Casecontrol designs in the study of common diseases

1
Case-control designs in the study of common
diseases alternative designs
JC Desenclos, F Simón, A Moren EPIET, Menorca,
Spain, October 9, 2006
2
Case-control studies

• Objective compare exposure in cases and in
  population origin of cases
• Sample of that population as controls
• Representative as for the exposure of interest
• Random sampling, regardless exposure or disease
  status
• Meaning of OR differs according to different
  control sampling schemes

3
Cohort populations origin of cases and controls
Rodrigues L et al. Int J Epidemiol.
199019205-13.
4
Origin of controls and measures of association
Inclusive (case-cohort)
Concurrent (density)
ExcIusive (traditional)
5
Inclusive design (case-cohort) OR estimates RR

• Controls representative proportion of total
  population at risk at the beginning of the study
  period
• ? including cases
• Sampling independent of the exposure and outcome
• A case may also be a control
• No need to asses disease status among controls
• Reasonable if source population is followed up
  for the same time period (ex OB of
  gastro-enteritis)

6
Concurrent design OR estimates Relative Rates

• Controls representative proportion of population
  at risk when the case appears (concurrent
  selection)
• Represent person-time at risk in exposed and
  unexposed
• A control can be a case later
• A person can be a control for several cases
• Matched analysis because of time matching
• Example Prolonged OB of hepatitis C in a
  dialysis unit selecting 3 controls per case among
  those at risk of infection at the same time as
  the case occurs

7
Traditional design (exclusive)

• Controls sampled from population still at risk at
  the end of the study period
• ORE of cases to controls ORD of exposed to non
  exposed
• OR good estimate of relative risk and relative
  rate if disease is rare

8
Example waterborne OB of gastro-enteritis
Attack rate 0,29 RR 3,04
Exclusive design
Case cohort design (inclusive)
OR 4,64 (CI 95 1.8 11.9)
OR 3,08 (CI 95 1.2 7.8)
9
Which design is best?

• Rear diseases similar results
• Common diseases
• Non-recurrent disease with high incidence
• Inclusive design (case-cohort) OR?RR
• Highly incident and recurrent disease or when
  probability of exposure changes along time or
  when the effect of exposure may change along time
  
• Concurrent design OR?RRate

10
(No Transcript)
11
Alternative designs

• Case to Case
•  Case - Crossover 
• Case-time-control

12
Case to case design
13
Two listeriosis outbreaks of 2 distinct PFGE
patterns, France, 1999-2000
Cases
October November December
January February March
1999 2000
de Valk H et al. Am J Epidemiol 2001154944-50
14
Listeriosis outbreak and sporadic cases by
routine PFGE pattern, France, 1999-2000
Cases
October November December
January February March
1999 2000
de Valk H et al. Am J Epidemiol 2001154944-50
15
Controls selected among sporadic cases for the
study of outbreak 2, France, 1999-2000 (Source
InVS-CNR)
Cases
October November December
January February March
1999 2000
de Valk H et al. Am J Epidemiol 2001154944-50
16
Food consumption multivariate analysis on 29
case-patients and 32 control-patients. Outbreak
of listeriosis France, December 1999 - February
2000.
adjusted for underlying condition, pregnancy
status and date of interview by logistic
regression
de Valk H et al. Am J Epidemiol 2001154944-50
17
Case-to-case study design

• Controls patient with non epidemic subtypes
• from same source population
• same susceptibility (underlying diseases)
• included as cases if they had the OB strain
• Information readily available
• Reduces the information (recall) bias
• Food-exposure collected before status is known

18
Case-Crossover design
19
Hospital and community OB of S. Typhimurium
September
October
November
December
January
Haegebaert S et al. Epidemiol infect 2003130,1-5
20
Case-Crossover design

• Same person taken as control (matched design)
• Compare exposure in a risk period to a prior
  control period of the same duration
• Matched analysis (discordant periods)
• Evaluates exposures that
• vary from time to time within a person
• triggering a short term effect, with abrupt onset
• Key issue the definition of the risk period

21
Case crossover  design applied to a prolonged
S. Typhimurium outbreak
Haegebaert S et al. Epidemiol infect 2003130,1-5
22
Food exposures from menu information in the risk
and control periods and matched OR for 17
nosocomial cases
Risk
Control
Matched
Foods
period
period
95 C.I.
OR
Exposed ()
Exposed ()
Veal
5 (29)
1 (6)
5
0,6 - 236,5
Pork
4 (23)
6 (35)
0,6
0,1 - 3,1
Hamburgers
13 (77)
5 (29)
5
1,1 - 46,9
Ham
6 (35)
5 (29)
1,5
0,2 - 17,9
Pâté
2 (12)
2 (12)
1
0,01 - 78,5
Chicken
2 (12)
3 (18)
1
0,01 - 78,5
Turkey
11 (65)
6 (35)
2,67
0,7 - 15,6
Cordon bleu
0 (0)
2 (12)
-
undefined
undefined
Lamb sausages
2 (12)
0 (0)
-
undefined
Poultry sausages
2 (12)
0 (0)
-
Haegebaert S et al. Epidemiol infect 2003130,1-5
23
Case-Crossover design

• For extended source outbreaks
• No need of a control group
• One to several control-periods per risk period
• Controls for between-persons confounding
• Very sensitive to recall bias unless data have
  been collected prior to onset (administrative
  databases)
• May be biased by time trend in exposure
  between-period confounding
• Case-time-control

24
Case-time control design
25
Between period confounding
Cyclical variation of exposure
ORa/ORb OR of exposure adjusted for time trend
26
Folic acid antagonists (FAA) in pregnancy and
congenital cardiovascular defects (CCD)

• Case Woman who had a child with CCD (N3870)
• Control Woman who had a child without CCD
  (N8387)
• Exposure FAA during 2nd 3rd month of pregnancy
• Case-crossover study for cases and controls
  independently

OR1.0 (0.5-2.0)
Case-time control OR 1/0.3 2.9 (1.2-7.2)
OR 0.3 (0.2-0.6)
Hernandez-Diaz S. Am J Epidemiol 2003158385-391
27
Conclusions

• If you do not need that OR estimates correctly
  the RR then traditional design
• Otherwise, if you need OR ? RR, identify the best
  design for each situation
• If you can not find or want to avoid controls
• Case to case
• Case-crossover

28
Find the foot fitting the glass slipper
29
References

• Rodrigues L et al. Int J Epidemiol 199019205-13
• de Valk H et al. Am J Epidemiol 2001154944-50
• Haegebaert S et al. Epidemiol infect
  2003131,809-813
• Hernandez-Diaz S et al. Am J Epidemiol
  2003158385-391
• Rothman KJ Epidemiology an introduction. Oxford
  University Press 2002, 73-93