Abstract
Public sector housing is often allocated on the basis of the ‘vulnerability’ of applicants. As part of assessing vulnerability, housing departments request assessments from applicants' general practitioners (GPs). GP assessments submitted over 3 years to a local authority housing department were analysed. The nature of the patient group and format of the GP report discriminate against accurate reporting and hence fair assignment of housing.
Introduction
POOR housing adversely affects health, and appropriate housing can significantly improve health.1,2 High quality public sector housing, however, is scarce and local authorities allocate housing on the basis of the ‘vulnerability’ of applicants. Under the 1985 Housing Act local authorities have considerable flexibility in terms of defining ‘vulnerability’, although it does include old age, mental illness, or handicap and physical disability. In Brighton and Hove, in order to assess vulnerability, general practitioners (GPs) are asked to submit assessments of applicants. It is anticipated that GPs will base their assessment on the patient's notes rather than an examination and they are paid a small fee for their reports, which are submitted on a tear-off reply sheet.
Method
This study is a retrospective survey of GP assessments of housing applicants submitted to a Brighton and Hove housing department. The GP assessments in this study were completed in free text. Free text data on medical, psychiatric, or social problems were coded using a local coding system, which matched text to ICD-9 codes, and entered into Epi-Info 6 by both authors. A sample of entries was cross-checked, and confirmed coding consistency. HIV-positive applicants were excluded as their reports gave no further health details. The range and prevalence of problems among housing applicants reported by GPs was compared with the results from studies on the health of homeless groups.
Results
A total of 1494 reports were submitted during the 3-year period of 1995–1998. Most applicants were males under 30 years of age. Almost half of applicants had a documented medical problem. Orthopaedic problems were most common, although asthma was the most common single medical condition recorded (n = 126, 8.4%), followed by back pain (n = 96, 6.4%), and epilepsy (n = 68, 4.6%).
Over three-quarters of applicants had documented psychiatric problems, with depression the single most common psychiatric complaint (n = 161, 10.7%). Of those applicants with psychiatric morbidity, 35% had a coexistent medical problem and 18% had one or more documented social problems. Eight per cent of the same group had an alcohol problem and 12% a drug problem, although it was not always clear to what extent drug use was problematic. Ninety-eight (6.6%) applicants suffered from a psychotic mental illness, of whom just under half had schizophrenia.
Overall, 346 (23.2%) applicants had a recorded drug problem. The specific drugs used were not always stated, although heroin was most commonly recorded, followed by benzodiazepines. Alcohol problems were recorded in 296 (19.8%) applicants; 40% of this group had a coexisting medical problem and 12.5% a co-existing social problem.
HOW THIS FITS IN
What do we know?
Homelessness is associated with a range of physical, psychological and social problems although little is known of the health and social wellbeing of housing applicants.
What does this paper add?
This study suggests that housing applicants face many of the same problems that homeless people face. General practitioners and housing departments need to work more closely together to ensure that assessments of housing applicants are accurate and housing is allocated as fairly as possible.
Just 17% of applicants had at least one documented social problem. Marital problems, family problems, domestic violence, and criminality were the most common. Unemployment was recorded in only six reports.
Discussion
Homeless people have been the subject of considerable study. The category of housing applicants, that is people who are potentially homeless, has, to our knowledge, not been previously studied. The information in this study also came from free text rather than structured questionnaires. It would be expected, therefore, that the findings for this group would differ from other studies of homeless groups. Nevertheless, the age–sex profile in this study was consistent with studies of rough sleepers, and the medical morbidity recorded did not differ substantially from other studies.2,3 The prevalence of severe mental illness in this study was also consistent with other studies,5 although, apart from substance misuse, the recording of concurrent comorbidity was low. Eight per cent of applicants in this study had a dual diagnosis of psychiatric illness and alcohol misuse and 12% had a dual diagnosis of psychiatric illness and substance misuse: these findings were also consistent with other studies.5,6
There were some notable differences. In this study, 76.1% of patients experienced minor psychiatric problems. The prevalence of mental illness among homeless people in the United Kingdom (UK) is estimated at between 30% and 50%.4,5 The frequency of recorded social problems was very low; for example, unemployment, although closely linked with housing difficulties, was recorded in just six instances. Accurate information on social circumstances is important in assigning housing and, indeed, many housing departments, including the one in this study, now make independent assessments of social circumstances. However, it is likely that many GPs are unaware of the level of detail that is required for these assessments.
This study suggests that GPs may not report social problems accurately. This is understandable as homeless people are very mobile; indeed, some GPs recorded that they were unfamiliar with the applicant and could only interpret past records. Many homeless people will have complex health and social needs, and proper assessment will require more than a case-note review or routine appointment. The free text nature of the reports did not encourage systematic and accurate assessment. The very high levels of minor psychiatric illness recorded may be real; people with housing difficulties do face considerable stress and it is not clear if the stress of actual homelessness is worse than the stress of potential homelessness. However, the high levels of minor psychiatric illness could also be the result of patient and/or doctor exaggeration in order to improve the chances of housing allocation.
Appropriate housing is vital to health, and the GP assessment is important in determining housing allocation. Some applicants may, therefore, have been unfairly disadvantaged due to an inadequate report, while others could have been unfairly advantaged as a result of exaggeration. Assessments of homeless people will always present practical difficulties. However, the current process could be improved with guidance on what information is required and a subsequent systematic assessment by the GP. On the basis of these findings, local authorities and GPs need to work closely to ensure that assessments are accurate and complete, and that public sector housing is allocated as fairly as possible.
Acknowledgments
Thanks to Dr Angela Iversen and the staff at Brighton and Hove City Council Housing Department.
- Received May 29, 2003.
- Revision received June 19, 2003.
- Accepted February 9, 2004.
- © British Journal of General Practice, 2004.