Background. To analyze how seven Canadian hospice palliative care (HPC) centres and one national surveillance dataset compare with respect to the collection of forty data elements. Research and service delivery implications of the findings are discussed. Methods. The data sources consisted of data element names and their definitions collected in a computer based format from seven HPC centres and one surveillance dataset. The data elements were structured into five themes: demographic, patient death, support, contact or informal caregiver; program/consultations/service request, and clinical. Each theme contains a number of data elements with a total of 40 elements included in the analysis. Comparative analysis was done on the data elements to compare their names and definitions. Results. Much variation exists in data collection around HPC delivery. Such variation prevents any timely and meaningful comparison of service and care delivery across HPC centres. Patient death data, service/program data and clinical data is particularly varied. Conclusion. Developing a common minimum data set is a logical starting point to help overcome data variations between care centres. Greater coordination is needed between care centres and the development of national standards and policies. Moving towards electronic data collection would help facilitate common policy and practice norms.