Journal of Pain and Symptom Management
Volume 39, Issue 1 , Pages e2-e4, January 2010

Use of the Palliative Prognostic Index in a Palliative Care Consultation Service in Melbourne, Australia

The Royal Melbourne Hospital, Melbourne, Victoria, Australia

published online 16 October 2009.

Article Outline

 

To the Editor:

The article by Lau et al., “Use of the Palliative Performance Scale (PPS) for End-of-Life Prognostication in a Palliative Medicine Consultation Service,” highlights the ongoing search for an appropriate tool to estimate survival and prognosticate accurately, both in the cancer and noncancer context in different palliative care settings with varying patient demographics.1, 2, 3, 4, 5, 6, 7

We recently analyzed results from pilot implementation of a standardized prognostication tool, used for all the patients at initial consultation by our hospital-based palliative care consultation service at The Royal Melbourne Hospital, a metropolitan tertiary and general hospital in Melbourne, Australia. The aims of our study were to assess the practical utility of a prognostic tool in routine clinical practice to assist with clinical decision making and discharge planning.

We used the Palliative Prognostic Index (PPI), a prognostication tool developed and validated for use in cancer patients by Morita et al., which is based on the Palliative Performance Scale (PPS), with four additional variables of prognostic significance: oral intake, dyspnea, delirium, and edema.8, 9, 10, 11 The variables are scored independently, and the score then combined with the PPS score, resulting in a total score between 0 and 15. A PPI score of more than 6 predicts survival of less than three weeks, a PPI score of 5 or 6 predicts survival of less than six weeks, and a PPI score of ≤4 predicts survival of more than six weeks. Although the PPI has not yet been validated in a noncancer setting, our group chose to use it, as it provides a temporal estimate of survival rather than a probability of survival at a certain time, which we have found anecdotally to be more informative for patients and carers. The PPI is also simple, brief, and noninvasive. For the purposes of our study, the PPI was contrasted to Clinician Prediction of Survival (CPS).

Over four weeks, our inpatient palliative care consultation team prospectively collected data from all the inpatients with incurable disease referred to our service. The sample size was 80 patients (45% female, 55% male), with a mean age of 74.1 years; 65% had cancer diagnoses and 35% had noncancer diagnoses.

At initial consultation, CPS was determined by consensus decision among medical and nursing members of our palliative care team and a PPI score was determined. Survival information was collected eight weeks after recruitment.

As previous studies have demonstrated, CPS at our center correctly predicted survival in 50% of cases.12, 13 When incorrectly predicted, the survival was overestimated in 75% of cases. Additionally, the predictive accuracy varied with survival time, with greater accuracy seen in very short-term survival (less than one week survival accurate in 95.7% of cases) and for longer-term survival (more than eight weeks survival accurate in 59.1% of cases). For survival times ranging from one week to eight weeks, accuracy varied from 12.5% to 16.7%. Accuracy was also greater for the noncancer population; however, the majority of these patients were seen for terminal care, with 74% of these patients having a CPS of less than one week.

Our audit reaffirmed the validity of the PPI in cancer patients, with survival curves forming three distinct bands corresponding to the three projected survival times. The sensitivity (81.5% for PPI score of 5 or 6, and 85.0% for PPI score of more than 6) and specificity (70.6% for PPI score of 5 or 6, 75.0% for PPI score of more than 6) of our data were also similar to that of the original work by Morita et al.8 The survival curves of the data from the noncancer set unfortunately did not reach statistical significance due to a small sample size. However, encouragingly, there was a trend observed for different survival profiles for the corresponding PPI score.

Further analysis of our data suggests that our PPI is a more sensitive tool in predicting survival less than three or less than six weeks, but the CPS is a more specific tool for predicting survival of more than three or more than six weeks. In addition, scatter plot analysis of the CPS and PPI in cancer and noncancer patients revealed a significant (P<0.001) correlation.

In conclusion, consistent with the literature, CPS is often accurate where survival time is very short (less than one week). However, for intermediate and longer survival times, CPS can be inaccurate and is frequently overly optimistic. We would suggest that it is for this population that prognostication tools provide the greatest utility. Our data revalidate the PPI in cancer patients and suggest that it may be useful for noncancer patients. This supports the need for further research, including a large-scale validation study in the noncancer setting.

Currently, there are numerous prognostication tools available, including the PPS and the PPI, but as demonstrated in our results and in Lau et al.'s article, the tools, as well as the studies assessing these tools, have their limitations that to date have not been adequately addressed.1, 9, 14 We agree that the tools have to be used in a “local” context, incorporating consideration of patient demographics, diagnosis, and palliative service setting to assess generalizability. Ultimately, the goal of reaching an accurate, individualized prognosis for each patient will continue to require clinical acumen, in addition to any prognostication tool—not only to enable their informed use, but also to guide clinicians in how to best apply this knowledge to enhance patient care and communicate this information when appropriate in a compassionate, timely, and effective way.10, 15

Back to Article Outline

References 

  1. Lau F, Maida V, Downing M, et al. Use of the Palliative Performance Scale (PPS) for end-of-life prognostication in a palliative medicine consultation service. J Pain Symptom Manage. 2009;37:965–972
  2. Anderson F, Downing GM, Hill J, et al. Palliative Performance Scale (PPS): a new tool. J Palliat Care. 1996;12:5–11
  3. Harrold J, Rickerson E, Carroll JT, et al. Is the palliative performance scale a useful predictor of mortality in a heterogeneous hospice population?. J Palliat Med. 2005;8:503–509
  4. Virik K, Glare P. Validation of the Palliative Performance Scale for inpatients admitted to a palliative care unit in Sydney, Australia. J Pain Symptom Manage. 2002;23:455–457
  5. Lau F, Downing GM, Lesperance M, et al. Use of Palliative Performance Scale in end-of-life prognostication. J Palliat Med. 2006;9:1066–1075
  6. Olajide O, Hanson L, Usher BM, et al. Validation of the Palliative Performance Scale in the acute tertiary care hospital setting. J Palliat Med. 2007;10:111–117
  7. Downing M, Lau F, Lesperance M, et al. Meta-analysis of survival prediction with Palliative Performance Scale. J Palliat Care. 2007;23:245–252
  8. Morita T, Tsunoda J, Inoue S, Chihara S. The Palliative Prognostic Index: a scoring system for survival prediction of terminally ill cancer patients. Support Care Cancer. 1999;7:128–133
  9. Viganò A, Dorgan M, Buckingham J, et al. Survival prediction in terminal cancer patients: a systematic review of the medical literature. Palliat Med. 2000;14:363–374
  10. Maltoni M, Caraceni A, Brunelli C, et al. Prognostic factors in advanced cancer patients: evidence-based clinical recommendations—a study by the Steering Committee of the European Association for Palliative Care. J Clin Oncol. 2005;23:6240–6248
  11. Stone CA, Tiernan E, Dooley BA. Prospective validation of the Palliative Prognostic Index in patients with cancer. J Pain Symptom Manage. 2008;35:617–622
  12. Viganò A, Dorgan M, Bruera E, et al. The relative accuracy of the clinical estimation of the duration of life for patients with end of life cancer. Cancer. 1999;86:170–176
  13. Twomey F, O'Leary N, O'Brien T. Prediction of patient survival by healthcare professionals in a specialist palliative care inpatient unit: a prospective study. Am J Hosp Palliat Care. 2008;25:139–145
  14. Lau F, Cloutier-Fisher D, Kuziemsky C, et al. A systematic review of prognostic tools for estimating survival time in palliative care. J Palliat Care. 2007;23:93–112
  15. Glare PA, Sinclair CT. Palliative medicine review: prognostication. J Palliat Med. 2008;11:84–103

PII: S0885-3924(09)00766-0

doi:10.1016/j.jpainsymman.2009.08.001

Journal of Pain and Symptom Management
Volume 39, Issue 1 , Pages e2-e4, January 2010

Article Tools