By Sydney Rosen, Research Professor, Department of Global Health, Boston University, and Co-Director of the Health Economics and Epidemiology Research Office (HE2RO), University of the Witwatersrand, South Africa
Among other anticipated benefits, differentiated service delivery (DSD) models for HIV treatment are expected to reduce the cost of service delivery to both the health system and the patient, especially for the large number of clinically stable patients in most national antiretroviral therapy (ART) programmes. This expectation follows logically from the notion that most DSD models are designed to be “less intensive” than conventional care and, therefore, presumably utilize fewer resources per patient served. Until recently, though, little empirical evidence on DSD model costs was available.
Weighing up costs
Now several newly published studies estimate the costs of DSD models implemented routinely (that is, not as pilot or demonstration projects or research studies), allowing us to compare the true costs and outcomes of DSD with conventional, facility-based care in a number of settings in sub-Saharan Africa. These studies include cluster-randomized trials and observational evaluations in Lesotho, Malawi , South Africa, Uganda, Zambia, and Zimbabwe. A summary brief of this evidence is available here.
Each of the studies reported 12-month retention-in-care rates. The costs of ART to the health system, usually including medications, clinic visits and DSD-specific interactions, laboratory tests, and infrastructure and other fixed costs, were estimated per patient and per patientretained in care 12 months after study enrolment. Most of the patients enrolled in the studies met national criteria for being clinically stable on ART; all were adults. In some cases, the cost per patient per year for treatment in the DSD model was compared with the cost of conventional care; in other cases, different DSD models were compared with one another. Some of the studies also estimated costs to patients themselves.
6MMD consistently less expensive
In most, but not all, cases, DSD models achieved roughly the same 12-month retention rates as did conventional care, or were reported as non-inferior to conventional care. Some models did slightly better than conventional care in terms of retention; few did worse. The cost to the health system per patient treated under each model varied, with some models more expensive than conventional care and others less expensive.
In general, the cost differences between models per patient were modest, mainly reflecting the large share of costs attributable to antiretroviral medications, whose cost does not vary with model of care. Several of the models proved not to be “less intensive” than conventional care. Moreover, in these models, resource allocation shifted from the clinic to the DSD model but did not diminish. Models that incorporated six-month medication dispensing (6MMD), however, were consistently less expensive than conventional care.
What it means for patients
More striking than the cost implications to health systems were the sharp falls in costs to patients themselves. In studies that surveyed patients about the costs they incurred for obtaining treatment, including transport fares and the value of the time spent, patients reported cutting their own out-of-pocket and/or opportunity cost expenditures by between a quarter and a half per year, generally due to the reduced number of full clinic visits required by DSD models. Some models, such as adherence clubs, did not reduce time costs substantially, but did save patients money for transport.
Even a small cost reduction to the health system per patient treated may be important in view of the very large numbers of patients enrolled in national ART programmes in many countries. Savings to patients may be even more important if they help improve long-term retention in care while making ART less of an economic burden.
Further research needed
An important limitation to all of the studies reported here is that costs are limited to patients who are stable on ART and therefore eligible for DSD models. These patients are likely to cost the health system less than average even when in conventional care; shifting them to DSD models may leave the more expensive patients in conventional care, simply reallocating overall health system costs, rather than reducing them. Future research is needed to understand the role of DSD models in improving health outcomes and lowering costs of ART programmes as a whole. Research is also needed to explore the integration of DSD models for HIV treatment with service delivery for other chronic needs, to optimize clinic efficiency and minimize the overall burden on patients to access healthcare.
 In press.