Alternative Study designs

1
Alternative Study designs
K Danis EPIET Introductory Course Menorca, Spain
12 October 2009
, Source EPIET (1995-2008), Alain Moren,
Epiconcept, Jean Claude Desenclos, InVS (2004)
2
Cohort study measuring risk
Denominator those present at beginning Usually
short duration outbreak (attack rate)
3
Cohort study measuring rate
Individuals contribute to different length of
time Denominator sum of times
4
Cohort study
Can we use a sample of the
denominator instead of the entire
denominator?
Time
Rodrigues L et al. Int J Epidemiol.
199019205-13.
5
Controls selection

• Controls sampled to mirror the exposure
  experience in the source population
• sampled from source population that gives rise to
  cases
• representative of exposure in source population
• Sampling independently of exposure status

6
OR in case-control studies the rare disease
assumption

• Case control study very efficient for rare
  diseases
• Initially used for testing significant
  differences in exposure without attempting to
  quantify the risk associated with
  exposure  Statistically do more lung cancer
  patients have a history of smoking than controls
  ? rather than  by how many times does smoking
  increase the risk of lung cancer ? 
• Cornfield (1961) if disease is rare OR RR
• Used more and more for common diseases
• Miettinen (1976), Greenland (1981), Smith (1984)
  if controls chosen appropriately, no rare
  disease assumption is needed for the OR to
  estimate the relative risk or rate !

7
Cohort study
Time
Rodrigues L et al. Int J Epidemiol.
199019205-13.
8
Cohort study
Traditional case control study Cases and
Sample of non cases Cases Controls E
Ce (Ne-Ce ) f Non E Cu (Nu-Cu ) f
Time
Rodrigues L et al. Int J Epidemiol.
199019205-13.
9
Traditional (exclusive) designMeasure of effect
Odds ratio

• Controls sampled from population still at risk at
  the end of the study period
• OR good estimate of risk ratio and rate ratio if
  disease is rare

10
Cohort study
Case cohort study Cases and
Sample of source population Cases Contr
ols E Ce (Ne) f Non E Cu (Nu) f
Time
Rodrigues L et al. Int J Epidemiol.
199019205-13.
11
Case-cohort design Measure of Risk ratio
(relative risk)

• Control group to estimate the proportion of the
  total population that is exposed may include
  cases
• In a fixed population controls selected from all
  individuals at risk at the start of the study
• Controls sampled regardless whether or not they
  will have developed the disease
• A person selected as a case may also be selected
  as a control and vice versa
• They are kept in both groups
• No need to document disease status among controls
• Example outbreak of gastro-enteritis with 30
  attack rate

12
Cohort study
Density case control study Cases and
Sample of source population still at
risk Cases Controls E Ce (Npye) f Non
E Cu (Npyu) f
Time
Rodrigues L et al. Int J Epidemiol.
199019205-13.
13
Density case control (concurrent) design OR
estimates the rate ratio

• Controls are selected concurrently from those
  still at risk when a case occur
• A person selected as a control can later become a
  case
• The opposite not possible a case no longer at
  risk
• A control who later becomes a case is kept in
  both groups
• Controls represent person years at risk
  experience among exposed and unexposed
• Match analysis on time of selection is necessary
  to give unbiased estimate of rate ratio

14
Cohort study
Time
Rodrigues L et al. Int J Epidemiol.
199019205-13.
15
Cohort populations measures of association
Cohort design
3 measures of association
16
Cohort populations measures of association
Measures of association
17
Cohort populations, control selection to estimate
each measure of association corresponding
designs
18
How to select controls to estimate the respective
measure of association
19
What design and when?

• Traditional case control - rare disease
• Case cohort - frequent disease - same
  denominator over time - non recurrent outcome
• Density case control - rare or frequent
  disease - exposure changes over time - non or
  recurrent outcome

20
Alternative designs

• Case-to-case
•  Case-crossover 

21
Case-to-case approach
Source Jean Claude Desenclos, Jet De Valk
22
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
23
Listeriosis outbreak cases and sporadic cases
distinguished by routine PFGE, France, 1999-2000
Cases
October November December
January February March
1999 2000
de Valk H et al. Am J Epidemiol 2001154944-50
24
Controls selected among sporadic cases for the
case to case control study, listeriosis 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
25
Food consumption of case-patients and
control-subjects, multivariate analysis on 29
case-patients and 32 control-subjects. 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
26
Case-to-case control study

• Possible if disease can be classified in
  subgroups that have specific risk factors
• May be the case for infectious agents subtypes?
• Controls cases with non epidemic subtypes
• from same source population?
• same susceptibility (underlying diseases)
• included as cases if they had the outbreak strain
• readily available
• Reduces the information (recall) bias
• Food-exposure collected before status is known

27
The case-crossover design
28
The case-crossover design

• Same person taken as its own control (matched
  design)
• Compare exposure in a risk period to a prior
   control period  of the same duration
• No control group needed
• Only pairs of period discordant for the exposure
  of interest used in the analysis
• Acute diseases exposures that change overtime
• Transient exposures (drug adverse events)
• Key issue the definition of the risk period

29
Cases Matched pairs 1 Discordant 0,
1 2 Discordant 1, 0 3 Concordant 1,
1 4 Concordant 0,0
30
Case crossover  design applied to a prolonged
Salmonella Typhimurium outbreak
Haegebaert S et al. Epidemiol infect 2003130,1-5
31
Food exposures from menu information in the risk
and control period 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
32
Case-crossover design

• No need of a control group
• One to several control-periods per risk period
• Controls for between-persons confounding
• Need of data collected prior to onset
  (administrative source)
• If exposure collected by interview then very
  sensitive to recall bias
• May be biased by time trend in exposure
  between-period confounding
• Case-time-control design

33
Case-time control design
34
Between period confounding
Cyclical variation of exposure
ORa/ORb OR of exposure adjusted for time trend
35
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
36
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 want to avoid controls
• Case to case
• Case-crossover

37
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 2003130,1-5
• Hernandez-Diaz S et al. Am J Epidemiol
  2003158385-391
• Rothman KJ Epidemiology an introduction. Oxford
  University Press 2002, 73-93
• Suisa S. The case-time-control design.
  Epidemioogy. 19956248-253.
• Greenland S. Confounding and exposure trends in
  Case-cross-over and case-time-control designs.
  Epidemiology. 1996 7231-239.
• Mittleman, Maclure, Robins. Control sampling
  strategies for case cross-over studies An
  assessment or relative effectiveness. A J
  Epidemiol. 142191-98.