Interaction

1
Interaction

• Selected reading

2

• Definition
• Types
• Interpretation
• Tracing

3
??? ???? ?? ?? ??????? ???? ?????

• ?????? ?? ??? ???? ?? ??????? ???? ????. ????
  ?????? ????? ???? ??? ? ????? ?? ????? ?? ?????.

4
???? ?????? ??? ???????? ?? ??? ??????
5
???? ?????? ??? ???????? ?? ??? ??????

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

6
??? ????? ?? ????? ?????? ?? ?????
7

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

8
??? ??? ?????? ???? ????? ??? ????? ?? ?????
?????? ?? ?????
9
????? ??? ?????
10

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

11
Interaction or effect modification
Variation in the magnitude of measure of
effect across levels of a third variable.
Happens when RR or OR is different between
strata (subgroups of population)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
(AB - aB) (Ab ab) 0 AB aB Ab ab
16

• Definition
• when the magnitude of a measure of association
  (between exposure and disease) meaningfully
  differs according to the value of some third
  variable
• Synonyms
• Effect modification
• Effect-measure modification
• Heterogeneity of effect

17


18
Additive Risk difference
(AB - aB) (Ab ab) 0 AB aB Ab ab AB
ab (Ab ab) (aB ab)
19
Multiplicative Risk Ratio
(AB / aB) (Ab / ab) AB / ab (Ab / ab) (aB
/ ab)
20
Smoking, Family History and Cancer Additive vs
Multiplicative Interaction

Crude
Family History Present
Stratified
Family History Absent
RRno family history 2.0 RDno family history
0.05
RRfamily history 2.0 RDfamily history 0.20

21
AB aB Ab ab (AB / aB) (Ab / ab)
22
Types of interaction

• Additive vs. Multiplicative
• Qualitative vs. Quantitative

23

24


25


26
Smoking, Caffeine Use and Delayed Conception
Additive vs Multiplicative Interaction

Crude
RR crude 1.7 RD crude 0.07
Stratified
No Caffeine Use
Heavy Caffeine Use
RRno caffeine use 2.4 RDno caffeine use 0.12
RRcaffeine use 0.7 RDcaffeine use -0.06
27
Additive vs. Multiplicative Interaction

• Assessment of whether interaction is present
  depends upon which measure of association is
  being evaluated
• ratio measure (multiplicative interaction) or
  difference measure (additive interaction)
• Absence of multiplicative interaction always
  implies presence of additive interaction

28

• Absence of additive interaction always implies
  presence of multiplicative interaction
• Presence of multiplicative interaction may or may
  not be accompanied by additive interaction
• Presence of additive interaction may or may not
  be accompanied by multiplicative interaction
• Presence of qualitative multiplicative
  interaction is always accompanied by qualitative
  additive interaction

29

• (AB / aB) (Ab / ab)
• ln (AB / aB) ln (Ab / ab)
• ln AB ln ab ln Ab ln ab

30
Smoking, Caffeine Use and Delayed Conception

Crude
RR crude 1.7
Stratified
No Caffeine Use
Heavy Caffeine Use
RRno caffeine use 0.7
RRcaffeine use 2.4
RR adjusted 1.4 (95 CI 0.9 to 2.1)
31
What are the Possible Effects of a 3rd Variable?

• Intermediary Variable
• Effect Modifier (interaction)
• Confounder
• No Effect

32
Intermediary Variable (conceptual decision)?
Report Crude Estimate
no
yes
Effect Modifier?
no
Report stratum-specific estimates
yes
Confounder?
Report adjusted summary estimate
No Effect Report Crude Estimate
yes
no
33
Chance as a Cause of Interaction?

Crude
OR crude 3.5
Stratified
Age gt 35
Age lt 35
ORage gt35 5.7
ORage lt35 3.4
34

• . cs delayed smoking
• smoking
• Exposed Unexposed
  Total
• ------------------------------------------------
  ---
• Cases 26 64
  90
• Noncases 133 601
  734
• ------------------------------------------------
  ---
• Total 159 665
  824
• Risk .163522 .0962406
  .1092233
• Point estimate 95
  Conf. Interval
• -------------------------------
  ---------------
• Risk difference .0672814
  .0055795 .1289833
• Risk ratio 1.699096
  1.114485 2.590369
• -----------------------------------------------
• chi2(1) 5.97
  Prgtchi2 0.0145

35

• . cs delayed smoking, by(caffeine)
• caffeine RR 95 Conf.
  Interval M-H Weight
• -------------------------------------------------
  -----------------
• no caffeine 2.414614 1.42165
  4.10112 5.486943
• heavy caffeine .70163 .3493615
  1.409099 8.156069
• -------------------------------------------------
  -----------------
• Crude 1.699096 1.114485
  2.590369
• M-H combined 1.390557 .9246598
  2.091201
• -------------------------------------------------
  -----------------
• Test of homogeneity (M-H) chi2(1) 7.866
  Prgtchi2 0.0050

36

Since the stratified ORs are different
(heterogeneous), there is multiplicative
interaction
37
6.52x 1.34 8.47

38

• (AB / aB) (Ab / ab)
• (AB / aB) gt (Ab / ab)
• (AB / aB) lt (Ab / ab)
• (AB aB) (Ab - ab)
• (AB aB) gt (Ab - ab)
• (AB aB) lt (Ab ab)

39
Types of interaction

• Additive vs. Multiplicative
• Qualitative vs. Quantitative
• Synergism vs. Antagonism

40
??? ?? ????? ???? ? ?? ????? ?????? ????? ????
??? ?????? ?????
41

• ???
• ?????? ?? ???? ????? ?? ??????? ?? ????? ?????? ?
  ????????? ?? ????? ????? ????? ?? ??? ??????? ??
  ??? DOTS

42
??? ?? ????? ???? ? ?? ????? ?????? ????? ?????
??? ?????? ?????
43

• Definition
• Types
• Interpretation
• Tracing
• Matching

44

• If heterogeneity is present is there
  interaction?
• If we concluded that there is interaction, what
  should we do?