PSYCHOMETRIC EVALUATION OF THE MODIFIED STRAIN IN NURSING CARE ASSESSMENT SCALE
Author(s)
Ciesla G1, Frank L1, Kleinman L1, Brodaty H2, Rupnow M3 , 1MEDTAP International, Bethesda, MD, USA; 2University of New South Wales, Sydney, Australia; 3Janssen Pharmaceutica Products, L.P, Titusville, NJ, USA
OBJECTIVES: The objective of this study was to evaluate the psychometric properties (reliability and validity) of the Modified Strain in Nursing Care Assessment Scale (M-NCAS), a 32-item scale for assessing dementia-related behaviors and the extent of job burden caused by those behaviors. METHODS: M-NCAS data were collected from nursing staff ratings of nursing home patients diagnosed with dementia (N=281), in a randomized double-blind clinical trial comparing risperidone to placebo (RIS-AUS-5 trial). For each item nurses rated the extent to which they agree the target patient exhibits the behavior (agree, partially agree, doubtful/unsure, don't agree) and the extent to which the behavior is difficult to cope with (very easy, easy, difficult, very difficult). Exploratory factor analyses were performed separately for "agree" and "cope" scales to determine if empirically and conceptually valid subscales exist. Internal consistency reliability was assessed via Cronbach's alpha. Correlations with the BEHAVE-AD and Cohen-Mansfield Agitation Inventory were examined to evaluate construct validity. RESULTS: Factor analysis resulted in identification of 3 agree subscales (difficulty, attention seeking, autonomy) and 5 cope subscales (predictability, self direction, neediness, job satisfaction, affect). Total scores were internally consistent, with alphas of 0.74 for the agree scale and 0.95 for the cope scale. Internal consistency reliability was also acceptable for the 3 agree subscales (0.69, 0.77, 0.57) and for the 5 cope subscales (0.89, 0.86, 0.83, 0.86, 0.79). The M-NCAS agree and cope total scale scores correlated most highly with the BEHAVE-AD aggressiveness and anxiety/phobia subscales (r=0.36, r=0.35; r=0.43, r=0.35 respectively: p<0.001) and the CMAI verbal/non-aggression and aggression subscales (r=0.45, r=0.32; r=0.34, r=0.31, p<0.001), indicating adequate construct validity. CONCLUSIONS: The 32-item M-NCAS is an internally consistent and valid scale for capturing both dementia-specific behaviors and nursing perception of job burden associated with those behaviors. The M-NCAS provides detailed item-level as well as subscale-level data.
Conference/Value in Health Info
2002-05, ISPOR 2002, Arlington, VA, USA
Value in Health, Vol. 5, No. 3 (May/June 2002)
Code
PMH34
Topic
Patient-Centered Research
Topic Subcategory
Patient-reported Outcomes & Quality of Life Outcomes
Disease
Mental Health