Validation of the Palliative Performance Scale for Inpatients Admitted to a Palliative Care Unit in Sydney, Australia

Abstract

To the Editor: The Palliative Performance Scale (PPS), which incorporates five observer-rated parameters (ambulation, activity and evidence of disease, self-care, intake, and level of consciousness) has been developed as a novel modification of the Karnofsky Performance Scale for measuring performance status in palliative care, allowing for the evolving physical deterioration of such patients.1Anderson F Downing G.M Hill J et al.Palliative Performance Scale (PPS) a new tool.J Palliat Care. 1996; 12: 5-11PubMed Google Scholar To date, it appears to have undergone limited reliability and validity testing.2Morita T Tsunoda J Inoue S et al.Validity of the Palliative Performance Scale from a survival perspective.J Pain Symptom Manage. 1999; 18 ([letter]): 2-3Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar, 3Morita T Tsunoda J Inoue S Chihara S Survival prediction of terminally ill cancer patients by clinical symptoms development of a simple indicator.Jap J Clin Oncol. 2000; 29: 156-159Crossref Scopus (82) Google Scholar We decided to test this new tool in consecutive admissions to a 14-bed specialist palliative care inpatient unit (PCU) in Sydney, Australia during a 6-month period. The tool was evaluated prospectively on consecutive admissions between 1 February and 31 July 2001. On the day of admission, the PPS score was determined by a specialist palliative care physician, in consultation with the registrar (at the end of their specialist training) who admitted the patient. During the evaluation period, there were 161 admissions of 139 individual patients to the PCU. The PPS was scored on the day of 154 admissions of 136 patients. It was not recorded in 7 admissions (7 patients; 4%) because of oversight, although 4 of the 7 had PPS scores recorded on other admissions during the evaluation period. Another patient who had a PPS score recorded on admission was omitted from the analysis because they were a day-only admission. For the 153 admissions that were analyzed (95% of all admissions), the median PPS score on admission was 30 (range 10–70). Death was the outcome of 107 (70%) of these admissions, and the mean length of stay of all admissions was 14.0 days (95% CI: 11.9–16.1). Fourteen (9%) of the admissions exceeded 30 days' duration. The overall post-admission survival is known in 145 (95%) of cases, and had a median of 13 and a mean of 29.1 days (95% CI: 26.9–31.3). Mode of discharge included return home (n = 33), transfer to acute care hospital (n = 6), and nursing home placement (n = 6). The median and range for length of stay (LOS) and overall survival according to each PPS score category is shown in Table 1, along with the median length of stay according to whether patients died or not.Table 1Palliative Performance Scale, Survival and Length of StayLOS—All AdmissionsLOS—Admissions Ending in DeathLOS—Admissions Ending in DischargeSurvival Post- Administration—All CasesPPS on AdmissionnDied (%)MedianRangeMedianRangeMedianRangeMedianRange101110 (91)10–1610–416an = 1.1610–62201816 (88)3.50–193.50–1922.5bn = 2.15,3040–19304942 (86)81–1625.50–162148–2281–162403823 (61)111–111121–3373–111181–204502310 (43)173–102249–71113–102309–16360116 (54)101–4121.510–4181–122910–1567020 (0)51,9——51–9145bn = 2.122,168Total153107 (70)90–16280–162111–111130–204a n = 1.b n = 2. Open table in a new tab As shown graphically in Figure 1, the percentage of patients who died during the admission decreased from over 90% with a PPS 10 to 0% with a PPS 70, similar to the findings of Anderson et al.1Anderson F Downing G.M Hill J et al.Palliative Performance Scale (PPS) a new tool.J Palliat Care. 1996; 12: 5-11PubMed Google Scholar For the majority of patients, with PPS scores ≤ 60, performance status steadily declined over the final month. As illustrated in Figure 2, the survival curves for each PPS category appear distinct (Note: PPS 60 includes the two patients with PPS 70) and survival appeared to be longer with each increase in PPS category (test for trend logrank χ2 with 1 df = 44.40, P < 0.0001). Unlike Morita et al.,2Morita T Tsunoda J Inoue S et al.Validity of the Palliative Performance Scale from a survival perspective.J Pain Symptom Manage. 1999; 18 ([letter]): 2-3Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar we did not find three distinct survival groupings. This may be because we obtained a single PPS evaluation on the day of admission, rather than multiple assessments during the admission. Our overall median post-admission survival was shorter than Morita et al.2Morita T Tsunoda J Inoue S et al.Validity of the Palliative Performance Scale from a survival perspective.J Pain Symptom Manage. 1999; 18 ([letter]): 2-3Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar (13 versus 27 days), suggesting that our patient population admitted to this PCU were further down the terminal phase trajectory. We found the PPS to be a valid and reliable tool. The PPS score on admission predicted outcome and time to death/discharge, and the prediction appeared to be irrespective of other clinical or demographic differences. Our results were very similar to those of Anderson et al.1Anderson F Downing G.M Hill J et al.Palliative Performance Scale (PPS) a new tool.J Palliat Care. 1996; 12: 5-11PubMed Google Scholar The PPS appears to be a useful adjunct to clinical decision making and prognostication in patients admitted to PCUs in Australia and Canada. Of interest would be the inter-observer reliability of the PPS when used by other physicians and/or nursing staff, as the ability of the PPS to convey meaningful prognostic information to patients, their families and palliative care health providers appears to be valuable.