Objective: Goal attainment scaling (GAS) is a measurement approach used extensively in mental health. It accommodates multiple individual patient goals, yet retains mathematical properties allowing comparisons between patients. This study was carried out to investigate the feasibility and measurement properties of GAS in a geriatric care setting.
Design: Prospective descriptive study.
Setting: The geriatric restorative care service and geriatric assessment unit at Camp Hill Hospital, a 350-bed tertiary care facility in Halifax, Canada.
Patients: Fifteen patients aged 65 to 94 who were consecutively admitted to the two geriatric services (mean age 79 years, mean length of stay 37 days, 9 females).
Intervention: Goal Attainment follow-up guides were developed independently for each patient by two geriatricians after a comprehensive assessment of the patient. These guides were later compared to assess level of agreement in goal setting and scale development. A single goal attainment follow-up guide was then developed for each patient by consensus of the two geriatricians. At the end of the follow-up, the guides were scored independently for each patient by one of the geriatricians and by the patient's primary care nurse.
Main outcome measures: GAS scores were determined on admission and discharge. Each patient also received admission and discharge ratings on the Barthel Index as well as a global rating of outcome (on a subjective 10-point scale) by a geriatrician who was blinded to the Goal Attainment follow-up score.
Results: GAS proved feasible, requiring 15-20 minutes to scale an average of six goals per patient. GAS also appears reliable. Of 87 goals, 71 (82%) were identified independently by two geriatricians, and the remainder were determined by consensus. This is also a measure of content validity. The physician-nurse inter-rater reliability was 0.87 (intraclass correlation). Concurrent validity was assessed by correlation with the Barthel Index (r = 0.86) and the global clinical outcome rating (r = 0.82). Content validity was also assessed by comparing our goal areas with those identified in recent consensus reports on geriatric assessment. Of these 13 assessment areas, 12 appeared to be reasonably well covered while one assessment area (sexual problems) was not identified for any of the 15 patients.
Conclusions: GAS appears to be a feasible method of goal setting and outcome evaluation in geriatric care settings, with promising reliability and validity.