On-line version ISSN 2078-5135
Print version ISSN 0256-9574
SAMJ, S. Afr. med. j. vol.98 n.4 Cape Town Apr. 2008
One of the first comprehensive reviews of the capacity and performance of any public sector ARV sites countrywide reveals that less than 20% of patients began ARV treatment within 3 months of referral - the period in which most deaths occur.
This has massive national implications for AIDS mortality and, in the opinion of some of the country's top HIV/AIDS experts, can be relatively easily addressed by a fundamental shift in approach. The latest evidence boosts their calls for the horizontal integration of officially accredited ARV sites into the overall primary health care system to speed up drug access for desperate people living with HIV/AIDS.
Clinicians on the ground have repeatedly asserted that HIV testing, CD4 staging, opportunistic illness prevention and drug readiness training can be adequately done at non-ARV-accredited primary health care sites, thus saving countless lives. Patients having had this 'work up' can then be referred to an official ARV site for their drugs.
Space and staff shortages were resulting in a waiting list for treatment initiation in one site, patients being turned away from the pharmacy in another, services being cut back and general staff burnout and dissatisfaction.
With doctor shortages at an all-time high (an estimated 70% of posts empty at most rural hospitals), the health department's insistence on rolling out ARVs only at the 362 accredited sites plus its reluctance to shift less complicated ARV drug prescribing to adequately trained nurses is causing deadly bottlenecks in delivery.
Described by Dr Francois Venter, chief of the Southern African HIV Clinicians Society as 'an excellent piece of operational research', the new study is of four Comprehensive Care Management and Treatment (CCMT) sites in Gauteng.
Until now most evidence on the feasibility and outcomes of ARV provision in South Africa has come from 'innovator' NGO programmes, led by motivated champions with external donor funding. The public sector has largely escaped expert examination.
Published by the Centre for Health Policy at the University of the Witwatersrand in January this year, the research, conducted on 'second-generation' 2-year-old CCMT roll-out sites in 2006, reported patient saturation already present at two of the sites.
There was 'a rapid growth in numbers', and expansion was 'continuing unabated'. Space and staff shortages were resulting in a waiting list for treatment initiation in one site, patients being turned away from the pharmacy in another, services being cut back and general staff burnout and dissatisfaction.
The authors said that because of these findings, it was 'distinctly possible' that some of the good performances observed in the study would not be maintained. There was a disturbingly high rate of loss to follow up, reaching 30% at one site. In the short term, attention to improving chronic disease care systems, such as simplified follow-up protocols, reorganisation of roles and task shifting, improving patient flow and, importantly, designing better monitoring and evaluation processes, would create efficiencies, allowing for larger numbers to be seen without loss of quality.
In the medium term however, the CCMT programme faced 'a demand that vastly outstrips supply'. The current model, based on dedicated CCMT sites in a few facilities, was unlikely to meet demand, much less need. 'There is now considerable evidence across the country of the feasibility of integrated, district-based approaches that mobilise the whole health care infrastructure and which achieve universal access while maintaining quality and outcomes,' they concluded.
Despite a plethora of reports and forms, monitoring and evaluation systems were weak across sites.
Scaling up such models would require a new cycle of planning that addressed, among other things, the need for new frameworks, methodologies and training, particularly comprehensive district-based approaches. Reviewing human resource policies and norms for the primary health care system, including core staff establishments to cater for expanded functions at this level, was vital.
Nurses the 'core' of delivery
When the data were collected in 2006, each site ranged from 600 to 1 700 patients and health care staff saw 50 -200 people a day. In terms of the norms set by the national CCMT plan, three of the sites had staffing shortfalls while all of them experienced a turnover of doctors and shortages of pharmacists, dieticians and social workers.Nurses formed the stable core of the sites and were key to setting the 'tone' of the service.
The study highlighted the clinical excellence of a public health approach, very high levels of treatment literacy (78 - 93% could state their latest CD4 result) and high disclosure rates, plus the constructive role of families in social support. Over 90% of patients agreed with the statements 'the health care workers I see cares about me', and 'the health worker provided you with feedback on whether the drugs were working or not'.
Within sites, VCT, TB, staging and ARV services, if not physically integrated, were managed closely together, making it relatively easy for patients to move from one service point to another. All sites had active systems of treatment preparation and adherence management while laboratory and drug supply systems functioned smoothly. Despite a plethora of reports and forms, monitoring and evaluation systems were 'weak' across sites.
Transport costs incurred in attending services were about R18 per visit at the community health centres and between R25 and R37 at the two hospitals. Patients spent between 4 and 6 hours at the clinic for each visit.
Venter said after studying the report that 'a lot of recent research, including from UCT, has shown that most deaths on the ARV treatment programme occur during the wait to access ARV treatment after referral'. Gauteng, together with KwaZulu-Natal, had the most number of patients on ARV in South Africa, making the study particularly relevant.
He picked out the poor use of viral loads to guide treatment, the lack of service integration and poor monitoring and evaluation findings of the study as eloquent testimony of country-wide problems. Venter said the poor use of viral loads to guide treatment was of concern as the test was 'expensive and operationally complex'. There was minimal evidence of health worker 'task shifting,' and an abundance of evidence of the ongoing human resource dilemma.
The document raised many operation issues, demonstrated the strength of a public health approach (excellent clinical results), the lack of mechanisms to chase down defaulters and the lack of urgency in ARV treatment initiation. The gender and age breakdown suggested that men and the youth were still not accessing these services.
'This is a hugely reflective and useful report. Now we need to discuss how to deal with these challenges,' he added.
The government's ambitious ARV rollout, aimed at universal access within 5 years, was launched in 2004. With less than half the people reached by the beginning of this year (ASSA estimates), the target date has now been revised to 2011.