Table 1.
Total sample, n = 6250a | PIF subsample n = 694b | Cases, n = 18b,c | Matched controls, n = 73b,c | Cox model, HR (95% CI), p valued | |
---|---|---|---|---|---|
Females | 52.2 (50.9‒54.0) % | 370 (53.3%) | 4 (22.2%) | 20 (27.4%) | – |
Age | 52.1 (51.7‒52.5) | 52.7 (14.1), 30‒90 | 48.1 (12.2), 30‒71 | 49.1 (10.9), 30‒73 | – |
Basic education only | 40.8 (39.4‒42.0) % | 291 (41.9%) | 3 (16.7%) | 23 (31.5%) | – |
Number of siblings | 3.8 (3.7‒3.8) | 3.9 (2.9), 0‒18 | 3.7 (3.1), 0‒10 | 3.5 (2.8), 0‒14 | – |
Toxo IgG seropositivity | 19.7 (18.3‒21.0) % | 172 (24.8%) | 4 (22.2%) | 15 (20.5%) | 1.91 (0.63, 5.76), 0.250 |
Toxo IgG antibody level | 27.3 (26.1‒28.5) | 31.0 (31.3), 3.4‒179.4 | 29.2 (30.6), 8.1‒88.4 | 23.6 (25.4), 4.4‒120.7 | 1.01 (1.00, 1.02), 0.128 |
Toxo IgM seropositivity | N/A | 119 (17.1%) | 4 (25.0%) | 13 (18.3%) | 1.38 (0.37, 5.10), 0.634 |
Toxo IgM antibody level | N/A | 2.0 (1.0), − 0.1 to 5.0 | 2.1 (1.0), 0.8‒4.0 | 1.9 (1.0), 0.4‒4.4 | 1.14 (0.66, 1.99), 0.643 |
CMV IgG seropositivity | 83.5 (82.5‒84.0) % | 583 (84.0%) | 15 (83.3%) | 59 (80.8%) | 1.03 (0.30, 3.57), 0.957 |
CMV IgG antibody level | 4.2 (4.1‒4.3) | 4.2 (2.2), 0.1‒9.9 | 3.4 (1.7), 0.4‒5.7 | 3.6 (2.0), 0.3‒7.6 | 0.89 (0.74, 1.07), 0.238 |
CMV IgM seropositivity | N/A | 114 (16.4%) | 2 (12.5%) | 6 (8.5%) | 1.24 (0.18, 8.79), 0.828 |
CMV IgM antibody level | N/A | 2.0 (1.0), 0.7‒6.1 | 1.8 (0.7), 0.8‒3.2 | 1.8 (0.8), 0.8‒4.2 | 0.88 (0.41, 1.89), 0.750 |
EBV IgG seropositivity | 97.9 (97.5‒98.0) % | 678 (97.7%) | 18 (100%) | 73 (100%) | N/A |
EBV IgG antibody level | 1.5 (1.4‒1.5) | 1.4 (0.7), 0.1‒4.5 | 1.5 (0.9), 0.4‒3.6 | 1.5 (0.7), 0.4‒3.7 | 0.98 (0.49, 1.96), 0.964 |
HSV-1 IgG seropositivity | 71.2 (70.0‒72.0) % | 496 (71.5%) | 14 (77.8%) | 47 (64.4%) | 1.69 (0.55, 5.22), 0.360 |
HSV-1 IgG antibody level | 3.7 (3.7‒3.8) | 3.8 (2.8), 0.1‒16.9 | 3.8 (2.5), 0.2‒8.5 | 3.7 (3.3), 0.2‒12.7 | 1.01 (0.88, 1.15), 0.942 |
HR Hazard Ratio (95% CI, confidence interval), p value for the infection variable
aCalculated with weights to adjust for the oversampling of individuals aged ≥ 80 years and non-response. Prevalences (95% CI), or mean (95% CI). Only IgG levels available from the total sample
bN (%), or mean (SD), range
cSuicide or self-harm diagnosis during the 15-year register follow-up (cases), and the control subjects for them, matched for age, gender, and health center district
dCox regression model predicting case status; adjusting for gender, age, educational level, region of residence, and number of siblings