Table 4

Pharmaceutical costs - results of two-step regression analysis

1. Probabilitya

N = 14, 531

2. Amount of costsb

N = 5, 791


Parameter

Odds ratio

Pr > Chi2g

Exp(est)

Pr > |t|g


Intercept

-

-

9.5571

< 0.0001


Sex: female

1.351

< 0.0001

0.849

< 0.0001


Age

0.728

< 0.0001

0.970

0.1559

Age squared

1.015

< 0.0001

1.000

0.6812


BMIc: very underweight

1.006

0.3100

2.461***

< 0.0001

underweight

1.077

1.387***

overweight

1.129

1.049

obese

1.122

1.237**


Socioeconomic statusd: high

1.075

0.4349

1.144**

0.0283

medium

1.034

1.069


Seasone: spring

0.960

0.3520

0.973

0.7188

summer

0.869

1.152

autumn

0.955

0.900


Migrantf

0.673***

< 0.0001

1.144*

0.0144


aLogistic mixed regression; bGeneralised linear mixed regression model (Gamma distribution with log-link);

Reference: cnormal weight, dlow socioeconomic status, ewinter, rnon-migrant.

gp-value of total effect.

Significance levels for individual effect levels: *** < 0.001, ** < 0.01, * < 0.05.

BMI, body mass index.

Wenig et al. BMC Health Services Research 2011 11:340   doi:10.1186/1472-6963-11-340

Open Data