Factors influencing hospice thromboprophylaxis policy: a qualitative study.

Despite level 1 evidence supporting the use of low-molecular weight heparin thromboprophylaxis in hospitalised cancer patients, only 7% of specialist palliative care units (SCPU) have thromboprophylaxis guidelines. The reasons for this are unclear. To explore specialist palliative care units (SPCU) directors' views on thromboprophylaxis in the inpatient unit, audiotaped semi-structured interviews were conducted with SCPU medical directors to explore factors influencing thromboprophylaxis practice. Purposive sampling of units known not to have thromboprophylaxis guidelines was conducted (as identified from previous research). The hospice directory was used to sample from units in each region of Great Britain and Ireland to ensure representation across the specialty. Interviews were transcribed and analysed using interpretative phenomenological analysis (IPA). Four major and four sub themes were identified. Participants were progressive in their attitudes to palliative care and comfortable with instigating active interventions for patient benefit. Symptomatic venous thromboembolism (VTE) was rarely seen and therefore not considered important enough to warrant guidelines. There was concern that evidence informing thromboprophylaxis guidelines in the general population was not transferable to the advanced cancer population and that the outcome measures from these studies were less meaningful to a palliative care patient. Thromboprophylaxis was considered a life prolonging intervention which may result in a poorer death than one because of VTE. Nevertheless, participants were receptive to change if presented with convincing evidence derived from a representative population. Until the true prevalence and symptomatic burden of VTE is known, the role of thromboprophylaxis in the SPCU setting will remain controversial. There is a need for a well-designed study to explore the utility of thromboprophylaxis in the palliative care inpatient setting. However, this will require meaningful outcome measures to be used within a clinically applicable population.