A Framework for Generalizability in Palliative Care

Abstract

Palliative medicine has only recently joined the ranks of evidence-based medical subspecialties. Palliative medicine is a rapidly evolving field, which is quickly moving to redress its historical paucity of high-quality research evidence. This burgeoning evidence base can help support the application of evidence-based principles in palliative and hospice clinical care and service delivery. New knowledge is generally taken into practice relatively slowly by established practitioners. At present, the translation of evidence into palliative and hospice care clinical practice lags behind emerging research evidence in palliative care at even greater rates for three critical reasons: 1) the application of research results to specific clinical subpopulations is complicated by the heterogeneity of palliative care study subpopulations and by the lack of a recognized schema for describing populations or services; 2) definitional issues in service provision are, at best, confusing; and 3) fundamental research concepts (e.g., external validity, effect size, generalizability, applicability) are difficult to apply meaningfully in palliative care. This article provides a suggested framework for classifying palliative care research subpopulations and the clinical subpopulations to which the research findings are being applied to improve the ability of clinicians, health planners, and funders to interpret and apply palliative care research in real-world settings. The framework has five domains: patients and caregivers; health professionals; service issues; health and social policy; and research.

Refining a Checklist for Reporting Patient Populations and Service Characteristics in Hospice and Palliative Care Research

“…The pattern of data provision in the most and least frequently reported sub-domains was examined in relation to whether the checklist [5] labelled them "core" or "noncore." Only one non-core item (primary data source) appeared in the 10 subdomains that were most frequently reported.…”

“…All papers were coded by one investigator (AG) using the framework. [5] One in ten papers was coded independently by a second investigator (DC). Any discrepancies were settled by consensus.…”

“…These divergences can create very different sub-populations of study participants, even in rigorously designed and executed randomised controlled trials, [3,4] and can limit the generalizability of reported evidence. [5] Palliative care is not unique in struggling to report key elements of study design, so as to facilitate the application of research findings in practice. In oncology, a recent review of randomized, controlled trials of best supportive care interventions found that only 11% of such studies published from 2005 to 2008 reported all data elements essential to translation of oncology clinical trial data into clinical practice.…”

“…[16] To complement the checklists above, given the variations in its populations and service models, palliative care research needs a discipline-specific checklist to evaluate external validity. This research group previously proposed a supportive and palliative care discipline-specific checklist; [5] the checklist comprises a small number of easily documented and understood measures, selected to reflect key variation between service delivery models and the populations referred to those clinical services. By assessing applicability based on service and patient population descriptors, use of this checklist should help in describing and comparing supportive and palliative care studies in the future, and should allow more rapid and appropriate uptake of new evidence into practice.…”

Palliative Cancer Care Research

“…This approach is considered to improve quality of care through a critical implementation of best practice to the right patient populations, at the right time and with the right amount and intensity. 2 There are some fundamental concepts that need to be fulfilled in order to reach the goal of EBP in palliative care. First of all, the scientific evidence to develop clinical guidelines needs to be available with an appropriate quality and for the target palliative care population.…”