Research article

Validity of measures of pain and symptoms in HIV/AIDS infected households in resources poor settings: results from the Dominican Republic and Cambodia

Gregory Pappas^{* 1} , R Cameron Wolf^{* 2} , Guy Morineau^{* 3} and Richard Harding^{* 4}

¹Department of Community Health Sciences, Aga Khan University, PO Box 3500 Stadium Road Karachi, Pakistan

²United States Agency for International Development, Office of HIV/AIDS, Washington, DC, USA

³Family Health International, Cambodia

⁴Dept of Palliative Care & Policy, Guy's King's & St. Thomas' School of Medicine, London, UK

author email corresponding author email* Contributed equally

BMC Palliative Care 2006, 5:3doi:10.1186/1472-684X-5-3

Published:	20 March 2006

Abstract

Background

HIV/AIDS treatment programs are currently being mounted in many developing nations that include palliative care services. While measures of palliative care have been developed and validated for resource rich settings, very little work exists to support an understanding of measurement for Africa, Latin America or Asia.

Methods

This study investigates the construct validity of measures of reported pain, pain control, symptoms and symptom control in areas with high HIV-infected prevalence in Dominican Republic and Cambodia Measures were adapted from the POS (Palliative Outcome Scale). Households were selected through purposive sampling from networks of people living with HIV/AIDS. Consistencies in patterns in the data were tested used Chi Square and Mantel Haenszel tests.

Results

The sample persons who reported chronic illness were much more likely to report pain and symptoms compared to those not chronically ill. When controlling for the degrees of pain, pain control did not differ between the chronically ill and non-chronically ill using a Mantel Haenszel test in both countries. Similar results were found for reported symptoms and symptom control for the Dominican Republic. These findings broadly support the construct validity of an adapted version of the POS in these two less developed countries.

Conclusion

The results of the study suggest that the selected measures can usefully be incorporated into population-based surveys and evaluation tools needed to monitor palliative care and used in settings with high HIV/AIDS prevalence.