Exploring the potential prognostic effect of various country-specific health service data on all-cause mortality using data from a large prospective study in schizophrenia


Correspondence to: A. Mittoux, H. Lundbeck A/S, Ottiliavej, 9, 2500 Valby, Denmark. E-mail: auat@Lundbeck.com



To explore whether the presence of specific health services impact the time to death in schizophrenia.


Cox's proportional hazard model was used to assess the effect of country-specific health care variables (total health care expenditure, public health care expenditure, number of hospital beds, number of physicians, and World Health Organization ranking) on time to death in schizophrenia patients from a large multinational study. Baseline patient characteristics (history of suicide attempts and monotherapy or polytherapy at the time of study entry) were also included in the model.


The number of physicians per 10,000 persons was the only health service variable associated with time to death. Each increase of one physician increased the hazard of dying by approximately 2% (95%CI, from 0.1 to 4.1%; p = 0.044) in Europe, whereas in Asia, it seemed to decrease the hazard of dying by approximately 3.6% (95%CI, from 9.9% decrease to 3% increase; p = 0.28). The effect of region as a function of the number of physicians indicated a turning point at 23 physicians per 10,000 persons: With fewer than 23 physicians per 10,000 persons, the risk of death was higher in Asia than that in Europe (hazard ratio > 1), whereas with more than 23 physicians per 10,000 persons, it was lower in Asia than that in Europe (hazard ratio < 1).


Some health services may have a significant prognostic effect on time to death in patients treated for schizophrenia, especially in Europe. The reasons for this need to be identified. Copyright © 2012 John Wiley & Sons, Ltd.