journal

journal club 15

journal

article

Abstract

Abstract

Method

figure article

internal external

validity

Discussion


20 - 30

internal external validity

Critical Appraisal

external validity

general population

1. ?

2. outcome

internal validity

Internal validity study design

experimental study clinical trial

common bias trial

1. allocation bias

confounding bias

randomization

allocate randomization

baseline characteristics Results

2. ascertainment bias investigator

bias

double-blinded

clinical trial equivalence, superiority non-

inferiority?


design

equivalence trial

A is not too different from B

equivalency

null hypothesis

reject null hypothesis A

reject A B

therapeutic trial

equivalency

non-inferiority trial

A is not much worse than B

much

non-inferior non-inferiority margin

null hypothesis margin

upper limit 95% CI margin reject

null hypothesis A is not much worse non-

inferior to B

one-tailed test unbounded

equivalence trial

superiority

trial

superiority trial

A is better than B

null hypothesis equivalence trial

equally effective

reject null hypothesis

superiority

reject A B

(abesence of proof is not proof of absence)

design

non-inferiority trial

reject null hypothesis

6 internal validity

1. ?
2. poor compliance drop out (

intention-to-treat analysis

3. non-reflective outcome? outcome

4. missing data?

5. follow-up time
?

6. set non-inferiority margin N

sample size
non-inferiority trial

margin expected event rate

event rate N

set margin N underpower

non-inferiority margin treatment effect

lower limit

reject null hypothesis

50% treatment effect

FDA 1.3

Intention-to-treat on treatment

analysis?

intention-to-treat randomized

follow-up period

on treatment per-protocol follow-

up

intention-to-treat allocation bias pragmatic

underestimate (true) treatment

effect superiority trial non-inferiority

study falsely conclude non-inferiority

non-inferiority trial on treatment

analysis editor manuscript

on treatment per protocol explanatory effect

safety analysis formulate

hypothesis future trials

bias disrupt

initial randomization
 study design analysis

follow up compliance drop out rate

internal validity

1. event rate

HR OR true event rate

expected event

rate sample size

event rate absolute risk (1/AR = NNT)

relative risk reduction magnitude

(clinical significance)

2. 95% CI HR

p value (statistical significance)

3. subgroup analysis subgroup analysis

subgroup analysis ?

subgroup analysis treatment effect (

HR, RR, AR ) specific end point

primary, secondary tertiary efficacy safety

end point

treatment effect

outcome

outcome

subgroup analysis

ticagrelor CV event

study placebo

CV event

Heterogeneity Heterogeneity

treatment

effect Heterogeneity

RCT
Heterogeneity

subgroup result were consistent

across all specified subgroup interaction p value

sig

appraise ()

HOPE-3

primary outcome study control

Heterogeneity

statistical test for interaction null

hypothesis (Ho) ""

significant p < 0.05 reject

null hypothesis ( )

treatment effects p value

p value for interaction p value for trend

practically

overall treatment effect HR 0.92

(0.79 - 1.06) significant

P Trend = 0.005 reject null

hypothesis treatment effect



> 143.5 mmHg

appraise subgroup analysis

1 subgroup analysis type I error

power

sample size

primary outcome

subgroup analysis
subgroup analysis

primary outcome secondary tertiary outcome

2 subgroup analysis

primary secondary outcome

Multiplicity Steven Nissen

NEJM

Circulation subgroup analysis

Editor
 3 prespecified post hoc analysis? prespecified

post hoc analysis bias internal

validity prespecified analysis

controlled

analysis post

hoc analysis

meta-analysis

clinical trials

systematic review ( systemic

system"a"tic)

meta-analysis

1.

HR 95% CI

% weight

sample size
 total HR

95% CI

heterogeneity test

chi-square, p-value I2

heterogeneity test

p-value total hazard ratio

chi-square p-value

null hypothesis trials

p value chi-square sig reject null

heterogeneity

p sig heterogeneity

heterogeneity test

p value

heterogeneity test

0.10 cut-off point 0.05

I2 quantitative test heterogeneity

 degree heterogeneity

25% (low), 50%(intermediate), 75%(high)

HR fixed random effect model

fixed effect model true effect size

heterogeneity

random effect model

meta-analysis

hazard ratio

heterogeneity HR

2.

Forest Plot

sample size

HR

95% CI

Survival Curve

survival analysis

censored data

event

survival analysis

survival analysis 100%

impact factor

Circulation NEJM

survival analysis

editor

1. Kaplan-Meier Method

conditional probability

P(n) = P1 x P2 x .... x Pn
 P2 = ( -

) /

Probability (P) Survival Time

(T)

median survival time

P = 0.5 T

P = 0

censored data

censoring

survival curve

assumption

. competing risk

survival

curve

control

multiple regression analysis

survival curve Cox's proportional hazards

model Cox Regression ( hazard -s

. Hazard ratio

. Censored Non-censored data survival prospect

appraise survival curve

KM method p value reject null

hypothesis

chi-square (log rank)

assumption null hypothesis

2. Cox's proportional hazards model (Cox Regression)

probability hazard

ln H(t) = h0(t) + n1x1 + n2x2 + ..... + npxp

x =

n = coefficients

ln log e plot log scale

cumulative hazard

coefficient ln antilog exponential

survival curve