Marcel Slockers

117 General discussion excess mortality among homeless people in Rotterdam is lower than in a study from Amsterdam, which showed a seven times higher mortality among 364 homeless people. The studied group from thisAmsterdam study was rather small, and therefore the estimate is rather uncertain with a wide 95% confidence interval (95% btbi:4.1-13.5). 11 The difference in mortality can possibly be explained by differences between the study populations. The Amsterdam study describes the excess mortality of rough sleepers who had been seen by a street doctor in a day and night shelter, whereas in the Rotterdam study this preselection -which leads to a more vulnerable population- was not made. Not much has yet been reported in literature about the effects of excess mortality on life expectancy among homeless people. In our study, we found that 20-year-old homeless men andwomen live 14 and 16 years shorter, respectively, than the general Rotterdampopulation. This effect is smaller than in a study from Denmark, which showed a 17 (for women) to 22 (for men) year shorter life expectancy. 7 The lower life expectancy of homeless people in Denmark -especially for men- can possibly be explained by the fact that the ten-year Danish study from 1999 onwards showed more psychiatric diagnoses (49%) and substance abuse (49%) among homeless people than our Dutch study. An important group of Danish homeless people consists of persons who originally come from Greenland. This former Greenlandic population has extremely high rates of addiction and psychiatric problems. In Denmark, there has been no similar action plan and the so-called harm reduction approach, aimed at limiting health damage, was started later than in the Netherlands. Studies showing male-female difference In our study, the adverse effect of homelessness on life expectancy was found to be greater in women than in men. This is consistent with the findings of a recent review study. 6 Homeless women may comprise a more extreme selection of the total female population, in terms of mental and other problems, than homeless men. Young homeless women are more likely to have an unhealthy lifestyle, more severe alcohol and drug abuse, and more psychiatric health problems than homeless men. 6,12,13 Homeless women are hospitalized more often, particularly for hepatitis and intoxication. 14 Studies on causes of death The most important cause of death in our cohort study was unnatural death (26%). This is also the case in a large Danish study (28%), and other international studies have also reported that accidents, suicide, and violence account for a much larger share of mortality than among the general population. 2,5,7,8 According to a review study, cardiovascular diseases, infectious diseases, and mental health problems are the main causes of death