Homeless adults who were offered a small cash incentive to undergo preventive therapy for tuberculosis completed the treatment at nearly double the rate of others, according to new research by a University of California, San Francisco team.
All study participants were from San Francisco and tested positive to a tuberculin skin test, indicating they had been exposed to TB. Participants did not yet have active contagious TB, and preventive therapy would keep them from getting sicker and becoming contagious.
The cash incentive was $5 per visit for therapy, which consisted of the antibiotic isoniazid, known as INH, administered in pill form on a regular basis over a six-month period.
"Our study findings indicate that there is a role for appropriate monetary incentives in TB control. From previous studies we know that the U.S. homeless population is at high risk for TB infection and that adherence to preventive therapy is difficult within this group," said Jacqueline Peterson Tulsky, MD, lead investigator of the research team and UCSF associate clinical professor of medicine.
Research results are reported in this week's issue (March 13) of the Archives of Internal Medicine. A specialist in TB and HIV/AIDS care, Tulsky treats patients at San Francisco General Hospital Medical Center.
INH is a well-established tool for TB control, and previous research has shown that adherence is higher-even in hard to treat populations with active TB--when patients are observed taking each dose of preventive medication, rather than when they take it solely on their own. The approach is known as directly observed preventive therapy, or DOPT.
The study involved 118 persons who were randomized into one of three groups:
--DOPT with monetary incentive: Medication was administered by staff at SFGHMC twice a week, with the cash award given at each visit.
--DOPT with a "peer health adviser": Medication was administered by a peer adviser-a homeless person specifically trained for this project-twice a week at any agreed upon location.
--Usual care: Persons in this group, which served as the control, received a month's supply of medication through the TB Clinic at SFGHMC along with instructions from the staff for self-administering daily doses. Participants were considered to be adherent to therapy if they picked up their monthly refill of medication.
Study results showed that 44 percent of participants in the monetary incentive DOPT arm completed the six-month INH therapy. This compared with a completion rate of 19 percent in the peer health adviser DOPT group and 26 percent in the usual care group.
The research is one of only a few randomized studies involving the homeless that compares the use of incentives for clinical treatment with a control group, according to Tulsky.
The low rate of adherence in the peer health adviser arm was a surprise to the research team, Tulsky said, because this strategy was developed with the idea that peer outreach workers have been used in other health care activities successfully and the homeless peers seemed well-suited to understand the problems of the homeless participants. Followup discussion found that many of the advisers felt they lacked some of the communication and life skills necessary for effectively dealing with study participants.
"From this finding, we realized that although it was an appealing idea to employ homeless adults to work with their peers, we need to develop a more extensive preparatory program--combining both support and training--in order for it to be effective." Tulsky said.
Participants were recruited for the study and screened for TB and HIV through a community project directed by Andrew Moss, PhD, of the UCSF Departments of Epidemiology & Biostatistics and Medicine at SFGHMC.
All study participants were considered "true homeless," defined as persons staying in emergency shelters, on the street, in a car or van, or in a park or other outdoor public space not designated for sleeping, or "marginally homeless," defined as indigent residents of low-cost residential hotels, often called single-room occupancy or SRO hotels.
Another variable that predicted adherence was housing status, the researchers found. Persons who lived in low-cost hotels had a higher completion rate in all three study arms. "Stabilized housing is often considered a key requirement for providing medical care in the homeless and our study supports this,' they noted.
The distance to the TB clinic at SFGHMC also may have had a negative effect on adherence, according to the researchers. The downtown location of most low-cost hotels and shelters and other neighborhoods where the homeless spend most of their time were about a 30-minute walk to the hospital. Another site for DOPT might improve adherence rates, the researchers speculate.
There were no significant differences in the demographics of the three groups. Overall, about 30 percent had completed college and 31 percent were veterans. Over 30 percent had a lifetime history of intravenous drug use, 64 percent had used crack cocaine, and 56 percent had a history of alcoholism. Current use of drugs and alcohol was high.
Some health care providers feel that it is inappropriate to give money to persons with known alcohol or drug addictions, Tulsky noted, but the goal of this study was to openly review the role of different types of clinical approaches for preventive TB therapy. "From our findings, we feel that a $5 incentive provides a positive effect at both the public health and personal levels and should be considered by jurisdictions with large numbers of TB cases among their homeless."
The research was supported by grants from the National Institute of Drug Abuse, the UC Universitywide AIDS Research Program, and the Kaiser Family Foundation.
In addition to Tulsky and Moss, the study team included Louise Pilote, MD, MPH, PhD, Montreal General Hospital/McGill University; Judith Hahn, MA, UCSF Department of Epidemiology & Biostatistics; Michele Burke, BS, RN, UCSF School of Nursing; Margaret Chesney, PhD, UCSF Center for AIDS Prevention Studies; and Andrew Zolopa, MD, Stanford University.
Journal
Archives of Internal Medicine