Table 3.

Results from the complete case (CC) and inverse probability weighted (IPW) models with estimates for the association of subject-specific factors on Self-Reported Health trajectories

	CC			IPW
	Coef.	95% CI.		Coef.	95% CI.
Body mass index
< 18.5 Kg/m2	0.238	−0.040	0.515	0.222	−0.091	0.536
18.5-22.99 Kg/m2	0.101	0.020	0.182	0.091	0.007	0.175
22-24.99 kg/m2	0.000	Reference		0.000	Reference
25-29.99 kg/m2	−0.042	−0.118	0.035	−0.033	−0.112	0.045
> 30 kg/m2	−0.173	−0.283	−0.063	−0.137	−0.252	−0.021
BMI*age
< 18.5 Kg/m2	−0.063	−0.112	−0.013	−0.058	−0.113	−0.002
18.5-22.99 Kg/m2	−0.015	−0.031	0.000	−0.014	−0.029	0.002
22-24.99 kg/m2	0.000	Reference		0.000	Reference
25-29.99 kg/m2	−0.001	−0.015	0.013	−0.003	−0.017	0.011
> 30 kg/m2	0.002	−0.017	0.022	−0.004	−0.024	0.016
age (pr 10 years)	0.951	0.137	1.766	0.933	0.055	1.810
age2	−0.333	−0.572	−0.095	−0.341	−0.597	−0.085
age3	0.043	0.013	0.073	0.045	0.013	0.077
age4	−0.002	−0.003	−0.001	−0.002	−0.004	−0.001
Observation time (year)	−0.005	−0.006	−0.003	−0.004	−0.006	−0.003
Gender	−0.018	−0.034	−0.002	−0.010	−0.026	0.007
Specific medical conditions	−0.071	−0.075	−0.067	−0.063	−0.068	−0.059
Mental distress (standardized)	−0.222	−0.230	−0.215	−0.210	−0.218	−0.201
Resting heart rate (Standardized)	−0.040	−0.047	−0.032	−0.040	−0.049	−0.032
Systelic blood pressure (pr 10 mmHg)	0.003	−0.001	0.007	0.000	0.000	0.000
Total cholesterol	0.000	0.000	0.001	0.001	−0.005	0.008
Physical activity (4 levels)
Inactive	0.000	Reference		0.000	Reference
Light	0.102	0.085	0.119	0.091	0.073	0.108
Moderate	0.177	0.159	0.196	0.158	0.139	0.178
Intensive	0.212	0.187	0.237	0.192	0.165	0.218
Daily smoking (yes/no)	−0.092	−0.108	−0.076	−0.085	−0.102	−0.068
Constant	2.321	1.307	3.335	2.376	1.281	3.471
Random part
Variance (constant)	0.111	0.105	0.117	0.165	0.158	0.171
Variance (Residual)	0.235	0.229	0.241	0.185	0.180	0.190

Complete case model (CC): Mixed-effects ML regression model, Wald chi2(23) =9184.16, Log likelihood = −25,102.758, P < 0.0001

Inverse probability weighted model (IPW): Mixed-effects ML regression model with inverse probability weighting for differential attrition, Wald chi2(23) =7683.26, Log likelihood = −29,943.958, P < 0.0001