Skip to main content

Table 1 Data from a hypothetical cohort study with multifactorial selection in exposure (E) originating from C (known source) and U (unknown source; Fig. 2C). The relative effect of E on the disease outcome D is constant across strata of C and U (RRED = 2.0). U influences the effect of C on D (RRCD = 1.5 when U = 0 but 4.0 when U = 1). Lack of adjustment for U therefore leads to bias in the estimated exposure effect also when C is adjusted for. Stratification for C may lead to false conclusions regarding the heterogeneity in the exposure effect

From: A novel framework for classification of selection processes in epidemiological research

U C E D = 1, n D = 0, n Risk RRED1 RRCD2
0 0 E = 1 20 80 20%   
   E = 0 90 810 10% 2.0  
  1 E = 1 90 210 30%   1.5
   E = 0 105 595 15% 2.0 1.5
1 0 E = 1 40 160 20%   
   E = 0 80 720 10% 2.0  
  1 E = 1 400 100 80%   4.0
   E = 0 200 300 40% 2.0 4.0
Totals   E = 1 550 550 50%   
   E = 0 475 2425 16% 2.32  
  0 E = 1 60 240 20%   
   E = 0 170 1530 10% 2.0  
  1 E = 1 490 310 61%   
   E = 0 305 895 25% 2.41  
  1. 1 Relative risk among exposed (E = 1) vs. unexposed (E = 0) across strata of C and U, overall and across strata of C
  2. 2 Relative risk among C-positive (C = 1) vs. C-negative (C = 0) across strata of E and U