Sub-Saharan Africa (SSA) has 25% of the world's disease burden but only 1.3% of the world's health workers.1 This shortage of human resources for health is a critical limitation to the provision of antiretroviral treatment (ART) to those in need of it in SSA, the region having the highest burden of HIV globally.2,3 The lack of qualified human resources is a major challenge to meeting the United Nations three 90% targets for testing, treatment coverage, and viral suppression by 2020.4
Pharmaceutical services also experience staff shortages and workload pressure, particularly as the ART program has expanded so rapidly during the last decade.5,6 The African region has the lowest density of pharmaceutical staff worldwide (0.8 per 10,000 population, almost 5-fold lower than the region with the next-lowest density).7 Pharmaceutical care is an important component of the ART program, and addresses potential pharmaceutical-related problems and promotes patient adherence.8 Excellent adherence is critical to individual patient's well-being and to prevent the development of viral resistance that may compromise the effectiveness of the ART program.9 The shortage of pharmacists also deprives the population of vital expertise in the management of drug-related problems.10 Factors contributing to the shortage of pharmacists include a shortage of training institutions, migration of pharmacists to developed countries, an urban/rural maldistribution of pharmacists and most pharmacists working in the private sector serving a small proportion of the population.6,8
To address staff shortages and facilitate increased scale-up of the ART program, task shifting for midlevel workers to perform the tasks of more highly trained health workers has been recommended by the World Health Organization (WHO).11 Task shifting the clinical management of ART patients from doctors to nurses has been shown to produce comparable patient ART outcomes.12 Training lower cadres of pharmacy workers has also been recommended to promote pharmaceutical care for ART programs in SSA.8
South Africa has the world's largest ART program with more than 2.6 million people who have started ART.3 There remains, however, a great need to initiate people onto ART as only 42% of adults living with HIV receive ART.3 Two pharmaceutical care models have recently developed in the country involving, first, indirectly supervised pharmacist assistants (ISPAs), and second, clinical nurse practitioners who issue pharmaceuticals. A previous economic evaluation has found the ISPA model to be the least costly pharmaceutical model in the ART program, but did not include measures of quality of care or patient outcomes.6 Evaluation of the quality of care is an important part of the quality assurance mechanism for task-shifting approaches.11 There are no published data comparing the quality of pharmaceutical care or the clinical effectiveness of ART among patients who access sites that have adopted these 2 models. The aim of this study was to compare the ISPA and nurse-managed dispensing of ART models in terms of: (1), the quality of pharmaceutical care; (2), clinical outcomes of ART patients accessing these services; and (3), the cost of providing each of the approaches from a health services perspective.
Study Design and Setting
A retrospective analysis of pharmaceutical care quality audits, patient clinical data, and staff costing data collected from 7 ISPA facilities and 8 nurse-managed facilities was undertaken in South Africa. All the facilities were primary health care sites. The ISPA facilities were located in the Cape Town metropolitan district, Western Cape, and the nurse-managed facilities were located in KwaDukuza subdistrict of Ilembe, KwaZulu-Natal province. All facilities were supported by Kheth'Impilo, a nonprofit organization that supports the South African Department of Health with health systems strengthening innovations and pharmaceutical services. The selected facilities were all the ISPA facilities supported by Kheth'Impilo, and all the primary health care facilities supported by Kheth'Impilo in KwaDukuza sub-district. (There were no ISPA facilities located in KwaDukuza and no nurse-managed facilities in Cape Town.)
Description of Pharmaceutical Care Models
ISPAs are post-basic pharmacist assistants who have received an additional 6–12 months of on-site training and mentoring by a registered pharmacist to work under the indirect supervision of a pharmacist. Post-basic pharmacist assistants are a cadre of midlevel workers, specifically trained to manage pharmaceutical stock control, issue medication, provide medication information to patients, advise on adherence, and intervene where needed regarding the legality of prescriptions. They are registered with the South African Pharmacy Council after achieving competence in accredited course material and completing 2 years of in-service training in a pharmacy registered for training. The Pharmacy Act allows post-basic pharmacist assistants to work under the indirect supervision of a pharmacist in primary health care settings under certain specific conditions: Only patient-ready packs can be dispensed; all pharmacy-related standard operating procedures must be available, read and signed by the ISPA; and a pharmacist must visit the site at least once per month and document the visit. The ISPAs are expected to take responsibility for pharmaceutical services within their scope of practice at site level. Responsibilities include the dispensing of ART, management of the dispensary, and management of all orders and stock in the facility. The Designated Supervisory Pharmacists (DSPs) supervise up to 5 ISPAs (who work in clinics other than where the DSP is based), visit the clinics once per week to oversee the ISPA functions, and perform an audit of the site pharmaceutical service monthly. Additional training and mentoring of ISPAs is provided by the DSP to ensure continued quality of care while working under indirect supervision.
To expand the ART program, nurse initiated and managed ART (NIMART) was introduced in South Africa in 2010.13 At certain sites, clinical nurse practitioners issue prescriptions, which are dispensed by pharmacists or ISPAs at the pharmacy or dispensary. However, in the nurse-managed pharmaceutical model, nurses issue pharmaceuticals during patient consultations, which was the model adopted at the nurse-managed sites included in this study during 2012, and is a model that has been widely implemented. Small supplies of pharmaceuticals are kept in the consultation rooms with additional medication fetched from the medicine room on a daily basis. Nurses also perform stock ordering, stock control, and general management of the medicine rooms. At the 2 largest sites, post-basic pharmacist assistant's supported nurses in the medicine rooms being involved with stock ordering and stock control, but they were not trained in indirect supervision and did not dispense pharmaceuticals. A roving DSP oversaw pharmaceutical care within the district and performed a quality audit at the sites on a monthly basis.
Data Collection and Analysis
Pharmacy Quality Audits
Standardized audit tools were developed to routinely assess the quality of pharmaceutical care at sites where ART is offered. The audit form assessed 4 components of pharmaceutical care, namely, (1) Good Pharmacy Practice (13 point score); (2) stock control (12 point score); (3) evaluation of prescription and patient folders (scores depended on the number of folders evaluated and ranged from 30 to 600, with 6 points per folder); and (4) patient exit interview (checking of medication label and standardized questions posed to 10–20 patients after they had received medication) (see audit form, Supplemental Digital Content 1, http://links.lww.com/QAI/A774). For folder evaluations, pharmacists randomly selected folders of existing ART patients to review, as well as all folders of new patients initiated on ART by nurses in the preceding month; thus, the number of folders evaluated per month per site varied. The pharmacists selected consecutive patients on the day that the pharmacist visited the clinic for exit interviews. Structured exit interview questions included assessing the patient's knowledge of the indication for the medication, knowledge of the use and storage of the medication, and rating the patient's satisfaction with the dispensing process. The results of routine monthly audits conducted by DSPs between January 2013 and December 2013 were analyzed. Scores for each section were calculated as a proportion, and proportions were compared using risk ratios (RRs) and 95% binomial exact confidence intervals (CI). Overall weighted scores combining all 4 sections were calculated for each dispensing model such that the score from each section contributed an equal weight to the overall score.
All ART-naive adults ≥16 years of age who commenced triple ART at the 15 clinics between 1 January 2012 and 6 months before site database closure were included in analyses. Site database closure varied slightly between the sites (between December 2013 and October 2014). Before April 2013, HIV-positive adults and adolescents were eligible to start ART if they had a CD4 cell count ≤200 cells per microliter and/or a World Health Organization stage IV defining illness; were pregnant or had tuberculosis with a CD4 cell count ≤350 cells per microliter; or had multidrug-resistant tuberculosis irrespective of CD4 cell count.13 From April 2013, ART eligibility criteria were expanded to include all patients with CD4 cell counts ≤350 cells per microliter and/or a WHO stage III or IV defining illness, and all pregnant women irrespective of CD4 cell count.14 Standardized first-line regimens consisted of 2 nucleoside reverse transcriptase inhibitors and 1 nonnucleoside reverse transcriptase inhibitor.
Routine, individual-level patient data were collected prospectively for patient monitoring purposes by designated site-based data capturers at each patient visit using standardized custom-designed electronic databases, with data being regularly pooled to a central data warehouse using standard operating procedures. Regular data cleaning and quality control procedures were implemented.
Patients were followed up from the start of ART until the earliest of loss to care (through mortality or loss to follow-up), transfer out to another facility, or 2 years after starting ART. Patients transferring to other facilities were censored on the date of last clinic visit. The clinical outcomes analyzed were patient attrition (through death or loss to follow-up) after starting ART and proportions of patients achieving virological suppression on ART. Patients were recorded as having died based on health care worker or community health worker reports. A patient was defined as lost to follow-up if no visits to the clinic occurred for 180 days or more.15 Virological suppression was defined as a viral load <400 copies per milliliter.
Kaplan–Meier estimates were used to analyze time until patient attrition. For patients who did not return to the clinic after starting ART, 1 day of survival time was added to include them in survival analyses.
An incremental ingredients costing method was used to estimate the mean human resource costs per patient visit and per item dispensed for the period January 2013 to December 2013, and then compared between the 2 pharmaceutical models of care. Salary unit costs were derived from staff salary records in Cape Town and estimated from Department of Public Service and Administration Circulars of Occupational Specific Dispensation salaries for KwaZulu-Natal with levels supplied by the Department of Health. Human resources costs pertaining to pharmaceutical-related activities only were considered. Pharmaceutical-related activities were defined as any staff activity pertaining to ordering and management of pharmaceutical stock, maintenance of medicine rooms, and time spent issuing medication and counseling patients regarding correct use of medication. Costs related to clinical consultation and medication costs were not considered. In both models, the cost of pharmacist supervision was also included (above-service level cost). All costs are presented in 2013 prices and were converted to US dollars at the average exchange rate in 2013 of US$ 1 = ZAR 10.34.16
At nurse-managed sites, professional nurses worked in the medicine room doing stock control and ordering, and also consulted patients and issued medicine. The full-time equivalent of nurses doing stock control and ordering at sites was estimated based on interviews with facility managers. The average proportion of time that consulting nurses spent on pharmaceutical-related activities was assumed to be 32%, derived from a previous time and motion study.6 A sensitivity analysis was also performed assuming lesser and greater proportions of 28% and 36%, respectively.6 At ISPA facilities, ISPAs spent most of their time managing and dispensing medication for ART patients, and cost calculations were limited to estimated expenditure related to ART patient visits. A proportion of their time (and that of the DSPs) was related to stock control and stock ordering for general primary health care pharmaceuticals. The full-time equivalent spent for ART patient–related activities was derived from interviews with DSPs and the district program manager, and was assumed to be 80% for ISPAs and 62.5% for DSPs.
Population and health services characteristics of the provinces and districts in which the 2 pharmaceutical models were located were compared using Pearson's χ2 test where numeric data were available. Data were analyzed using Microsoft Excel and Stata version 13 (College Station, TX). Ethical permission for the study was provided by the University of Cape Town Human Research Ethics Committee. Permission for the study was also granted by the City of Cape Town health and the KwaZulu-Natal department of health research committee.
Key characteristics of the population and health services of the districts in which the 2 pharmaceutical models were located are tabulated in Supplemental Digital Content 2 (see http://links.lww.com/QAI/A774). The region in which the nurse-managed clinics were located had a higher antenatal HIV prevalence rate, higher HIV testing coverage, a lower provincial rate of loss to follow-up among ART patients and equivalent viral suppression among ART patients, compared with the region in which the ISPA clinics were located.
Quality of Pharmaceutical Care
The results of the pharmaceutical care audits are shown in Table 1. Good Pharmacy Practice scores were 15% higher (relatively) at ISPA compared with the nurse-managed clinics [overall 84.0% at ISPA sites vs. 73.3% at nurse clinics; RR = 1.15 (95% CI: 1.09 to 1.20); P < 0.0001]. Stock control scores were 12% higher at ISPA sites [90.2% at ISPA sites vs. 80.4% at nurse clinics; RR = 1.12 (95% CI: 1.08 to 1.17); P < 0.0001]. Evaluation of prescription and patient folder scores were similar between models, RR = 1.01 (95% CI: 1.00 to 1.03); P = 0.071. Standardized patient exit interview scores were 17% higher at ISPA sites [89.1% at ISPA sites and 75.9% at nurse clinics; RR = 1.17 (95% CI: 1.16 to 1.19); P < 0.0001]. The overall weighted score was 11% higher at ISPA sites [88.8% at ISPA and 79.9% at nurse clinics; RR = 1.11 (95% CI: 1.09 to 1.13); P < 0.0001]. Figure 1 shows patient exit interview scores according to month during 2013.
A total of 10,751 patients were included in clinical analyses, of whom 5406 (50.3%) attended ISPA sites. Characteristics at the start of ART were similar between patients at ISPA and nurse sites: The median age was 32.9 years at ISPA sites and 33.4 years at nurse sites; men accounted for 37.5% of patients at ISPA sites and 39.6% at nurse sites; the median CD4 cell count was 220 cells per microliter (interquartile range: 121–300) at ISPA sites and 226 cells per microliter (interquartile range: 129–303) at nurse sites; and in both groups 47% of patients were in WHO clinical stages 3 or 4.
During 11,514 patient-years of follow-up, 217 deaths were recorded and an additional 1965 patients became lost to follow-up. Figure 2 shows the Kaplan–Meier cumulative probabilities of patient attrition according to the duration of ART. After 2 years, cumulative attrition at ISPA and nurse-managed sites was 20.7% (95% CI: 19.4% to 22.0%) and 31.5% (30.0%–33.2%), respectively. After 2 years, the Kaplan–Meier cumulative probabilities of reported mortality were 1.7% (95% CI: 1.3% to 2.1%) and 3.8% (95% CI: 3.2% to 4.6%) at ISPA and nurse-managed sites, respectively; and the cumulative probabilities of loss to follow-up were 19.3% (95% CI: 18.0% to 20.6%) and 28.8% (95% CI: 27.2% to 30.4%) at ISPA and nurse-managed sites, respectively.
A total of 8897 viral load results were available to analyze. After 1 year of ART, the proportion of patients in the cohort achieving virological suppression at ISPA and nurse-managed sites was 89.6% (95% CI: 87.9% to 91.1%) and 85.9% (95% CI: 83.7% to 88.0%), respectively.
Table 2 shows the human resources involved with pharmaceutical-related activities and mean provider staff costs per patient visit and per item dispensed. The mean cost per patient visit was US$ 1.35 at ISPA and US$ 1.89 at nurse-managed facilities, ie, 29% lower at ISPA facilities. The mean cost per item dispensed was US$ 0.43 at ISPA and US$ 0.84 at nurse-managed facilities, ie, 49% lower at ISPA facilities.
Table 3 shows the sensitivity analysis when using low and high assumptions regarding the proportion of time that dispensing nurses spent on dispensing-related activities. The proportionate savings at ISPA facilities per patient visit varied between 21% and 35%, and per item dispensed varied between 44% and 54%, ie, they were reasonably stable using the differing assumptions.
Pharmaceutical care is an important determinant of the success of the ART program,8,17 and will play an important role for SSA achieving the UNAIDS goal of 90% of diagnosed HIV-positive people receiving ART and 90% being virologically suppressed by 2020.4 In light of the shortage of pharmacists in SSA, task-shifting pharmaceutical care from pharmacists to lower cadres of workers has become essential to address the human resources required for the ART program. This is the first study to our knowledge that has compared the quality of pharmaceutical care and patient clinical outcomes between 2 task-shifting approaches to pharmaceutical care in SSA.
Pharmaceutical quality of care was found to be higher at ISPA sites than at nurse-managed sites. There have been anecdotal reports of possible compromised quality of pharmaceutical care at nurse-managed sites,6 and a lack of maintenance of pharmacy records.18 Quality may be reduced where the prescriber and the dispenser are the same person, due to a lack of quality control. The focussed training, support, and mentoring regarding pharmaceutical care which ISPAs receive likely enhances the quality of pharmaceutical management and dispensing at ISPA sites. Although in a previous study qualitative exit interviews found patients preferred the anonymity of receiving ART in the consultation room at nurse-managed sites instead of at the dispensary at ISPA sites,6 quantitative patient exit interview scores in this study (which included client satisfaction) were 17% higher at ISPA sites. It may be that patients are more comfortable receiving medication at dispensaries in more recent years as a result of reduced stigma.
Patient clinical outcomes were observed to be better at ISPA managed sites. A previous study found that patient clinical outcomes were equivalent between patients at ISPA sites and patients accessing sites having a full-time pharmacist.19 As the clinics using the 2 pharmaceutical care models in the present study were located in different provinces, local differences in health systems and population differences may have influenced clinical ART outcomes. HIV prevalence was higher in the province in which the nurse-managed clinics were located; however, provincial ART outcomes during the study period were not inferior in the province in which the nurse-managed sites were situated.20 Geographic differences may, nevertheless, have contributed to differences in the clinical ART outcomes observed in this study, and it was not possible to quantify the effect of the pharmaceutical care model attributable to observed differences in clinical outcomes independently from other health systems and population level factors.
Provider staff costs for pharmaceutical-related activities were lower for the ISPA model per patient visit and per item dispensed. The ISPA model was also considerably less expensive in an economic evaluation conducted in 2009/2010.6 Cost differences between the models may be explained due to the considerably lower salaries of ISPAs than professional nurses. Compared with the evaluation in 2009/2010, costs per patient visit were considerably lower for both models. This is likely related to the doubling and tripling of patient to staff ratios between the evaluations as the number of patients on ART has increased considerably.5 The 33% depreciation of the South African Rand to the USD between the 2 evaluations further contributed to this difference.
Nurses who manage ART patients in South Africa have high workloads and have much work pressure.18,21 A vacancy rate of 40% among nurses has been reported in South Africa,22 with estimates that up to 50% of nurses time may be required for ART program data collection and management.23 The importance of supporting nurses who manage ART patients with other lower staff cadres has previously been highlighted.18 The 2 pharmaceutical models are not mutually exclusive, and considering that the salary of an ISPA is lower than that of a professional nurse, nurse-managed sites should be able to cost-effectively use ISPAs to take over the role of pharmaceutical care to ease the workload burden of nursing staff, which may result in improvements in both pharmaceutical care and patient clinical care. A dispensing ISPA used at nurse-managed sites would free up nurse time, which may allow a greater number of patients to receive clinical consultations, thus expanding the capacity of the overburdened ART program. The ISPA model could thus compliment the NIMART program, which has been extensively rolled out in South Africa.
The strengths of the study include that 3 aspects of the pharmaceutical models were compared, that data were drawn from a relatively large number of sites, and that clinical data from a large patient cohort were collected prospectively. The limitations of the study include that sites adopting the 2 pharmaceutical models were situated in different provinces of South Africa, and that routine data were used to calculate patient clinical outcomes. For the cost analysis, assumptions had to be made regarding the proportion of time that ISPAs and DSPs spent on non–ART-related activities as accurate measurements were not available.
The ART program in SSA requires considerable expansion using cost-effective interventions to reach those in need of treatment and achieve the targets set by UNAIDS for 2020 and 2030 while simultaneously sustaining a high quality of care for patients currently receiving treatment. Expansion of pharmaceutical care capacity is a necessary component of scaling-up this program. This study has found that facilities that adopted the ISPA model had a better quality of pharmaceutical care, were less costly to manage, and were possibly associated with improved clinical outcomes. Further research comparing facility characteristics such as patient waiting times and investigating the ease of scale-up and sustainability of the models is warranted. Further qualitative research using interviews with nursing staff may be helpful to assess the burden on nurses regarding consulting and dispensing medication simultaneously. In the interim, further expansion of the ISPA pharmaceutical model and integrating it with NIMART care would likely enhance the cost-efficient scale-up of the ART program in SSA and other regions where expansion of treatment access is urgently needed.
The authors acknowledge participants included in the study, the departments of Health of the Western Cape and KwaZulu-Natal, the City of Cape Town Health, the Elton John AIDS Foundation, the United States Agency for International Development and the President's Emergency Plan for AIDS Relief.
1. Commission for Africa. Our common interest: report of the commission for Africa. 2005. Available at: http://www.commissionforafrica.info/2005-report
. Accessed March 4, 2015.
2. Bärnighausen T, Bloom DE, Humair S. Human resources for treating HIV/AIDS: needs, capacities, and gaps. AIDS Patient Care STDS. 2007;21:799–812.
4. UNAIDS. Fast-track—ending the AIDS epidemic by 2030. 2014. Available at: http://www.unaids.org/en/resources/documents/2014/JC2686_WAD2014report
. Accessed April 14, 2015.
5. Bekker L, Venter F, Cohen K, et al.. Provision of antiretroviral therapy in South Africa: the nuts and bolts. Antivir Ther. 2014;19(suppl 3):105–116.
6. Foster N, McIntyre D. Economic evaluation of task-shifting approaches to the dispensing of anti-retroviral therapy. Hum Resour Health. 2012;10:32.
7. World Health Organization. World health statistics. 2015. Available at: http://www.who.int/gho/publications/world_health_statistics/2015/en/
. Accessed October 14, 2015.
8. King RC, Fomundam HN. Remodeling pharmaceutical care in Sub-Saharan Africa (SSA) amidst human resources challenges and the HIV/AIDS pandemic. Int J Health Plann Manage. 2010;25:30–48.
9. Nachega J, Mills E, Schechter M. Antiretroviral therapy adherence and retention in care in middle-income and low-income countries: current status of knowledge and research priorities. Curr Opin HIV AIDS. 2010;5:70–77.
10. Matowe L, Duwiejua M, Norris P. Is there a solution to the pharmacist brain drain from poor to rich countries? Pharm J. 2004;272:98–99.
11. World Health Organization. Task Shifting: Global Recommendations and Guidelines. 2008. Available at: http://www.who.int/workforcealliance/knowledge/resources/taskshifting_globalrecommendations/en/
. Accessed February 23, 2015.
12. Sanne I, Orrell C, Fox MP, et al.. Nurse versus doctor management of HIV-infected patients receiving antiretroviral therapy (CIPRA-SA): a randomised non-inferiority trial. Lancet. 2010;376:33–40.
13. South African National Department of Health. Clinical guidelines for the management of HIV & AIDS in adults and adolescents. 2010. Available at: http://www.fidssa.co.za/Guidelines/2010_Adult_ART_Guidelines.pdf
. Accessed March 10, 2011.
14. South African National Department of Health. The South African Antiretroviral Treatment Guidelines. 2013. Available at: http://www.sahivsoc.org/practise-guidelines/national-dept-of-health-guidelines
. Accessed March 4, 2015.
15. Chi BH, Yiannoutsos CT, Westfall AO, et al.. Universal definition of loss to follow-up in HIV treatment programs: a statistical analysis of 111 facilities in Africa, Asia, and Latin America. PLoS Med. 2011;8:e1001111.
16. Average Yearly Exchange Rates. Available at: www.oanda.com
. Accessed November 15, 2014.
17. Abah I, Ojeh V, Falang K, et al.. Pharmaceutical care outcomes in an outpatient human immunodeficiency virus treatment center in Jos, Nigeria. J Basic Clin Pharm. 2014;5:57–61.
18. Davies NE, Homfray M, Venables EC. Nurse and manager perceptions of nurse initiated and managed antiretroviral therapy (NIMART) implementation in South Africa: a qualitative study. BMJ Open. 2013;3:e003840.
19. Monteith L, Fatti G, Grimwood A. Improved ART pharmacy service and patient care by utilising indirectly supervised post basic pharmacist assistants. Presented at: 5th South African AIDS conference 2011; June 7–10, 2011, Durban, South Africa.
20. National Department of Health of South Africa. ART Programme Analysis: Reviewing the ART Programme From April 2004 to March 2014. Pretoria, South Africa: National Department of Health of South Africa; 2014.
21. Georgeu D, Colvin C, Lewin S, et al.. Implementing nurse-initiated and managed antiretroviral treatment (NIMART) in South Africa: a qualitative process evaluation of the STRETCH trial. Implement Sci. 2012;7:66.
22. George G, Quintan T, Reardon C. Human Resources for Health in South Africa. Durban, South Africa: HEARD & University of Kwa-Zulu Natal; 2009.
23. Hirschhorn L, Oguda L, Fullem A, et al.. Estimating health workforce needs for antiretroviral therapy in resource-limited settings. Hum Resour Health. 2006;4:1.