Skip to main content

Table 4 Differences in prescribing behaviour of cardiovascular medication between family practices (n = 45; n = 157,137 patients) with unadjusted random effect estimates, adjusted random effects estimates for family practice characteristics (i.e. urbanization and practice type), random effects estimates for patient characteristics (i.e. age, gender, social economic status and insurance type), and adjusted for practice characteristics and patient characteristics

From: Primary prevention of cardiovascular diseases: a cost study in family practices

 

τ2

OR range

Patients with CVD and endocrine diseases (n = 9,954)

  

Unadjusted

0.15

0.61 - 1.65

Adjusted for FP characteristics only

0.13

0.63 - 1.59

Adjusted for patient characteristics only

0.14

0.62 - 1.62

Adjusted for FP and patient characteristics

0.12

0.64 - 1.56

Patients with CVD (n = 20,976)

  

Unadjusted

0.6

0.73 - 1.36

Adjusted for FP characteristics only

0.5

0.76 - 1.32

Adjusted for patient characteristics only

0.5

0.75 - 1.33

Adjusted for FP and patient characteristics

0.4

0.77 - 1.29

Patients with endocrine diseases (n = 9,287)

  

Unadjusted

0.13

0.63 - 1.58

Adjusted for FP characteristics only

0.11

0.65 - 1.53

Adjusted for patient characteristics only

0.10

0.67 - 1.50

Adjusted for FP and patient characteristics

0.08

0.69 - 1.44

Patients without CVD and endocrine diseases (n = 116,920)

  

Unadjusted

0.29

0.50 - 2.00

Adjusted for FP characteristics only

0.25

0.52 - 1.91

Adjusted for patient characteristics only

0.23

0.54 - 1.86

Adjusted for FP and patient characteristics

0.19

0.57 - 1.75

  1. Notes: 1) OR = odds ratio; the OR range is calculated as the range between exp(-1.29*τ) and exp(1.29*τ);
  2. 2) FP = family practice; 3) The variance τ2 estimated in the random effects model is a measure of the between-FPs differences, and indicated the spreading of prescribing behaviour of the individual FPs; 4) CVD = cardiovascular disease.