Marcel Slockers

67 The effects of the Action Plan for Social Support on the health of homeless people transition period (2003–08) and one after full implementation of the new policy (2008–09). This also did not reveal differences, nor did considering the year-by-year changes provide an indication of any systematic change in mortality rates during the study period. The lack of a reduction in mortality among homeless people after 5 years implies that the policy efforts had no short-term effect on mortality among people who were already homeless in 2001. This is contrary to the success of these policies in terms of the objectives of the policy intervention, which were all met, 14 as well as on other relevant parameters, such as the safety index of Rotterdam. After 2006, the safety index of the city as a whole and of several problem areas has rapidly improved, reflecting reductions in violent crimes, street use of illegal drugs and feelings of unsafety, among others. 20 This provides an indication of improved quality of life of all Rotterdam citizens, including the (former) homeless. Moreover, a 75% decrease in tuberculosis incidence among homeless persons has been observed. 21 However, in spite of these successes, these policies were so far not successful in reducing mortality among homeless people in our cohort. This could be due to incomplete reach of the program and persistent excess mortality in a subgroup remaining homeless. Another possible explanation is that policy efforts will have their beneficial effects on mortality in the long term, but that this is not yet reflected in reduced mortality up to 2010. The cohort of homeless persons established in 2001 carries the burden of a long previous history of homelessness, unhealthy living circumstances and lifestyles (e.g. use of alcohol and illicit drugs, heavy smoking) and related chronic health problems (e.g. damage to the cardiovascular and respiratory system, such as atherosclerosis and chronic obstructive pulmonary disease). Providing housing, daytime activities and controlling addictions of homeless persons have certain benefits but cannot provide a survival advantage for the chronically disadvantaged. Combating other health problems that result from long-term exposure to unfavourable lifestyles and environmental factors (e.g. cardiovascular and respiratory diseases) may be a preventative measure able to produce the largest health benefits. 22 Our results indicate that reducing the still very high mortality rates of homeless people asks for additional policies beyond the provision of housing and other services, and attention to the prevention of homelessness seems needed. Ethics statement: The Medical Ethical Review Committee of the Erasmus MC declared that this study was not subject to the Law on Medical Research with human beings and that it had no objections to the performance of this study. Funding This study was supported by a grant from the Municipal Public Health Service (GGD) Rotterdam-Rijnmond.