Kenyan study shows people with HIV can provide safe, effective community management of ART

Community-based care delivered to adults living with HIV by people living with HIV using mobile technologies provided care as safe and effective as clinic-based care, researchers report in the advance online edition of the Journal of Acquired Immune Deficiency Syndromes.

This prospective community randomised clinical pilot study was conducted in villages surrounding a rural clinic in western Kenya from March 2006 to April 2008.

While the benefits of antiretroviral treatment in resource-poor settings are known and well-described, scale-up of antiretroviral treatment is slow. Of the estimated 9.5 million adults living with HIV in need of ART in low- and middle-income countries over 5 million are still without treatment.

Task-shifting is one example of a strategy developed to help resolve the lack of human and financial resources needed to facilitate scale-up. The World Health Organization guidelines advocate for task-shifting from physicians and nurses to community health workers including people living with HIV to provide HIV services at the community level. Community health workers take on some responsibilities usually assigned to physicians or nurses. 

Limited evidence exists about the use of mobile health technology (mobile phones, personal digital assistants) as an effective and cost-efficient component of task-shifting to help people living with HIV provide health care in resource-poor settings.

The authors evaluated how task-shifting affected the clinical outcomes of HIV-infected adults enrolled in an innovative HIV care delivery system where people living with HIV were engaged as community care co-ordinators (CCC).

The CCCs had secondary school education and used an electronic decision support tool (personal digital assistant) to deliver medications and provide follow-up care to patients on antiretrovirals in the community.

This prospective community randomised clinical pilot study was integrated into an extensive HIV/AIDS care network. Begun in 2001 the United States Aid for International Development and the Academic Model for Providing Access to Health care programs (USAID-AMPATH) partnership currently manages 23 main clinics and 23 satellite clinics in western Kenya.

The setting for the pilot study was the HIV clinic and the community surrounding the Mosoriot Rural Health Clinic. The HIV clinic serves the Kosirai Division, a community of 60,000 in a province with an estimated HIV prevalence of 7.4%.

Kosirai Division has 24 geographic and administrative areas called sub-locations. Each sub-location is approximately 4 kilometres in diameter and can be crossed by foot in one to two hours. Ideal for individual CCCs to manage, the sub-location was chosen as the unit of randomisation.

Eligibility for the study included enrollment at the Mosoriot HIV clinic, aged 18 years or above, being clinically stable on antiretroviral treatment for at least three months with no adherence issues, having household members aware of the patient’s HIV status, living in Kosirai Division and having agreed to participate.

The standard of care model in use at the time of the study involved monthly clinic visits which meant contact with at least three health care providers (nurse, clinical officer and pharmacist). Transportation and wait times resulted in an expensive and time-consuming visit for the patient.

Criteria for recruitment of community care co-ordinators (CCCs) included: being part of the HIV clinic population at Mosoriot, having a secondary-level education, being clinically stable with self-reported 100% adherence over the preceding six months and considered by clinical staff to be a good role model and mentor.

CCC training included the use of a personal digital assistant (PDA). The PDA was pre-programmed to collect a symptom review, vital signs, adherence, food security and domestic violence.

The PDA also had a built-in decision support programme with pre-programmed alerts

• To return the next day to the clinic for patient re-evaluation
• Transport the patient to the clinic for urgent evaluation, or
• Call the clinical officer for consultation.

This support programme was critical to the successful co-ordination of care, the authors stress.  This allowed the CCCs to focus on the collection of symptoms and signs and undertake interventions as needed and for medical decisions to be made by clinical officers or physicians. With or without alerts CCCs were able to call medical officers for advice.

A two-month clinical mentorship followed the training.

The intervention group had monthly PDA supported home assessments by people living with HIV (community care co-ordinators) and clinic appointments every three months. The control group had standard of care monthly clinic visits.

In addition each patient had a research visit co-scheduled with the clinic visit.  Assessment of WHO stage, Karnofsky score (an assessment of an individual's health and wellbeing, based on a performance index of physical ability http://www.aidsmap.com/page/1277865/ ), antiretroviral adherence history, herbal medication use as well as opportunistic infections were made at enrollment and then every three months until the final visit at 12 months.  At enrollment and the final visit HIV viral load and CD4 cell counts were recorded, with an additional CD4 cell count at six months. 

At the end of one year no significant differences in clinical or laboratory outcomes were seen between the intervention group (87 patients) and the control group (102 patients). Both groups showed high levels of self-reported adherence.

The authors note, to their knowledge, this is the first randomised control study to look at the efficacy of using HIV-infected individuals with secondary-school level education to provide antiretrovirals and monitor HIV therapy. A similar published study in Uganda with similar findings had one significant difference. The lay workers providing home-based care all had college degrees or diplomas.

“In rural areas the ability to decentralise HIV care to lay persons, the majority of whom will only have a secondary education, is of critical importance.” note the authors.

The intervention group (community care co-ordinators), however, had significantly fewer clinic visits compared to the control group (6.2 and 12.4, p<0.001). This study showed that task-shifting decreased the number of clinic visits, so decongesting the clinic and allowing for more patients to get care with fixed resources, note the authors.

The CCC group were in a better position to identify issues that affected HIV care, for example food insecurity, domestic violence, alcohol abuse and disclosure issues.

As with community health workers (accompagnateurs) in the Partners in Health programme in Haiti, CCCs not only served as active and valued members of the health care team but provided “social linkages and support for their patients” note the authors.

The authors found higher than anticipated rates of unplanned pregnancies in both the intervention and control groups. They suggest that future home-based care programmes incorporate reproductive health issues including family planning and pregnancy.

Limitations of the study, note the authors, include

• The small sample size
• Participants had to be stable on antiretrovirals for at least three months before enrollment so the results cannot be generalised to those just starting ART
• Karnofsky score has never been validated in Africa
• The study was not designed to assess the effect of the intervention on adherence.

The main strength, the authors stress, is the use of a community randomised design which allows for real-time comparison of the control and intervention groups.

The authors conclude “this pilot study suggests that task-shifting and mobile technologies can deliver safe and effective community-based care to people living with HIV, expediting roll-out and increasing access to treatment while expanding the capacity of health care institutions in resource-constrained environments.” However, they add “larger scale studies will be needed to support our findings.”