A COMMUNITY NEEDS ASSESSMENT - THE CARE NEEDS ASSESSMENT PACK FOR DEMENTIA (CARENAPD) - ITS DEVELOPMENT, RELIABILITY AND VALIDITY

Objective. To develop and evaluate a multidisciplinary needs assessmen t tool for people with dementia living in the community and their care rs. Design. The measure was developed through applying a theory of nee d, generating content, consultation with potential users and refinemen t and evaluation, Validity was established incrementally through the d evelopment process, Setting. The development and evaluation was conduc ted in a variety of settings, including multidisciplinary dementia com munity care teams, social work departments, day hospitals, and inpatie nt and residential care. Patients. The evaluation included community p atients with a formal diagnosis of dementia (N = 34) and consultation with a multidisciplinary group of potential users (N = 23). The develo pment process included inpatients with a formal diagnosis of dementia (N = 157) and consultation with potential users (A: = 170) from a rang e of professions including both health and social cars. Measures. Inte rrater reliability was assessed using the kappa statistic. Social vali dity was estimated using a measure developed far this purpose as part of the development process, Results. The evaluation of interrater reli ability demonstrated that three-quarters of assessors agreed on at lea st 85% of items in the CarenapD. The kappa statistic demonstrated that agreement for 76.2% of items in the CarenapD was 'good' or better tie kappa >0.75), for 12.4% of items it was 'fair' or 'moderate' tie kapp a 0.35-0.60) and for the remaining 12 (11.4%) items for which kappa co uld not be calculated there was low intra-item variance and high agree ment (>90%). There was good evidence for social validity. Conclusions, The CarenapD is a reliable and valid multidisciplinary assessment of need for people with dementia living in the community and their carers . (C) 1998 John Wiley & Sons, Ltd.