease: CHD)
Exposure
Outcome/Disease
?9
CHD
<6?9
CHD
4:
?9
:65:
2
84
2916
3000
84/3000
:6<65:
2
87
4913
5000
87/5000
H 46
J 2 ;P
(RR)
84/3000
87/5000
1.6
34?4
1. 0R> selectiion bias H
2<6
randomization G@268?0RHH
2. 0R>
5> 34<0H cohort
3. <6 =
@K 9
2;<?4
;
43 ?46 2 ;B242H
H= 85 019
;;42H=
J<?
4. 4 H4
? outcome <?
285H
6
B4<60RHH2
5.
@K 6 H 6 5
Case-Control Study
@K 7 case-control 01
@K 74 (retrospective study) 01
@K 2:6
;P
34
8
B4<?
0RHH
?8;P
9?
2
@K 2 8;P
(9
B4
4
2H)
850S:6 M 6 case 0
:62<68;P
(9
B44
2H) G@2
6
control 7=
?6 :6 76:6 <?
8
B4<?
0RHH (exposure) 4? 64
B4<6 7= = ?63494 34852?9
7<?
exposure 70
94
34852<6?9
76<?
exposure 6 9?
:0B4
@K 7 case-control H:6 0
?A;B24 6
0RHH2
H@K (exposure) 01 : (cause) 34
?8;P
(outcome)
B4<6
@K 7 case-control =
@K 9
24:
2= (rare disease)
B401
9
2;<?4
; (latency period) 69
A B24H
@K 7MH01PB4
46 H 2?9
7
A 7 K
@K 7 case-control
@K 7 case-control MH01
@K 7?4 (retrospective study) G@2
2
H
?B40
46 28;P
7<68;P
74<0@K 6
<?
8
22H@K 6
B4<6
M
@K 7 case-control
J= <?
2345J5A
@7<03 (prospective study) :6 (cohort) 285H? 345<03 B24
AHM
? 345 85H
J=
@K 29
(case) 72<69
(control) 7
=
@K 6 4
<?
28 (exposure) M76
6 4:6
B4<6
@K K
M
6 y nested case-control study z
Cohort of
Norway
(CONOR)
NSAIDs
oral cancer
(case)
No oral cancer
(control)
No NSAIDs
(exposure)
Past
(outcome)
Present
Future
A 8 K
@K 7 nested case-control
<
345 case-control
@K 7MH
0
:6 case :6 control ;B24@K 6 0RHH276
Exposure
Outcome
case
Control
:620RHH
:62<60RHH
a+c
b+d
@K 7 case-control 0
46 2@K J5:6 H 2019
7<6019
?MH@<64 H 4:
34
?9
<? ;
<6
J4<? 6 2 exposed H= 6 M H
?9
3@M2
;76 6 H= 019
6 M2 exposed 2 6 B4
<6
J
2;P
(relative risk: RR) <?9?
46 <
A
@K 7 case-control H
J
4J@ 2<?6 9?0
:62 0RHH2:62<60RHH2 82<?
6
odds ratio (OR) = 6 odds J@ 94 2:
MH?3@M64?94 2:
MH<6
?3@M = 6 odds ratio J@ 4
6 (ratio) 34 odds
34 case-control study 01?M
Odds ratio (OR)
(a/a+c)
(c/a+c)
ad/bc
(b/b+d)
(d/b+d)
a/c
b/d
4
B49
2@K 4
?2= (rare event) 6 OR H RR
46
@K ;P
34
:
? Endometrial cancer > >20:S
Exposure
Outcome
? Endometrial cancer
:62:
77
64
:62<6:
42
11
119
75
=
OR = 0.32 6
77 x 11
42 x 64
0.32
B4
JA-A345 case-control
34?
1.
J20R> ? H
P
34?4
1.
B4:6:2 = <?
;
<6
J:70
:6
@K 6 :6:<?:
2. 2H0X?85H 2J ;3485
34501 case
B4 control H@4 H?4<?
2.
J=
@K <?9
2;4/<6
64= 2;P
<?9?
7
latent period
3. 0
? <?8
A B240
3. 8
@K 3@M456 5
343452
H
?<03
B4
H
=
@<4?
?4
4.
H47<3345= <?
B4<6
J= <?
345 J! (experimental study)
507
@K ?4 (experimental study) HHH?
2H@K G@2 JB4
6 01 intervention 76850S
= ?
<?
intervention 2
?M
-
7 randomization
7<6 randomization (
B4 quasi-experimental study)
@K ?401
@K 2
?8<03 (prospective study) 9?=
0
4
? event
B48;P
H
<?
intervention
6 :6@K G@2 <?
intervention :6
control 2<6<?
intervention 9?2<0
@K 0
MH
447
@K
?: ;B24?4
6 N24 H?3@MH : ?M8
@K ?4H@ 6 B24JB4
4J@0
P8
34 6<??6
@K 9?_; 46 2
@K 2
:6 (randomization) 34
intervention 76850S
Randomized controlled trials (RCT) 01
507
@K ?42 6 B24JB4 2:?
@K H
2 H 850S
B4853
6
@K 2 <6 event G@2018;P
H
2H :6
46 2 K }
?B4 (inclusion criteria 7 exclusion criteria) HJ5:6
(randomization) ;B24<?
intervention ?6 N (4 H 6 2 :6) 6
0
6
placebo
B4
C 2
K 9
MN
@K 7 randomized controlled trials 764401 2
50764N B4
1. 345 randomized controlled trials
J cross-over
@K
507M 853
6
@K :
H4<?
intervention 6 1 ?
F
&GH%
random
IJ'KL 1 N A
IJ'KL 2 N B
Evaluation 1
Wash out peroid
IJ'KL 1 N B
IJ'KL 2 N A
Evaluation 2
9 randomized controlled trials -). cross-over
IJ&S-'KL
&T
H-).'KL 2
(.
U2
5VW(.
U
IJ2HJ'KLFJ(.
U
random
H)
(SW
&K
'K F
502 10
@K 7 randomized controlled trials ? cross-over
!Y;D
Z
[
70
J@ :K
B4:3422
B4<6
J?<? <?
B47H7H
<? 7:K
B4:6 M0270<? 6 70
?
6 <?76M= 5 4 :
?95
?M= B4? 670
? HH <?76
5@
5@? ;@
;4H642 6 N 70
? 4B2N 6
?
@K C 01 G@2 85H 2 ?
70
6 M44 016
B436 70
6 M 76 76:776:6
B4<6
7= K276 M
6 76 446 <
;P
46 <
J;
9?2<0
J766? (measurement scales) 3470
6 N 4401 2 ? B4
1. Categorical data (
B4 discrete data)
6 3470
?MH01767:6 7?34 subject 6 ; G@26 201<0<? B4
7>6 M <6
J 6 2456
6 7> 70
?M
J766444 H=
:6<?01
- ? dichotomous : 70
2<?; 2 6 6 M 6 ; ; 7;>
- ? polychotomous : 70
2 6 2 6 6 ;:P
4
B4
56B4? :6 4 94 4 01
70
?M4 H76<?446 B470
7 norminal 7 ordinal G@2H6 64<0
2. Continuous data
01345K64B24 <6 2M:? 7766 ;23@M626 H4566 6 N 6
01G
6 01 1 2 3 ... G
<0
B24N <62M:? 7
6 6 B46
76
6 1 2 6 3 4 01 70
? continuous
J764401 2 ? B4
70
? interval 7 ratio G@2 H6 64<0
4H M4 H76?3470
?
?70
G@2
J76
?
?70
1.
(nominal scale) 01
?
?2 2:?
J4 76 <?9?K34
7:66 M <6
J4J@ 76 ? : ;
B4:6 <? 6 70
76
4401:62JB4 6 6 ;:P
4 G@2 76 34:66 M<6
J
:
44 013
B4
J7? 76 ? ?6 B46 ?46 <? B4 370
?
M<6
701345: ; 46 6
;
BM4
7601
7601
>
< H 4? 4
01
2. 4?
(ordinal scale) 0170
2
J7? 76 K24<?6 6 4
6 ?6 2 = 6 76<6
J4
4?<? 6 76 M6 6 ? 70
?M;64
H P :3 <?76 J 70
2
5@6 N
?
@K 46 6
;
7601 <6;
;
6 ;
M?
?A
z
A?46 2 A? <6A? <6A?46 2
?
@K z
<6<?@ K 0
J@K P@K 4 @K 4:?@K
3.6
(interval scale) 0170
2
J?
4?34:K276 44 01
3 73M:
B4
J 5
<?
J
: 76
<?6 6
B446 6 ? 6
4:5
?0R>> (IQ)
70
4:5
J?<?01327?J@4:55 2= 7
J0
<?6
4:5 80 =C 56 4:5 60 =C 01 76
?
6K_; B4 6 35
016 2
85H= ??4 <666 5
27H
B24H 6 5
34
?6
7?J@6
3470
M456 <6<6 46 6 4:5 0 =C
B4 0 =F <6<? J@<64:5
4. 4
6
(ratio scale)
01
?
?3470
252:?
70
J?44 013 G@2:
J 7?6 6 46 <? 7 6 5
016 5
27H
(absolute zero) B4<6
6 M= 015
J@<6 M= 70
2
J?<?
?4
6
01
0170
2 234 3
:3 ; 6 M= 65 4 : ?95
?M=
B4? 01
4H ?3470
7
H 34345A012H= 01
B4J
@K M N G@2
502 11 3452
H 70
H 7<60
502 12 3452
H 7<60
!Y;D
Z
[
:0
B4J;BMC H
Type of data
Non parametric data
Nominal
Ordinal
Describe one
group
Compare one
group
Compare two
independent
groups
Compare two
dependent
groups
Compare three
or more
independent
groups
Compare three
or more
matched groups
Correlation
between two
variables
Proportion
Chi-square or
Binomial test
Fisher's test
(chi-square
for large
samples)
Parametric data
Survival time
Median,
interquartile range
Mean, SD
Kaplan Meier
survival curve
Wilcoxon test
One-sample t test
Mann-Whitney test
Unpaired t test
Log-rank test or
Mantel- Haenszel
McNemar's test
Wilcoxon test
Paired t test
Conditional
proportional
hazards regression
Chi-square test
Kruskal-Wallis test
One-way ANOVA
Cox proportional
hazard regression
Cochrane Q
Friedman test
Contingency
coefficients
Spearman
correlation
Repeated
measures
ANOVA
Pearson
correlation
Conditional
proportional
hazards regression
-
;;3
3434: 8.?
. P
> : ;K
7
.?
. >: 02 =
345
4
0
4
4 =
,
? =
7
PH
0
1. 7H64:
.
? ;BMC . 3 : 9G;
; 2540.
2. Fletcher RH, Fletcher SW, Wagner EH. Clinical Epidemiology: the essentaials. 3 rd ed.
Maryland: William&Wilkins; 1996.
3. Grimes DA, Schulz KF. Cohort studies: marching towards outcomes. Lancet 2002 Jan 26;
359: 34145.
4. Schulz KF, Grimes DA. Case-control studies: research in reverse. Lancet 2002 Feb 2; 359:
43134.