Skip to main content

Table 2 Population attributable risk (PAR) of risk factors in the original and extended cohort

From: The potential for prevention of dementia across two decades: the prospective, population-based Rotterdam Study

 

RS-I

RS-II

 

n/N 624/7,003

n/N 145/2,953

 

HR (95 % CI)

PAR (95 % CI) per category

Combined PAR per risk factor (95 % CI)

HR (95 % CI)

PAR (95 % CI) per category

Combined PAR per risk factor (95 % CI)

Body mass index

  

NA

  

NA

 <18.5

1.25 (0.62–2.56)

0.00 (0.00–0.08)

 

1.41 (0.19–10.49)

0.00 (0.00–0.67)

 

 25–30

0.96 (0.81–1.15)

NA

 

0.90 (0.61–1.32)

NA

 

 >30

0.84 (0.65–1.08)

NA

 

0.81 (0.48–1.36)

NA

 

Hypertension

1.07 (0.89–1.27)

 

0.04 (0.00–0.44)

1.25 (0.81–1.94)

 

0.16 (0.02–0.62)

Diabetes mellitus

1.31 (1.05–1.63)

 

0.04 (0.01–0.09)

1.51 (0.96–2.37)

 

0.06 (0.02–0.19)

Cholesterol/HDL cholesterol

  

0.03 (0.00–0.73)

  

NA

 Quartile 2

1.04 (0.83–1.30)

0.01 (0.00–0.82)

 

0.64 (0.39–1.04)

NA

 

 Quartile 3

1.08 (0.86–1.35)

0.02 (0.00–0.30)

 

0.86 (0.53–1.39)

NA

 

 Quartile 4

1.04 (0.83–1.32)

0.01 (0.00–0.70)

 

0.72 (0.42–1.21)

NA

 

 Lipid lowering medication

0.64 (0.28–1.45)

NA

 

1.03 (0.60–1.74)

0.00 (0.00–1.00)

 

Smoking

  

0.07 (0.02–0.23)

  

NA

 Former

1.10 (0.90–1.34)

0.03 (0.00–0.23)

 

1.00 (0.68–1.48)

0.01 (0.00–1.00)

 

 Current

1.33 (1.04–1.71)

0.04 (0.02–0.10)

 

0.86 (0.48–1.52)

NA

 

Educational level

  

0.07 (0.00–0.90)

  

0.12 (0.00–0.89)

 Intermediate

0.99 (0.68–1.45)

NA

 

1.01 (0.57–1.78)

0.00 (0.00–1.00)

 

 Low

1.13 (0.78–1.63)

0.08 (0.00–0.66)

 

1.32 (0.73–2.36)

0.11 (0.02–0.51)

 

Total

  

0.23 (0.05–0.62)

  

0.30 (0.06–0.76)

  1. Estimates represent hazard ratios (95 % CIs) and population attributable risks (PARs; 95 % CIs). Models were adjusted for age, sex, and for body mass index, hypertension, diabetes mellitus, total cholesterol/HDL cholesterol ratio, lipid-lowering medication, smoking, and educational level, if appropriate. For some risk factors, PARs could not be calculated as these risk factors were not related to dementia in the expected direction; these PARs are therefore not applicable (indicated with NA) [25]
  2. Abbreviations: RS-I Rotterdam Study I, original cohort, RS-II Rotterdam Study II, extended cohort, HR Hazard ratio, CI Confidence interval, n Number of cases, N Number of people at risk, PAR Population attributable risk, NA Not applicable, HDL High density lipoprotein