British Journal ofPsychiatry (1996), 169,317—321 Commentaries on: An Index of Need for Psychiatric Services Based on In-Patient Utilisation PAULLELUOT1BRIANJARMAN,MADHAV1 BAJEKALandGYLESR.GLOVER PAUL LEWOTT This is an important paper in that it reports the results of the study which developed the resource allocation formula recently adopted by the NHS to distribute money for mental health services between English Health Authorities. If its application leads to even partial redress of the inequities of resource allocation which have left some inner city services dangerously under-funded, it is to be welcomed. The results have face validity in that they do identify the same inner areas as having greatest need as do other indices which have been used for service planning and research. Thus the York top ten are the same districts as those identified as the ten most deprived on Jarman UPA8 Scores and they also score high on the Mental Illness Needs Index developed by Glover. As the authors point out, their index differs from others in that they have attempted to account for supply factors, i.e. the number of psychiatric beds available topeopleliving inageographical area.It is the manner in which this has been done which raises the most questions. The main measure of supply is “¿ access to long stay beds (that is, beds other than maternity and acute)―. Although it is unclear whether this includes beds other than psychiatric (e.g. learning disability, geriatric), it does appear to include psychiatric beds of all lengths of stay and for all ages. This will create distortions. In most services only “¿ acute― admission beds are effectively available to a local population; long stay beds are either occupied by a residuum of patients who have been there many years or act, in effect, as tertiary referral centres for people who cannot easily be resettled from admission wards back into the community. The number of beds, and particularly long stay beds, in an area varies enormously, more in line with historical developments than any demand. As the authors point out, “¿ many long stay beds are geographically clustered―.This is because despite their closure having been Government policy for nearly 40 years, about one-half of all beds are still based in the old “¿ water tower―hospitals (Davidge et a!, 1993). Paradoxically, from the point of view of demand, these old hospitals are often in relatively affluent areas ringing large cities; this is the case in London. Although not detailed in the paper, the method used to derive the supply variable for any particular “¿ synthetic ward of reference―assumes that the closer a population is to a hospital with beds the greater its access to them. This is a very approximate measure of supply which might create considerable distortions. Access to long stay beds is determined by historical ties, and demand mediated by purchasing, as well as by proximity. In inner cities, this is also becoming the case for short stay beds as some patients are admitted to distant hospitals as extra-contractual referrals because local beds are full (MILMIS Project Group, 1995). Inshorthemethodusedtocontrol forsupplyof beds does not accord with the way in which modern mental health services are organised. Although many are sectorised for the purposes of community care, the pattern of location of beds is very varied. A final point concerns the use of the variable “¿ proportion of persons born in new Commonwealth― as a correlate of utilisation. This presumably includes both people who were born in the West Indies, who have a greater than expected number of admissions (McGovern & Cope, 1987), and those born in the Indian sub-continent who probably do not. Its incorporation in the index is therefore likely to under-predict need in areas whose new Commonwealth population are predominantly Afro-Caribbean. References DAVIDGE, M., EUAS, S., JAYES, B., ci a! (1993) Survey of Menial Illness HospItals March 1993. Inter-Authority Comparisons and Consultancy, Health Services Management Centre, University of Birmingham, Birmingham B15 2RT. McGovrax, D. & Cain, R. V. (1987) First admission rates for first and second generation Afro-Caribbeans. Social PsychIatry, 22, 139—149. MILMIS PRQJECrGROUP(1995)Monitoring inner London mental illness services. Psychiatric BulletIn, 19, 276-280. Paul Lelllott, MRCPSyCII, Royal College of Psychia trists' Research Unit, 11 Grosvenor Crescent, London SW1X lEE 317