Bias

1
Bias Confounder

• Ram Rangsin
• MD MPH DrPH

2
????????????? 2 ????? ??????

• ????????????
• (?????????????) ??????????
• ??????? Prevalence
• ????????? Incidence
• ???????????????????????????????????
• (??????????????????????) ?????????????
• Relative Risk gt Cohort Study
• Odds Ratio gt Case-Control Study
• Prevalence Rate Ratio gt Cross-Sectional Study

3
?????????????????????????????????????????????
Cause ??????
Effect ??

• RISK FACTOR (????????????)
• Cigarette

DISEASE (??????????) Lung Cancer
4
?????? ??????????????? ???

• ????????????????????????????????? 3 ??????
• I. ??????????? CHANCE
• II. ???? BIAS
• III. ????????? 3 ???? ?????? 3rd VARIABLES
  Confounder

5
I. ??????????? Chance

• ???????????????????????????????????????????
• ????????????? ?????????????????????????
  ?????????????????????????????????????????
  ???????????????????????
• Number of sample Validity

6
Size of the problems

• Prevalence
• Incidence
• Mean
• Prevalence of HT in this population
• 30
• 95 CI 25 - 36
• Mean fasting blood glucose of the population
• 126 mg/mL
• 95 CI 80 - 160

7
Confidence interval of the estimates

• Prevalence of HT in this population
• 30 (95 CI 25 - 36
• ??????????????????????????????????? 1000 ??
  ???????????????????
• ???????????????? ??????? 1000 ????????????????????
  ?????? 100 ????? ??? Prevalence ?????????? 95
  ???????????????????? 25 - 36
• ?????????????????? 95 ??? Prevalence of HT
  ????????????????????????? 25 - 36

8
I. Chance Precision of CI

• Larger sample size provides more narrow interval
• N1000 30 (95CI 25 - 36)
• N100 30 (95 CI 9 - 58 )

N 1000
N 100
30
9
I. Chance P value

• ???????????????????? ??????? 2 ???????????
• P value ????? ?????? ????????????????
  (probability) ?????????????
• ????????? 25 ???? 0.05
• ???????????????????
• ????????????????????????????? (????)
• ?????????????????????????
• RR 3.6
• - c2 , P value 0.35

10
II. ???? (Bias)

• ????????????????????
• ??????????????????????????????????????????????
• ???????????????? exposure ???? outcome
  ??????????????
• ??????????????????????????? Relative Risk, Odds
  Ratio, Prevalence Rate Ratio ????????????????????
  ?????????????

11
???? (bias) ?? 2 ????

• ??????????????????????????????????????????????
  (Selection bias)
• ???????????? exposure ???? outcome ??????????????
  (Information bias)

12
Selection bias

• ??????????????????????????????????????????????????
  ???????????????????
• ??????????????????????????????????????????????????
  ????????????????????????????????????????
  ???????????????????????????????
• ??????????????? case ??????? control
• ????????????? ????????? ?????????????????????

13
II. Bias

• A systematic error that results in an incorrect
  estimate of the association between exposure and
  outcome
• Note Chance is random error -- not systematic.
  Therefore, tests exist based on probability
• There is no way to correct for bias later!

14
????????????? VS ???????????

• Cases ??????????????????????????????????????????
  ???????????
• Controls ???????????????????????????????????????
  ????????
• Odds Ratio ?????????????????????????????????? gt 5
  cups/day 2.6
• ???????????????????????????????????????????? Why?

15
Selecting Controls

• Coffee and pancreatic cancer, MacMahon B et al.
  NEJM 1981
• ??????????????????????????????????????????????????
  ?
• Controls ?????????????????????????????????????????
  ??????????????????????????????????????????
• ??????????????????????????????????????????????????
  ??????????????????????????? ????????????????????
  ????????????????????? ??? ??????????????????
  ??????????????????????????????????????????????????
  ???????
• ????? Control ???????????????????????
  ??????????????????????????????????????????????????
  ??????????

16
Case
Control
A
B
No Pancreatic CA
Pancreatic CA
Exposed
D
C
Not Exposed
Coffee Consumption
NO Coffee Consumption
Unrealistic Odds Ratio (BIAS)
17
???? (bias) ?? 2 ????

• ??????????????????????????????????????????????
  (Selection bias)
• ???????????? exposure ???? outcome ??????????????
  (Information bias)

18
Information Bias
19
?????????????????? VS ?????????????? X-ray
????????????

• ???????????????????????????????? X-ray
  ????????????
• ??????????????????????????????????????????????????
  ???????????????????????????? ?????????????????????
  ???????????????
• ??????????????????????????????????????????????????
  ????????????????????????????
  ?????????????????????????

20
Case
Control
Malformation
No Malformation
A
B
Exposed
D
C
Not Exposed
Hx of radiation
No Hx of radiation
Unrealistic Odds Ratio (BIAS)
21
Confounder?????? (????????? 3)
22
III. Confounder

• ???????????????????????????????????????? ?
• ??????????????????????? 3 ????????????????
  ??????????????????? ?????????????
• ??????? Alcohol VS ????????????????
• ????????? 3 ? ????????????
• ????????? 3 ???????????????????????????????????
  ????????? 2 ?????? - Confounder
• - Effect modifier or Interaction

23
Alcohol VS Lung Cancer
CA Lung CA Lung-
Alcohol 800 200
Alcohol- 200 800
1000 1000
OR 800x800 4
200x200
24
Alcohol VS CA Lung

• Odds Ratio 4
• ?????????? Alcohol ???????????????????????????????
  ????????? 4 ????
• ???????????????????????????????? Bias
• ?????????????????????????????
• ??????????????????????????????????????????????????
  

25
Alcohol VS CA Lung

• ??????????????????????????????????????????????????
  ?? ? ????
• ?????????????????
• ???????????????????????? Alcohol
  ????????????????????
• ????????????????????????
• Alcohol ??????????????????????????????????????? ?
  ???
• ??????????????????????????????????????
• ????????? ???????????????????? ???????????????????
  ??
• ??????? Alcohol ??? ?????????

26
CA Lung CA Lung-
Alcohol 800 200
Alcohol- 200 800
1000 1000
OR 800x800 4 200x200
SMOKE
NON SMOKE
CA Lung CA Lung-
Alcohol
Alcohol-
800 100
CA Lung CA Lung-
Alcohol
Alcohol-
200 900
OR 1
OR 1
27
Confounder

• Alcohol CA Lung ? OR 4 (Crude OR)
• ??????????????????????
• Alcohol CA Lung ? OR 1 (Adjusted OR)
• No association between Alcohol and CA lung
• ?????????????????????????
• Alcohol CA Lung ? OR 1 (Adjusted OR)
• No association between Alcohol and CA lung

28
Confounder

• Alcohol CA Lung ? OR 4
• ???????????????????? (Stratified by smoking
  status)
• ???????????????????????????????? Alcohol and CA
  lung
• ??????????????????????????????????? (After
  adjusting for
• smoking status)
• ???????????????????????????????? Alcohol and CA
  lung

29
Confounder

• Smoking ? Alcohol Drinking
• Smoking ? CA Lung
• ???????????????? confounder ??????????????????
• ??????? Alcohol ??? CA lung

30
Confounder

• Association between CHD and Smoking
• smokers nonsmokers
• CHD 305 58 363
• no CHD 345 292 637
• 650 350 1000
• RR 2.8

31
Confounder

• Gender M/F might be a confounder of the
  association
• Gender ? CHD
• Gender ? Smoking

32
Stratification by gender

• What is the RR among Men?
• What is the RR among Women?

33
Calculate stratum specific measures of
association...

• STRATUM 1 MEN
• smkrs nonsmkrs
• CHD 300 50 350
• NO CHD 300 150 450
• 600 200 800
• STRATUM 2 WOMEN
• smkrs nosmkrs
• CHD 5 8 13
• NO CHD 45 142 187
• 50 150 200

RR 2.0
RR 1.9
34
IS THERE A CONFOUNDER?

• CRUDE RR for smoking and CHD 2.8
• STRATUM-SPECIFIC RR for smoking and CHD with
  gender
• as a potential confounder...
• MEN RR 2.0
• WOMEN RR 1.9 roughly the same
• Do Breslow- Day tests
• Gender confounds the association between smoking
  and CHD because the
• crude RR of 2.8 is NOT the same as the
  stratum-specific RRs of approx. 2.0

35
Smoking VS CHD

• Smoking ? CHD RR 2.8 (Crude RR)
• After adjusting for gender
• Smoking ? CHD RR 2 (Adjusted RR)
• Gender confound the association between smoking
• and CHD
• Gender or Sex increases the effect of smoking

36
???????????? Confounder

• ???????????????????????????? Confounder ????????
• ?????????????????????????????????????????
• Matching (in case-control study)
• Matched by Smoking status
• Matched by Gender
• Stratified analysis ?Mantel Haenszel Method
• Multivariate analysis

37
Confounder
CA Lung
Alcohol

• Related to the disease outcome

SMOKING
CA Lung
OR or RR gt 1

• Related to the exposure

SMOKING
Alcohol
OR or RR gt 1
38
Confounder
CA Lung
Alcohol
SMOKING
CA Lung
Alcohol
SMOKING
39
Confounder

• Not be an intermediate of the association between
  the exposure and the outcome

Time to AIDS Death
HIV-1 RNA
40
Confounder

• Not be an intermediate of the association between
  the exposure and the outcome

HIV-1 RNA
Time to AIDS Death
CD4 Level
41
Confounder

• Not be an intermediate of the association between
  the exposure and the outcome

Time to AIDS Death
HIV-1 RNA
CD4 Level
Time to AIDS Death
HIV-1 RNA
CD4 Level
42
Exercise

• The effect of Non-regular exercise during young
  adult on the development of MI was different in
  each gender
• The effect was higher among Men than Women.
• Among men, those who had a history of non-regular
  exercise during young adult had 4.8 times more
  likely to develop MI than those who had a history
  of regular exercise
• Among women, those who had a history of
  non-regular exercise during young adult had 1.7
  times more likely to develop MI than those who
  had a history of regular exercise