The development, validity, reliability and practicality of a new measure of palliative care : the Support Team Assessment Schedule
Hospices and support teams offering palliative care have increased by over four fold in the U.K. during the last decade. However, evaluations have been limited, often because of a lack of suitable outcome measures. This study aimed to develop and test the validity and reliability of a measure of palliative care provided by support teams. Through detailed discussion of work objectives with care staff, assisted by review of the literature, the Support Team Assessment Schedule (STAS) was developed. STAS has 17 items, each scaled 0 (best) to 4 (worst), with definitions for each scale point. Median time to complete a STAS rating was 2 minutes (range <1- 15). Face validity was demonstrated by use in five settings. Criterion validity was assessed by comparing team ratings with ratings from patients and family members (or other carers) collected through interview. Patient and team ratings showed moderate correlations (rho ranging 0.45 - 0.66 for 5 out of 7 items): team ratings were usually closer to those of patients than those of family members. As a test of construct validity, quality of life (HCRA-QL) index items were shown to correlate with similar STAS items in patients more than four weeks before death (a STAS sub-scale of six items was correlated with the total HRCA-QL, Spearman rho -0.45). Reliability of STAS was assessed by comparing the ratings of different staff. Out of a total of 45 patient assessments, 16 items showed agreement or ratings within one score in 88% or more cases, Cohen's Kappas were greater than 0.48 (up to 0.87) and were highly significant (p < 0.0005). There were high correlation coefficients (Spearman's rho ranged 0.65 - 0.94). Split-half reliability and internal consistency was assessed using Spearman Brown coefficients and Cronbach's alpha for ratings at referral, at death or discharge, and on all weeks (combined). The coefficients ranged 0.68 - 0.89: slightly higher than the coefficients found for the HRCA-QL index. Serial scores of patients that improved (the majority) and deteriorated (a minority) under care, and results from first and last assessments of patients under the care of five support teams, indicate that the STAS was discriminating in practice. Although STAS was used to audit support team care the items are relevant to the assessment and evaluation of palliative care in other settings.