Rapid Responses to:

PRIMARY CARE: Merete Nordentoft and Nina Wandall-Holm 10 year follow up study of mortality among users of hostels for homeless people in Copenhagen BMJ 2003; 327: 81 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Social Exclusion?

Phil Geis   (16 July 2003)

Deaths in homeless men in Glasgow, UK

David S Morrison   (7 August 2003)

Social Exclusion? 16 July 2003
Phil Geis, NA NA Send response to journal: Re: Social Exclusion?	The authors offer an apparently gratuitous recommendation as the final sentence of the article. "The prevention of social exclusion should start early in life" "Social exclusion" is not defined here and appears no where else in this report. Importanty, this concept was not identified as a predictor of early death or associated with any such factors identified and no means of its prevention were offered or discussed. Where the BMJ editors asleep? Competing interests:   None declared
Deaths in homeless men in Glasgow, UK 7 August 2003
David S Morrison, Consultant in Public Health Medicine Greater Glasgow NHS Board, Dalian House, PO Box 15329, 350 St Vincent Street, Glasgow G3 8YZ. Send response to journal: Re: Deaths in homeless men in Glasgow, UK	In the United Kingdom, Glasgow has a high prevalence of single homelessness, with 11 889 homeless presentations to the local authority made by 7 690 individuals (repeat presentations are common) in 2001-02. Although historical data are not available to perform a similar 10-year cohort study to that published by Nordentoft and Wandall-Holm (1), a comparative estimate of indirectly Standardised Mortality Ratios (SMRs) among male homeless hostel dwellers in Glasgow is possible. Person-time at risk of death data were obtained by summing all age- specific lengths of stay of local authority hostel dwellers over the three years between 1999 and 2001, using the local authority housing management information system. The high turnover of hostel beds means that this technique gives a more accurate representation of person-time than a simple hostel bed count. Females were excluded because they comprise only 12% of the local authority hostel population and absolute numbers of deaths are very small. All death records for individuals who died within the local authority hostels were obtained for the period from Greater Glasgow NHS Board (GGNHSB). One main cause of death was ascribed for each individual. Expected death rates for age-specific males were obtained from General Register Office for Scotland death records. 78 male deaths were recorded over the 3 years 1999 to 2001. The all- cause age standardised mortality ratio was 2.3 (95% CI 1.8 to 2.8). This is very close to Nordentoft’s (1) SMR of 2.8 for men. The commonest cause of death (23/78, 29%) was poisoning by drugs, usually heroin, for which the SMR was 7.9 (4.7 to 11.1). Alcohol was the second commonest cause of death (18/78, 23%) with an SMR of 2.4 (95% CI 1.3 to 3.5). The SMR for suicide was 77.4 (n=7, 95% CI 20.1 to 134.8). This is much greater than Nordentoft’s estimate of 6.0, but close to the figure of 62.9 for unknown causes, and may reflect differences in diagnostic and coding practices. Deaths from circulatory diseases (n= 10, SMR 1.1, 95% CI 0.4 to 1.8) and neoplasia (n=7, SMR 0.8, 95% CI 0.2 to 1.4) were not significantly outwith the Scottish population rates although they suffer from greater imprecision because numbers are smaller. Glasgow’s male homeless hostel dwellers share similarly increased risks of death to those in the Copenhagen cohort. Death rates from drugs, alcohol, and suicide are particularly high and standardisation masks particularly high rates among younger hostel residents. Glasgow’s plan to close its large local authority homeless hostels in favour of smaller more appropriate alternatives has already begun. These alternatives should reduce the violence, intimidation, and exposure to drugs and alcohol that hostels add to a homeless person’s problems (2). The effect on deaths from these causes will be closely monitored as one measure of the success of new alternatives to homeless hostels. Thanks to Tricia Richardson, Information Services, Greater Glasgow NHS Board, who supplied death records. (1) Nordentoft M, Wandall-Holm N. 10 year follow up study of mortality among users of hostels for homeless people in Copenhagen. BMJ 2003;327:81. (2) Morrison DS. The extent, nature, and causes of homelessness in Glasgow. Glasgow, Greater Glasgow NHS Board, 2003. Competing interests:   None declared