Psychometric properties of the Dutch version of the Health of the Nation Outcome Scales for older adults (HoNOS 65+) in daily care

Background: The Health of the Nation Outcome Scales (HoNOS) is widely used to evaluate mental health care outcomes. For appropriate use and interpretation in routine clinical practice, further validation of the adapted version for older clients (HoNOS 65+) is needed. Objective: The aim of this study is to compare scoring profiles produced by different categories of professionals, assess the internal consistency of the sum score and proposed subscales, and concurrent validity of the total sum score of the Dutch version of HoNOS 65+. Methods: We used baseline data from fourteen mental health care organizations participating in the MEntal health care Monitor Older adults (MEMO), a nationwide routine outcome monitoring system. A total of 767 older clients, referred for gerontopsychiatric disorders, were administered HoNOS 65+ by non-academic (primarily nurses, n = 430) or academic professionals (psychologists/physicians, n = 337). Demographics and full DSM-IV classification, including the Global Assessment of Functioning (GAF), were derived from the electronic medical dossier. Results: HoNOS 65+ seemed to discriminate between clients with and without a depressive disorder, adjustment disorder, anxiety disorder and psychotic disorder on the items expected. In clients suffering from a depressive or psychotic disorder, nurses/social workers and physicians/psychologists did not differ in scoring on all items of HoNOS 65+. In clients with an adjustment disorder or anxiety disorder, professions differed in rating on two items. Confirmatory Factor Analyses supported neither the factor structure of the original HoNOS nor the initially reported structure of HoNOS 65+ version. Cronbach's alpha of the total sum score was 0.60. Internal consistency of previously identified subscales was low. A new set of subscales could not be identified satisfactorily. A medium-sized correlation of the HoNOS 65+ sum score and the GAF was found (r = -.30, p<.001). Conclusions: It is preferable to use individual HoNOS 65+ items, to evaluate care outcomes. The HoNOS 65+ items discriminate between clients with and without a particular diagnosis. Nonetheless, in some of the most prevalent gerontopsychiatric disorders rating differs between professionals with different educational backgrounds.