Skip to main content

Table 3 Estimates of regression coefficient of multilevel regression analysis for practice and patient characteristics (number of patients = 6997; number of practices = 96)

From: Quality of life of Australian chronically-ill adults: patient and practice characteristics matter

Parameters (reference category)

Estimate for the main model

 

Physical components score (PCS-12)

Mental components score (MCS-12)

 

Regression Coefficients (Standard Error)

Regression Coefficients (Standard Error)

Patient main effect

  

Intercept

35.44

33.74

Female patients (male)

2.78 (0.46)‡

1.54 (0.51)†

Age, years

  

40–59 (18–39)

-4.65 (0.40)‡

2.22 (0.44)‡

>59 (18–39)

-7.64 (0.45)‡

6.02 (0.50)‡

Good or very good health (very bad, bad or fair health)

10.83 (0.23)‡

7.34 (0.25)‡

Owner-occupier (rented)

0.79 (0.29)†

1.72 (0.32)‡

College/university (elementary/high School)

1.02 (0.25)‡

-0.05 (0.28)

Employed patients (unemployed)

7.29 (0.48)‡

4.81 (0.52)‡

Retired patients (unemployed)

3.99 (0.48)‡

3.69 (0.53)‡

Married/cohabiting (single/separated/divorced/widowed)

0.47 (0.25)

0.63 (0.27)*

Born in Australia (non-English countries)

-0.68 (0.35)

1.46 (0.39)‡

Born in USA/UK/Canada/NZ (non-English countries)

0.23 (0.44)

0.87 (0.48)

Diabetes (two or more conditions)

2.66 (0.37)‡

1.49 (0.41)‡

Ischaemic heart disease/hypertension (two or more conditions)

1.24 (0.34)‡

0.55 (0.37)

Asthma (two or more conditions)

1.01 (0.42)*

1.42 (0.46)†

Disease unknown (two or more conditions)

3.83 (0.31)‡

1.22 (0.34)‡

Overall satisfaction with care

-0.30 (0.23)

1.20 (0.25)‡

Patient interaction effect

  

Female × employed

-3.46 (0.59)‡

-3.34 (0.65)‡

Female × retired

-3.34 (0.58)‡

-1.38 (0.64)*

Practice main effect

  

Size 1–3 general practitioners (4 or more GPs)

-0. 99 (0.27)‡

-0.55 (0.26)*

Urban (Rural)

0.53 (0.28)

0.16 (0.27)

  1. Note: *P < 0.05, † P < 0.01, ‡ P < 0.001
  2. Interactions not shown in the table were not included in the model.
  3. NZ = New Zealand
  4. Patient characteristics were collected independently using GPAS22 for the same respondents.