This article was downloaded by: [UZH Hauptbibliothek / Zentralbibliothek Zürich] On: 23 December 2014, At: 13:19 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
African Journal of AIDS Research Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/raar20
Rationing access to public-sector antiretroviral treatment during scale-up in South Africa: implications for equity Nandipha Jacobs , Helen Schneider & H.C.J. Van Rensburg Published online: 11 Nov 2009.
To cite this article: Nandipha Jacobs , Helen Schneider & H.C.J. Van Rensburg (2008) Rationing access to public-sector antiretroviral treatment during scale-up in South Africa: implications for equity, African Journal of AIDS Research, 7:1, 19-27, DOI: 10.2989/AJAR.2008.7.1.3.431 To link to this article: http://dx.doi.org/10.2989/AJAR.2008.7.1.3.431
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Copyright © NISC Pty Ltd
African Journal of AIDS Research 2008, 7(1): 19–27 Printed in South Africa — All rights reserved
AJAR
EISSN 1727–9445 doi: 10.2989/AJAR.2008.7.1.3.431
Rationing access to public-sector antiretroviral treatment during scale-up in South Africa: implications for equity Nandipha Jacobs1*, Helen Schneider1,2 and HCJ Van Rensburg2 Centre for Health Systems Research & Development, University of the Free State, PO Box 339, Bloemfontein 9300, South Africa 2 Centre for Health Policy, University of the Witwatersrand, PO Box 1038, Johannesburg 2000, South Africa * Corresponding author, e-mail: [email protected]
Downloaded by [UZH Hauptbibliothek / Zentralbibliothek Zürich] at 13:19 23 December 2014
1
As with other large-scale antiretroviral therapy (ART) programmes initiated in developing countries, rationing of ART in South Africa was inevitable in the face of limited human and infrastructural resources. This paper discusses rationing of ART and its implications for equity in South Africa, with a particular emphasis on the national programme in the Free State Province. We examine both the explicit and implicit dimensions of ART rationing, while considering the content of policy, programme design and implementation processes. The explicit rationing criteria included in the national comprehensive plan, as well as the associated treatment guidelines, direct and influence rationing decisions practised locally. However, implicit processes of rationing, which limit demand, are also wittingly and unwittingly introduced in several ways, and at all levels of the health service. Consequences of rationing manifest in the high numbers of patients lost to the system and the difficulties faced by the most impoverished clients in gaining access to ART services on an ongoing basis. Not losing sight of the equity element within rationing is important, and this aspect will have to be formally monitored for eventual success of the country’s ART programme. ART roll-out processes also require ongoing efforts to build consensus and confidence in rationing, accomplished with a view to optimal fairness. Keywords: accessibility, Free State Province, HAART, patient selection criteria, programme implementation
Introduction Many African countries have begun to develop and implement large-scale highly active antiretroviral therapy (HAART) programmes amid limited human and infrastructural resources. As a result of these limited resources, access to HAART for AIDS illness is being rationed in many developing countries (Loewenson, 2003; McCoy, 2003; Stewart, Paradath & Milford, 2006). Rationing is “a process of distributing scarce services within a population when it is not possible to provide each patient with every health care service appropriate to his/her medical need” (Krízová & Simek, 2002, p. 308). This is often the case when the demand for a service or a commodity outstrips its supply. In such cases rationing strategies are required in order to manage the healthcare delivery process efficiently. The literature divides rationing strategies into explicit (stated) and implicit (unstated) arrangements (Krízová & Simek, 2002; Rosen, Sanne, Collier & Simon, 2004a; Bennett & Chanfreau, 2005): “Explicit rationing occurs when defined and widely understood criteria such as age-related or insurance-related criteria are used to determine access,” and it is more likely to be developed as part of national policies (Bennett & Chanfreau, 2005, p. 542). Examples of explicit rationing for HAART include citizenship requirements, defining certain age-groups or geographical areas for priority access, psychosocial criteria (such as proof of adherence) or requiring co-payments
for care (Rosen, Sanne, Collier & Simon, 2005). Implicit rationing, on the other hand, “lacks any overarching plan or clearly defined criteria, but rather depends on subtle decisions, many of which are made by health care providers” (Bennett & Chanfreau, 2005, p. 542). Examples of implicit rationing include limitations on access to underlying healthcare infrastructure to deliver HAART, scheduling patients on a first-come-first-serve basis, appointment systems, and using waiting lists or queuing systems, the latter being particularly susceptible to local or provider influences. Rationing has implications for both the effectiveness and equity of HAART programmes. For example, prioritising mothers of young children (as opposed to older women or men) for treatment would increase the societal benefits (effectiveness) of a HAART programme by reducing the number of orphans and vulnerable children (Loeweson, 2003; McCoy, 2003). Similarly, treating only skilled workers may have macro-economic benefits, but would be regarded as highly inequitable in that it excludes poor and unemployed populations (Rosen et al., 2005). In general, without clear criteria or an explicit strategy for the rationing of treatment, access to treatment will tend towards reflecting underlying socio-economic inequities and social biases (Loewenson, 2003). If they are to be fair or equitable, therefore, rationing strategies used in treatment
Downloaded by [UZH Hauptbibliothek / Zentralbibliothek Zürich] at 13:19 23 December 2014
20
programmes require clearly defined and transparent procedures within an overarching process of reducing unfair social and geographic healthcare disparities (Loewenson, 2003; McCoy, 2003; Loewenson & McCoy, 2004; Rosen, Sanne, Collier & Simon, 2004b). In November 2003, the Comprehensive Plan for Comprehensive HIV and AIDS Care, Management and Treatment for South Africa (National Department of Health [NDoH], 2003a) (henceforth, the ‘Comprehensive Plan’) was adopted with the purpose of providing a national framework for expanded treatment for people with HIV or AIDS. As with the WHO/UNAIDS ‘3 by 5’ initiative to extend antiretroviral treatment to 3 million people by 2005 (World Health Organization [WHO], 2003), universal access for clinically eligible people is an explicit policy objective of the Comprehensive Plan (Loewenson, 2003; NDoH, 2003a). “The South African constitution and government require that implementation of a programme of such importance be carried out in a universal and equitable manner” (NDoH, 2003a, p. 18). Universal access (thus, also equitable implementation of the plan) to care and treatment — irrespective of race, gender and economic status — is one of the core explicit guiding principles of the Comprehensive Plan, and accordingly, on the surface, access to ART in South Africa requires no rationing. Moreover, Cleary, Chitha, Jikwana, Okorafar & Boulle (2005) show that government conditional grants, at least for the first few years of programme implementation, would be sufficient to deliver ART nationally. In practice, however, it is widely recognised that human and infrastructural resource constraints will prevent a match between supply of and demand for (and even less the need for) ART for some years to come. In 2005, the projected population of the Free State Province was approximately 2.8 million, including 381 860 HIV-infected people, of whom 40 706 were eligible for ART (WHO stage-IV disease) and 8 177 were urgently in need of ART (Actuarial Society of South Africa [ASSA], 2005). By October 2005, just over a year after commencement of the ART programme in the Free State, only 3 322 (8%) of clinically eligible people were receiving ART (Free State Department of Health [FSDoH], 2005). In this context, rationing appeared to have played an obvious part. Against the backdrop of the gap between supply and demand of ART, Stewart et al. (2006) raised concerns that the provisioning of ART will exaggerate existing inequities in resource allocation and service provision, thereby undermining the very system that the Comprehensive Plan was supposed to strengthen through the introduction of ART. They also pointed to emerging threats to achieving the equity goals in the implementation of the ART programme, and in particular equity among provinces, gender groups, urban and rural areas, and paediatric versus adult ART access. This paper analyses rationing of ART in South Africa, and the implications of such rationing for equity. The focus is on the Free State Province, one of the nine provinces of South Africa. We examine both the explicit and implicit dimensions of rationing ART, while considering the content of policy,
Jacobs, Schneider and Van Rensburg
programme design, and implementation processes during the first years of ART roll-out in the public sector. Methods and data sources The paper draws on secondary information from a comprehensive search of electronic databases, in an effort to identify policy and programme texts that specifically examine issues of rationing, patient selection and ART. These sources include national and provincial policies, treatment guidelines and scientific papers. Empirically, the paper is based on a series of data-gathering initiatives conducted by The Centre for Health Systems Research and Development for a research project entitled, ‘Documenting, monitoring, evaluating and facilitating implementation of the national Treatment Plan in the Free State, South Africa.’ We have collated primary data from individual interviews with high-ranking and mid-level officials of the Free State Department of Health (FSDoH) and medical specialists serving on the provincial ART Task Team (TT), as well as from notes of discussions taken at provincial meetings related to ART implementation, expansion and management. From these, direct citations are provided to illustrate, emphasise and substantiate findings on forms of rationing.† In addition, longitudinal data generated by ongoing appraisals of facilities providing ART in the Free State Province are used, including individual and group interviews with health workers. Guided by the availability and relevance of information, various sources of information are triangulated. Moreover, extensive discussion and agreement regarding the findings, our emphasis and interpretation has transpired between two of the authors (NJ and HCJ VR) who actually participated widely in data-gathering and in the research team. The paper describes the forms of rationing emerging and evident in the Free State public-sector ART programme, and examines both policy and practice contexts. Finally, the consequences of ART rationing strategies for service delivery, access and equity are discussed. ART rationing and its implementation in the Free State Province Phased geographical access An obvious way of limiting access to treatment is to offer ART only to patients residing in specified geographic catchment areas (McCoy, 2003; Rosen et al., 2004b; McGough, Reynold, Quinn & Zenilman, 2005). Ntuli, Ijumba, McCoy, Paradath & Berthiaume (2003) argued that most sub-Saharan countries would scale up treatment programmes incrementally through first offering ART at a few facilities and then adding more over time. This process was evidently also considered necessary to achieving the goal of equitable ART access in South Africa (NDoH, 2003a; Doherty, Loveday, Stewart & Thomas, 2005). As stated in the Comprehensive Plan (NDoH, 2003a), the ART programme was to be phased in over a five-year period, with the aim of achieving universal coverage of new
† Statements or remarks made during meetings of the provincial task team (TT), or in other official meetings related to the ART programme, are anonymously referred to as ‘FSDoH official’ or ‘FSDoH clinician’, TT meeting, with the date (day/month/year) of the statement or remark.
Downloaded by [UZH Hauptbibliothek / Zentralbibliothek Zürich] at 13:19 23 December 2014
African Journal of AIDS Research 2008, 7(1): 19–27
AIDS cases by the end of the 2008/09 financial year. The NDoH anticipated that by 2004 every health district in the country would have at least one service point,1 and within five years these service points would be expanded to every local municipality. In the Free State, one service point in each of the five districts was envisaged within the first two years, and thereafter a steady expansion to all appropriate facilities (20 local municipal areas in total) (FSDoH, 2003). The phased approach was endorsed by the Joint Health and Treasury Task Team, which viewed it as “acceptable as long as the state worked towards the realisation of a programme to which everyone in need will ultimately have access” (NDoH, 2003b, p. 20). In the early stages, one directive in the Free State, with strong tones of exclusion and rationing, was to ‘make sure that only patients in the catchment areas of an ARV site come to the site’ (TT meeting, 13.10.04). The obvious rationing strategy here is that the majority of those patients not living within those few initial catchment areas would be excluded unless they were willing to relocate (Doherty et al., 2005). Nationally, the ART programme was seeking to achieve some kind of geographical equity regardless of differences in capacity between provinces and districts. However, the approach was a fairly crude one in that it did not include specific considerations of need, such as the geographic concentration of the HIV epidemic, population size and density, thus potentially negatively affecting access and equity. In the Free State, ART sites were allocated equally to the five districts, irrespective of the important geographic and demographic differentials characterising the epidemic in the province. Population sizes (and density) range from a high of 736 292 (52.6/km2) in Motheo District to a low of 132 070 (3.86/km2) in Xhariep District, with sub-districts showing even greater extremes, ranging from population sizes of 654 922 to 38 604, and population densities of 104.24/km2 to 3.55/km2 (Barron & Asia, 2001, pp. 30–31). In effect, this means that although there was an equal spread of facilities providing ART (all equally staffed), and thus supply of ART per district — a strikingly inequitable spread of service-provisioning in respect of patient numbers, need, and demand was created. Selection and accreditation of ART sites Ensuring the presence of essential staffing and infrastructure for delivering a quality ART service is central to the success of the programme. This has been implemented through a rigorous process of selection and accreditation of sites. The NDoH was centrally involved in the accreditation of ART sites nationwide, setting particular norms and standards that sites had to meet before providing ART to the public (NDoH, 2003a). The identification and selection of potential ART sites in the Free State commenced in early 2003, driven by the provincial ARV Task Team and with the involvement of district managerial structures (FSDoH, 2003). Qualifying facilities had to have the appropriate level of staffing, with adequately functioning and efficient support services, specifically drug procurement and distribution services, and sufficient and suitable physical space to accommodate the programme (FSDoH, 2003). As one FSDoH official observed, ‘The most important criterion was
21
to find appropriate space’ (FSDoH official, TT meeting, 12.01.04). By December 2003, potential ART sites in all five districts had been provisionally assigned and their accreditation was pending (TT meeting, 01.12.03). Eventually, important changes were effected regarding the final site selection in some of the districts. Van Rensburg (2005 and 2006) pointed out that the site selection for the first phase of the ART programme implementation in the Free State was very much a top-down, management-directed approach with little consultation with facility governing structures or communities. Decisions were rushed both managerially and politically, and neither equitable distribution nor optimal geographic access appeared to have played an overt role in decisions on ART service sites. It is not clear whether other more strategic considerations, such as the capacity of facilities to scale up rapidly or to provide a good focus of learning, were part of decision-making, although the perceived accreditation requirements naturally promoted those areas and facilities within the districts with better buildings and infrastructure. The accreditation process was handled by a joint team of officials from the NDoH and FSDoH. Four service points (i.e. one each in the Lejweleputswa, Motheo, Thabo Mofutsanyane and Fezile Dabi districts) consisted of a referral treatment site serviced by three referring assessment sites — the ‘1 x 3’ model (Van Rensburg, 2006). Clinics and community health centres (CHCs) serve as the primary sites for entry to the service for the diagnosis, staging and routine follow-up of ART patients (NDoH, 2003a, pp. 60–61). Patients who meet the criteria for ART are referred to the district or regional hospital that serve as the treatment site, where detailed assessments are required and a medical practitioner, in consultation with other staff, decide whether the patient will commence with ART (FSDoH, 2003).2 This model allowed for some decentralisation of decision-making, with the NDoH team involved in conducting the accreditation of treatment sites, while the FSDoH team was responsible for accrediting the assessment sites (TT meeting, 08.12.03). Potential ART sites that did not meet the accreditation standards at the time of the first visit had to be strengthened, and financial and technical resources were mobilised for such purposes (NDoH, 2003a). Accreditation feedback sessions took place for purposes of devising detailed strategies for strengthening sites that did not meet the accreditation standards. Only accredited sites would provide ART treatment, as emphasised by a member of the NDoH team: ‘If you are not accredited, you cannot start treatment — if it is green, it is green, not a tint of green’ (NDoH official, accreditation feedback session, 23.03.04). This categorical, ‘yes/no’ approach to accreditation ensured that facilities would not be overloaded without having the necessary infrastructure, but it often introduced severe delays. Numerous bottlenecks were experienced in efforts to ensure the ability of selected sites to meet all the accreditation requirements, and some of the sites were only provisionally accredited, or not accredited at all, thus functioning for varying periods without all the requirements being fully in place. Delays in becoming operational due to refurbishing and infrastructural installations (over which the Department of Health sometimes had little control) were
Downloaded by [UZH Hauptbibliothek / Zentralbibliothek Zürich] at 13:19 23 December 2014
22
Jacobs, Schneider and Van Rensburg
experienced to varying degrees by all the ART sites, and almost a year passed between the original selection of the sites to the time of their final accreditation as qualifying ART sites, meanwhile excluding many patients who could have qualified for treatment (Table 1) (FSDoH, 2004c). Even then only four of the five selected ART service points were eventually ‘officially’ fully accredited as operational. In particular, the ART service points in rural and sparsely populated Xhariep District struggled to meet the prescribed accreditation requirements (TT meeting, 26.01.05). Although service points in this district were operational from September 2004, only one facility (Bophelong CHC) was fully accredited; the final accreditation of the other two facilities (Ethembeni and Itumeleng) happened only in December 2006, even though they were already functioning as combined treatmentassessment sites. This sort of situation not only reflects, but, as Barron (2003) warned, perpetuates longstanding inequities between urban and rural health facilities. Individual patient criteria In the initial stages of the programme in the Free State, after confirming the HIV status of a patient, the next criterion was the patient’s citizenship, the only social criterion of programme exclusion. This was implemented practically by requesting patients to present their identity documents (ID) when accessing ART. However, this criterion was challenged when it became clear that its implementation by providers was experienced as arbitrarily unjust. By mid-2005, the Task Team (TT) revisited the ID issue, where a participant noted the following: ‘…Nurses have been trained to insist on an ID. Hundreds of people have been turned away because they don’t have IDs. The husband is on treatment, but the wife is dying because she doesn’t have an ID. Now, the fact that a relative of the Premier has been turned away for not having an ID, the whole issue has been highlighted. This person has a CD4 of 7 and will die before an ID can be organised. The nurses want someone to write a clear policy on the ID issue. They feel bad because they have sentenced hundreds of people to death’ (TT meeting, 22.06.05). At the Mangaung University Community Partnership Project (MUCPP) community health centre there was even a notice displayed, stating: ‘No ID, no treatment.’ Thus, it was emphasised, ‘We must modify the system so that people can get treatment, ID or not’ (FSDoH official, TT meeting, 22.06.05). This exclusion criterion was subsequently dropped. Shortly afterwards, it was directed that the message should be spread that it was ‘not necessary to use ID for access, rather for data purposes. ID should not be really a condition for access to treatment.’ This directive also made it possible for citizens of Lesotho
(a neighbouring country bordering the Free State) to access the public-sector ART programme in the Free State. Within sites, the criteria for treatment access follow the clinical guidelines. Table 2 shows the patient selection criteria for adult patients, as stated in the national and provincial ARV clinical guidelines. Both sets of guidelines lean towards a hierarchical process of determining who gains access to treatment. These ARV treatment guidelines apply two sets of core technical criteria, namely clinical or medical eligibility criteria and adherence to treatment criteria. Clinical or medical eligibility criteria Clinical or medical eligibility criteria (as defined by WHO) are applied to all patients, and are similar in all respects. Beginning with the assessment of patients’ pre-treatment to confirm medical eligibility, the responsibilities of doctors were clearly stated in the FSDoH treatment guidelines (FSDoH, 2004a). As such, the certifying doctor or clinician is the key role player in certifying a patient’s medical eligibility. Despite their apparent objectivity, clinical criteria for ART can involve value judgements and should be examined and understood as such (Bennett & Chanfreau, 2005). Interviews conducted with doctors involved in the ART programme as part of the facility appraisals revealed that, while the majority of doctors were satisfied with the patient selection criteria, concerns were raised regarding how to deal with the waiting periods for eligible patients, drug shortages, managing borderline cases, the exclusion of alcohol-dependent patients, and confirming the eligibility of tuberculosis (TB) patients for ART. When asked to rate degrees of feeling solely responsible for determining the eligibility of patients for ART their responses varied from ‘not at all,’ ‘a little,’ ‘somewhat’ to ‘very much’ responsible, suggesting some doctors understood their role as involving discretionary decisions on resource allocation. This perception of responsibility appeared to be determined by the extent and accessibility of clinical support for doctors and the period of time the facility had been providing ART. In particular the lack of stated criteria for allocating treatment among those who already meet the clinical or medical criteria puts clinicians in the uncomfortable position of deciding who receives ART first (Krízová & Simek, 2002; Loewenson, 2003; Rosen et al., 2004a). This became the subject of considerable debate. One recommendation of the provincial ARV Task Team was to select patients most likely to survive over others with severely depressed immune systems, severe opportunistic infections, or TB. The preference for treatment-naïve patients also emerged as a criterion. Such decisions resulted in directives to front-
Table 1: Accreditation and opening dates for ART sites in the Free State, South Africa
Accreditation dates Opening dates Time difference
Lejweleputswa
Motheo
Thabo Mofutsanyane
Xhariep
Fezile Dabi
30 March 2004 3 May 2004 20 days
22 June 2004 28 June 2004 3 days
4 August 2004 10 August 2004 2 days
18 March 2005 13 September 2004 *
3 December 2004 6 December 2004 Weekend
* The Bophelong community health centre was accredited post-implementation Note: The time difference excludes weekends and public holidays Source: FSDoH (2004c)
African Journal of AIDS Research 2008, 7(1): 19–27
23
Table 2: Comparison of clinical guidelines for adult patients receiving ARVs according to the South African National Department of Health (NDoH) and the Free State Department of Health (FSDoH) NDoH Treatment Guidelines
FSDoH Treatment Guidelines
i. ii.
i. ii.
Downloaded by [UZH Hauptbibliothek / Zentralbibliothek Zürich] at 13:19 23 December 2014
iii.
iv. v.
vi.
HIV-positive individual. Meet medical criteria: • CD4
African Journal of AIDS Research Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/raar20
Rationing access to public-sector antiretroviral treatment during scale-up in South Africa: implications for equity Nandipha Jacobs , Helen Schneider & H.C.J. Van Rensburg Published online: 11 Nov 2009.
To cite this article: Nandipha Jacobs , Helen Schneider & H.C.J. Van Rensburg (2008) Rationing access to public-sector antiretroviral treatment during scale-up in South Africa: implications for equity, African Journal of AIDS Research, 7:1, 19-27, DOI: 10.2989/AJAR.2008.7.1.3.431 To link to this article: http://dx.doi.org/10.2989/AJAR.2008.7.1.3.431
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Copyright © NISC Pty Ltd
African Journal of AIDS Research 2008, 7(1): 19–27 Printed in South Africa — All rights reserved
AJAR
EISSN 1727–9445 doi: 10.2989/AJAR.2008.7.1.3.431
Rationing access to public-sector antiretroviral treatment during scale-up in South Africa: implications for equity Nandipha Jacobs1*, Helen Schneider1,2 and HCJ Van Rensburg2 Centre for Health Systems Research & Development, University of the Free State, PO Box 339, Bloemfontein 9300, South Africa 2 Centre for Health Policy, University of the Witwatersrand, PO Box 1038, Johannesburg 2000, South Africa * Corresponding author, e-mail: [email protected]
Downloaded by [UZH Hauptbibliothek / Zentralbibliothek Zürich] at 13:19 23 December 2014
1
As with other large-scale antiretroviral therapy (ART) programmes initiated in developing countries, rationing of ART in South Africa was inevitable in the face of limited human and infrastructural resources. This paper discusses rationing of ART and its implications for equity in South Africa, with a particular emphasis on the national programme in the Free State Province. We examine both the explicit and implicit dimensions of ART rationing, while considering the content of policy, programme design and implementation processes. The explicit rationing criteria included in the national comprehensive plan, as well as the associated treatment guidelines, direct and influence rationing decisions practised locally. However, implicit processes of rationing, which limit demand, are also wittingly and unwittingly introduced in several ways, and at all levels of the health service. Consequences of rationing manifest in the high numbers of patients lost to the system and the difficulties faced by the most impoverished clients in gaining access to ART services on an ongoing basis. Not losing sight of the equity element within rationing is important, and this aspect will have to be formally monitored for eventual success of the country’s ART programme. ART roll-out processes also require ongoing efforts to build consensus and confidence in rationing, accomplished with a view to optimal fairness. Keywords: accessibility, Free State Province, HAART, patient selection criteria, programme implementation
Introduction Many African countries have begun to develop and implement large-scale highly active antiretroviral therapy (HAART) programmes amid limited human and infrastructural resources. As a result of these limited resources, access to HAART for AIDS illness is being rationed in many developing countries (Loewenson, 2003; McCoy, 2003; Stewart, Paradath & Milford, 2006). Rationing is “a process of distributing scarce services within a population when it is not possible to provide each patient with every health care service appropriate to his/her medical need” (Krízová & Simek, 2002, p. 308). This is often the case when the demand for a service or a commodity outstrips its supply. In such cases rationing strategies are required in order to manage the healthcare delivery process efficiently. The literature divides rationing strategies into explicit (stated) and implicit (unstated) arrangements (Krízová & Simek, 2002; Rosen, Sanne, Collier & Simon, 2004a; Bennett & Chanfreau, 2005): “Explicit rationing occurs when defined and widely understood criteria such as age-related or insurance-related criteria are used to determine access,” and it is more likely to be developed as part of national policies (Bennett & Chanfreau, 2005, p. 542). Examples of explicit rationing for HAART include citizenship requirements, defining certain age-groups or geographical areas for priority access, psychosocial criteria (such as proof of adherence) or requiring co-payments
for care (Rosen, Sanne, Collier & Simon, 2005). Implicit rationing, on the other hand, “lacks any overarching plan or clearly defined criteria, but rather depends on subtle decisions, many of which are made by health care providers” (Bennett & Chanfreau, 2005, p. 542). Examples of implicit rationing include limitations on access to underlying healthcare infrastructure to deliver HAART, scheduling patients on a first-come-first-serve basis, appointment systems, and using waiting lists or queuing systems, the latter being particularly susceptible to local or provider influences. Rationing has implications for both the effectiveness and equity of HAART programmes. For example, prioritising mothers of young children (as opposed to older women or men) for treatment would increase the societal benefits (effectiveness) of a HAART programme by reducing the number of orphans and vulnerable children (Loeweson, 2003; McCoy, 2003). Similarly, treating only skilled workers may have macro-economic benefits, but would be regarded as highly inequitable in that it excludes poor and unemployed populations (Rosen et al., 2005). In general, without clear criteria or an explicit strategy for the rationing of treatment, access to treatment will tend towards reflecting underlying socio-economic inequities and social biases (Loewenson, 2003). If they are to be fair or equitable, therefore, rationing strategies used in treatment
Downloaded by [UZH Hauptbibliothek / Zentralbibliothek Zürich] at 13:19 23 December 2014
20
programmes require clearly defined and transparent procedures within an overarching process of reducing unfair social and geographic healthcare disparities (Loewenson, 2003; McCoy, 2003; Loewenson & McCoy, 2004; Rosen, Sanne, Collier & Simon, 2004b). In November 2003, the Comprehensive Plan for Comprehensive HIV and AIDS Care, Management and Treatment for South Africa (National Department of Health [NDoH], 2003a) (henceforth, the ‘Comprehensive Plan’) was adopted with the purpose of providing a national framework for expanded treatment for people with HIV or AIDS. As with the WHO/UNAIDS ‘3 by 5’ initiative to extend antiretroviral treatment to 3 million people by 2005 (World Health Organization [WHO], 2003), universal access for clinically eligible people is an explicit policy objective of the Comprehensive Plan (Loewenson, 2003; NDoH, 2003a). “The South African constitution and government require that implementation of a programme of such importance be carried out in a universal and equitable manner” (NDoH, 2003a, p. 18). Universal access (thus, also equitable implementation of the plan) to care and treatment — irrespective of race, gender and economic status — is one of the core explicit guiding principles of the Comprehensive Plan, and accordingly, on the surface, access to ART in South Africa requires no rationing. Moreover, Cleary, Chitha, Jikwana, Okorafar & Boulle (2005) show that government conditional grants, at least for the first few years of programme implementation, would be sufficient to deliver ART nationally. In practice, however, it is widely recognised that human and infrastructural resource constraints will prevent a match between supply of and demand for (and even less the need for) ART for some years to come. In 2005, the projected population of the Free State Province was approximately 2.8 million, including 381 860 HIV-infected people, of whom 40 706 were eligible for ART (WHO stage-IV disease) and 8 177 were urgently in need of ART (Actuarial Society of South Africa [ASSA], 2005). By October 2005, just over a year after commencement of the ART programme in the Free State, only 3 322 (8%) of clinically eligible people were receiving ART (Free State Department of Health [FSDoH], 2005). In this context, rationing appeared to have played an obvious part. Against the backdrop of the gap between supply and demand of ART, Stewart et al. (2006) raised concerns that the provisioning of ART will exaggerate existing inequities in resource allocation and service provision, thereby undermining the very system that the Comprehensive Plan was supposed to strengthen through the introduction of ART. They also pointed to emerging threats to achieving the equity goals in the implementation of the ART programme, and in particular equity among provinces, gender groups, urban and rural areas, and paediatric versus adult ART access. This paper analyses rationing of ART in South Africa, and the implications of such rationing for equity. The focus is on the Free State Province, one of the nine provinces of South Africa. We examine both the explicit and implicit dimensions of rationing ART, while considering the content of policy,
Jacobs, Schneider and Van Rensburg
programme design, and implementation processes during the first years of ART roll-out in the public sector. Methods and data sources The paper draws on secondary information from a comprehensive search of electronic databases, in an effort to identify policy and programme texts that specifically examine issues of rationing, patient selection and ART. These sources include national and provincial policies, treatment guidelines and scientific papers. Empirically, the paper is based on a series of data-gathering initiatives conducted by The Centre for Health Systems Research and Development for a research project entitled, ‘Documenting, monitoring, evaluating and facilitating implementation of the national Treatment Plan in the Free State, South Africa.’ We have collated primary data from individual interviews with high-ranking and mid-level officials of the Free State Department of Health (FSDoH) and medical specialists serving on the provincial ART Task Team (TT), as well as from notes of discussions taken at provincial meetings related to ART implementation, expansion and management. From these, direct citations are provided to illustrate, emphasise and substantiate findings on forms of rationing.† In addition, longitudinal data generated by ongoing appraisals of facilities providing ART in the Free State Province are used, including individual and group interviews with health workers. Guided by the availability and relevance of information, various sources of information are triangulated. Moreover, extensive discussion and agreement regarding the findings, our emphasis and interpretation has transpired between two of the authors (NJ and HCJ VR) who actually participated widely in data-gathering and in the research team. The paper describes the forms of rationing emerging and evident in the Free State public-sector ART programme, and examines both policy and practice contexts. Finally, the consequences of ART rationing strategies for service delivery, access and equity are discussed. ART rationing and its implementation in the Free State Province Phased geographical access An obvious way of limiting access to treatment is to offer ART only to patients residing in specified geographic catchment areas (McCoy, 2003; Rosen et al., 2004b; McGough, Reynold, Quinn & Zenilman, 2005). Ntuli, Ijumba, McCoy, Paradath & Berthiaume (2003) argued that most sub-Saharan countries would scale up treatment programmes incrementally through first offering ART at a few facilities and then adding more over time. This process was evidently also considered necessary to achieving the goal of equitable ART access in South Africa (NDoH, 2003a; Doherty, Loveday, Stewart & Thomas, 2005). As stated in the Comprehensive Plan (NDoH, 2003a), the ART programme was to be phased in over a five-year period, with the aim of achieving universal coverage of new
† Statements or remarks made during meetings of the provincial task team (TT), or in other official meetings related to the ART programme, are anonymously referred to as ‘FSDoH official’ or ‘FSDoH clinician’, TT meeting, with the date (day/month/year) of the statement or remark.
Downloaded by [UZH Hauptbibliothek / Zentralbibliothek Zürich] at 13:19 23 December 2014
African Journal of AIDS Research 2008, 7(1): 19–27
AIDS cases by the end of the 2008/09 financial year. The NDoH anticipated that by 2004 every health district in the country would have at least one service point,1 and within five years these service points would be expanded to every local municipality. In the Free State, one service point in each of the five districts was envisaged within the first two years, and thereafter a steady expansion to all appropriate facilities (20 local municipal areas in total) (FSDoH, 2003). The phased approach was endorsed by the Joint Health and Treasury Task Team, which viewed it as “acceptable as long as the state worked towards the realisation of a programme to which everyone in need will ultimately have access” (NDoH, 2003b, p. 20). In the early stages, one directive in the Free State, with strong tones of exclusion and rationing, was to ‘make sure that only patients in the catchment areas of an ARV site come to the site’ (TT meeting, 13.10.04). The obvious rationing strategy here is that the majority of those patients not living within those few initial catchment areas would be excluded unless they were willing to relocate (Doherty et al., 2005). Nationally, the ART programme was seeking to achieve some kind of geographical equity regardless of differences in capacity between provinces and districts. However, the approach was a fairly crude one in that it did not include specific considerations of need, such as the geographic concentration of the HIV epidemic, population size and density, thus potentially negatively affecting access and equity. In the Free State, ART sites were allocated equally to the five districts, irrespective of the important geographic and demographic differentials characterising the epidemic in the province. Population sizes (and density) range from a high of 736 292 (52.6/km2) in Motheo District to a low of 132 070 (3.86/km2) in Xhariep District, with sub-districts showing even greater extremes, ranging from population sizes of 654 922 to 38 604, and population densities of 104.24/km2 to 3.55/km2 (Barron & Asia, 2001, pp. 30–31). In effect, this means that although there was an equal spread of facilities providing ART (all equally staffed), and thus supply of ART per district — a strikingly inequitable spread of service-provisioning in respect of patient numbers, need, and demand was created. Selection and accreditation of ART sites Ensuring the presence of essential staffing and infrastructure for delivering a quality ART service is central to the success of the programme. This has been implemented through a rigorous process of selection and accreditation of sites. The NDoH was centrally involved in the accreditation of ART sites nationwide, setting particular norms and standards that sites had to meet before providing ART to the public (NDoH, 2003a). The identification and selection of potential ART sites in the Free State commenced in early 2003, driven by the provincial ARV Task Team and with the involvement of district managerial structures (FSDoH, 2003). Qualifying facilities had to have the appropriate level of staffing, with adequately functioning and efficient support services, specifically drug procurement and distribution services, and sufficient and suitable physical space to accommodate the programme (FSDoH, 2003). As one FSDoH official observed, ‘The most important criterion was
21
to find appropriate space’ (FSDoH official, TT meeting, 12.01.04). By December 2003, potential ART sites in all five districts had been provisionally assigned and their accreditation was pending (TT meeting, 01.12.03). Eventually, important changes were effected regarding the final site selection in some of the districts. Van Rensburg (2005 and 2006) pointed out that the site selection for the first phase of the ART programme implementation in the Free State was very much a top-down, management-directed approach with little consultation with facility governing structures or communities. Decisions were rushed both managerially and politically, and neither equitable distribution nor optimal geographic access appeared to have played an overt role in decisions on ART service sites. It is not clear whether other more strategic considerations, such as the capacity of facilities to scale up rapidly or to provide a good focus of learning, were part of decision-making, although the perceived accreditation requirements naturally promoted those areas and facilities within the districts with better buildings and infrastructure. The accreditation process was handled by a joint team of officials from the NDoH and FSDoH. Four service points (i.e. one each in the Lejweleputswa, Motheo, Thabo Mofutsanyane and Fezile Dabi districts) consisted of a referral treatment site serviced by three referring assessment sites — the ‘1 x 3’ model (Van Rensburg, 2006). Clinics and community health centres (CHCs) serve as the primary sites for entry to the service for the diagnosis, staging and routine follow-up of ART patients (NDoH, 2003a, pp. 60–61). Patients who meet the criteria for ART are referred to the district or regional hospital that serve as the treatment site, where detailed assessments are required and a medical practitioner, in consultation with other staff, decide whether the patient will commence with ART (FSDoH, 2003).2 This model allowed for some decentralisation of decision-making, with the NDoH team involved in conducting the accreditation of treatment sites, while the FSDoH team was responsible for accrediting the assessment sites (TT meeting, 08.12.03). Potential ART sites that did not meet the accreditation standards at the time of the first visit had to be strengthened, and financial and technical resources were mobilised for such purposes (NDoH, 2003a). Accreditation feedback sessions took place for purposes of devising detailed strategies for strengthening sites that did not meet the accreditation standards. Only accredited sites would provide ART treatment, as emphasised by a member of the NDoH team: ‘If you are not accredited, you cannot start treatment — if it is green, it is green, not a tint of green’ (NDoH official, accreditation feedback session, 23.03.04). This categorical, ‘yes/no’ approach to accreditation ensured that facilities would not be overloaded without having the necessary infrastructure, but it often introduced severe delays. Numerous bottlenecks were experienced in efforts to ensure the ability of selected sites to meet all the accreditation requirements, and some of the sites were only provisionally accredited, or not accredited at all, thus functioning for varying periods without all the requirements being fully in place. Delays in becoming operational due to refurbishing and infrastructural installations (over which the Department of Health sometimes had little control) were
Downloaded by [UZH Hauptbibliothek / Zentralbibliothek Zürich] at 13:19 23 December 2014
22
Jacobs, Schneider and Van Rensburg
experienced to varying degrees by all the ART sites, and almost a year passed between the original selection of the sites to the time of their final accreditation as qualifying ART sites, meanwhile excluding many patients who could have qualified for treatment (Table 1) (FSDoH, 2004c). Even then only four of the five selected ART service points were eventually ‘officially’ fully accredited as operational. In particular, the ART service points in rural and sparsely populated Xhariep District struggled to meet the prescribed accreditation requirements (TT meeting, 26.01.05). Although service points in this district were operational from September 2004, only one facility (Bophelong CHC) was fully accredited; the final accreditation of the other two facilities (Ethembeni and Itumeleng) happened only in December 2006, even though they were already functioning as combined treatmentassessment sites. This sort of situation not only reflects, but, as Barron (2003) warned, perpetuates longstanding inequities between urban and rural health facilities. Individual patient criteria In the initial stages of the programme in the Free State, after confirming the HIV status of a patient, the next criterion was the patient’s citizenship, the only social criterion of programme exclusion. This was implemented practically by requesting patients to present their identity documents (ID) when accessing ART. However, this criterion was challenged when it became clear that its implementation by providers was experienced as arbitrarily unjust. By mid-2005, the Task Team (TT) revisited the ID issue, where a participant noted the following: ‘…Nurses have been trained to insist on an ID. Hundreds of people have been turned away because they don’t have IDs. The husband is on treatment, but the wife is dying because she doesn’t have an ID. Now, the fact that a relative of the Premier has been turned away for not having an ID, the whole issue has been highlighted. This person has a CD4 of 7 and will die before an ID can be organised. The nurses want someone to write a clear policy on the ID issue. They feel bad because they have sentenced hundreds of people to death’ (TT meeting, 22.06.05). At the Mangaung University Community Partnership Project (MUCPP) community health centre there was even a notice displayed, stating: ‘No ID, no treatment.’ Thus, it was emphasised, ‘We must modify the system so that people can get treatment, ID or not’ (FSDoH official, TT meeting, 22.06.05). This exclusion criterion was subsequently dropped. Shortly afterwards, it was directed that the message should be spread that it was ‘not necessary to use ID for access, rather for data purposes. ID should not be really a condition for access to treatment.’ This directive also made it possible for citizens of Lesotho
(a neighbouring country bordering the Free State) to access the public-sector ART programme in the Free State. Within sites, the criteria for treatment access follow the clinical guidelines. Table 2 shows the patient selection criteria for adult patients, as stated in the national and provincial ARV clinical guidelines. Both sets of guidelines lean towards a hierarchical process of determining who gains access to treatment. These ARV treatment guidelines apply two sets of core technical criteria, namely clinical or medical eligibility criteria and adherence to treatment criteria. Clinical or medical eligibility criteria Clinical or medical eligibility criteria (as defined by WHO) are applied to all patients, and are similar in all respects. Beginning with the assessment of patients’ pre-treatment to confirm medical eligibility, the responsibilities of doctors were clearly stated in the FSDoH treatment guidelines (FSDoH, 2004a). As such, the certifying doctor or clinician is the key role player in certifying a patient’s medical eligibility. Despite their apparent objectivity, clinical criteria for ART can involve value judgements and should be examined and understood as such (Bennett & Chanfreau, 2005). Interviews conducted with doctors involved in the ART programme as part of the facility appraisals revealed that, while the majority of doctors were satisfied with the patient selection criteria, concerns were raised regarding how to deal with the waiting periods for eligible patients, drug shortages, managing borderline cases, the exclusion of alcohol-dependent patients, and confirming the eligibility of tuberculosis (TB) patients for ART. When asked to rate degrees of feeling solely responsible for determining the eligibility of patients for ART their responses varied from ‘not at all,’ ‘a little,’ ‘somewhat’ to ‘very much’ responsible, suggesting some doctors understood their role as involving discretionary decisions on resource allocation. This perception of responsibility appeared to be determined by the extent and accessibility of clinical support for doctors and the period of time the facility had been providing ART. In particular the lack of stated criteria for allocating treatment among those who already meet the clinical or medical criteria puts clinicians in the uncomfortable position of deciding who receives ART first (Krízová & Simek, 2002; Loewenson, 2003; Rosen et al., 2004a). This became the subject of considerable debate. One recommendation of the provincial ARV Task Team was to select patients most likely to survive over others with severely depressed immune systems, severe opportunistic infections, or TB. The preference for treatment-naïve patients also emerged as a criterion. Such decisions resulted in directives to front-
Table 1: Accreditation and opening dates for ART sites in the Free State, South Africa
Accreditation dates Opening dates Time difference
Lejweleputswa
Motheo
Thabo Mofutsanyane
Xhariep
Fezile Dabi
30 March 2004 3 May 2004 20 days
22 June 2004 28 June 2004 3 days
4 August 2004 10 August 2004 2 days
18 March 2005 13 September 2004 *
3 December 2004 6 December 2004 Weekend
* The Bophelong community health centre was accredited post-implementation Note: The time difference excludes weekends and public holidays Source: FSDoH (2004c)
African Journal of AIDS Research 2008, 7(1): 19–27
23
Table 2: Comparison of clinical guidelines for adult patients receiving ARVs according to the South African National Department of Health (NDoH) and the Free State Department of Health (FSDoH) NDoH Treatment Guidelines
FSDoH Treatment Guidelines
i. ii.
i. ii.
Downloaded by [UZH Hauptbibliothek / Zentralbibliothek Zürich] at 13:19 23 December 2014
iii.
iv. v.
vi.
HIV-positive individual. Meet medical criteria: • CD4
