Outcome research in palliative care: could it represent a new dimension of clinical research or clinical practice?

Using Laboratory Test Results at Hospital Admission to Predict Short-term Survival in Critically Ill Patients With Metastatic or Advanced Cancer

CONTEXT

Accurately estimating the life expectancy of critically ill patients with metastatic or advanced cancer is a crucial step in planning appropriate palliative or supportive care.

OBJECTIVES

We evaluated the results of laboratory tests performed within two days of hospital admission to predict the likelihood of death within 14 days.

METHODS

We retrospectively selected patients 18 years or older with metastatic or advanced cancer who were admitted to intensive care units or palliative and supportive care services in our hospital. We evaluated whether the following are independent predictors in a logistic regression model: age, sex, comorbidities, and the results of seven commonly available laboratory tests. The end point was death within 14 days in or out of the hospital.

RESULTS

Of 901 patients in the development cohort and 45% died within 14 days. The risk of death within 14 days after admission increased with increasing age, lactate dehydrogenase levels, and white blood cell counts and decreasing albumin levels and platelet counts (P < 0.01). The model predictions were confirmed using a separate validation cohort. The areas under the receiver operating characteristic curves were 0.74 and 0.70 for the development and validation cohorts, respectively, indicating good discriminatory ability for the model.

CONCLUSIONS

Our results suggest that laboratory test results performed within two days of admission are valuable in predicting death within 14 days for patients with metastatic or advanced cancer. Such results may provide an objective assessment tool for physicians and help them initiate conversations with patients and families about end-of-life care.

A literature review of randomized controlled trials of the organization of care at the end of life

We searched nine electronic databases for randomized controlled trials (RCTs) about care at the end of life and found 23 RCTs. We assessed their quality using the criteria of the Cochrane Collaboration. The RCTs researched three themes: (a) the effect of providing palliative care through dedicated community teams on quality of life, on the management of symptoms, on satisfaction with care, on the duration of the palliative period, and on place of death; (b) the effects of specific palliative care interventions-advanced planning of care for the end of life, patient-held records, providing quality-of-life data to patients and physicians, grief education for relatives, palliative care education for nurses, and palliative care for patients with dementia; and (c) the costs of palliative compared to conventional care. We identify difficulties in conducting research on palliative care and solutions and discuss future possible research themes.

Use of the palliative outcome scale in Argentina: a cross-cultural adaptation and validation study

Although measuring outcomes is essential to ensuring palliative care effectiveness, there is an absence of properly validated measures in many countries. We undertook a cross-cultural adaptation and validation of the Palliative Outcome Scale (POS) into a Spanish (Argentina) language and cultural context. The methodology used a sequence of phases: 1) verification of conceptual equivalence (literature review, professional interviews, and patient focus groups); 2) multiple translations; 3) committee review; and 4) field testing. Psychometric analysis entailed evaluation of quantitative content validity, construct validity, staff and patients' ratings comparison, internal consistency, test-retest reliability, and responsiveness to change. Conceptual equivalence was achieved. Multiple changes were introduced after the translations and field testing in 65 patients and 20 professionals. Content validity was high for all but one item. Construct validity against a validated quality-of-life measure (European Organization for Research and Treatment of Cancer Quality of Life C-30) was confirmed (rho=0.74, P<0.0005). There was acceptable agreement between staff and patients (Cohen's weighted kappa >0.3) for 5/10, 8/10, and 6/9 items at each of three time-point evaluations and good correlation for all but one item (Spearman coefficient >0.7). Internal consistency was acceptable (Cronbach's alpha=0.68-0.69 and 0.66-0.73) for patient and staff ratings, respectively, and test-retest reliability showed very high agreement for every item (>0.80). The Argentine POS showed adequate responsiveness to change, although significant difference was reached for only 3 out of 10 items for patients and staff, respectively. Completion of the POS did not take more than 12 and 6 minutes for patients and staff, respectively. This study indicates that the Argentine POS is a valid and reliable measure of palliative care outcomes with advanced cancer patients.