The MRC Needs for Care Assessment: progress and controversies

The MRC Needs for Care Assessment: progress and controversies 1 In the 1980s researchers at the Medical Research Council Social Psychiatry Unit developed and refined methods for measuring needs for psychiatric care.Building upon the initial studies of Wykes el al. (1982) and Wykes & Sturt (1986), the most recent and highly researched instrument is the MRC Needs for Care Assessment (Brewin el al. 1987;Brewin & Wing, 1989).In this commentary we will review some of the knowledge that has been gained about this instrument and discuss some recent controversies concerning its use.The MRC Needs for Care Assessment was developed specifically for measuring the needs of the long-term mentally ill living in the community and in touch with psychiatric services.It involves two distinct stages of data collection.In the first information is sought from the best-placed sources, who may include members of staff, relatives, and the patients themselves, about the presence of problems in 20 areas of clinical and social functioning.In the second stage a member of staff is questioned about a standard list of items of care relevant to each identified problem.Questioning includes details of whether each item of care has been tried, and about its appropriateness, effectiveness, and acceptability to the patient.Judgements of the primary need status (met need, unmet need, no need, or no meetable need) follow algorithmically from these ratings.In order that the assessment should not be artificially constrained by current practice or resources, resulting in over-conservative judgements, raters are instructed not to take account of any training or resource implications.The assessment is thus explicitly concerned with normative needs that are independent of those of any particular service setting.Both in conception and use the instrument aims to set minimum standards of care based on a workable synthesis of current knowledge and practice rather than to strive for some notional ideal care.From the outset we were clear that judgements indicating a mismatch between the care recommended by the assessment and the care actually provided ('unmet need' and 'overprovision') can only be 'recommendations for action ' (Brewin el al. 1987, p. 980).In other words, it is not possible at one assessment to specify exactly the nature of a patient's needs, but only to make recommendations consistent with good practice.Whether there is actually an unmet need or an overprovision will depend on staff actually offering or withdrawing the relevant item of care.Only by this means is it possible to establish that a recommended action is effective and acceptable to the patient, or that an item of care can be discontinued without harm.A comprehensive needs assessment, then, usually depends on completing the audit cycle, i.e. assessing outcomes in order to evaluate the success of any interventions based on the initial assessment.The quality of the initial assessment depends in part on taking account of the acceptability of specific items of care, and in the manual interviewers are instructed to enquire about this.For some groups of patients, such as those who are homeless or out of contact with psychiatric services, this will of course be particularly important.Extension of the MRC Needs for Care Assessment methodology to these groups is therefore likely to require additional questioning to establish acceptability and to take account of patients' responses to past interventions.Another important issue relates to the guiding philosophy of care incorporated in the assessment.All needs assessments incorporate, whether explicitly or implicitly, a philosophy of care, and the results of the assessment have to be interpreted in that light.This means that an assessment should not be selected unless the users are in broad agreement with the model of care it incorporates.The