Effectiveness and Acceptability of Delivery of Antiretroviral Treatment in Health Centres by Health Officers and Nurses in Ethiopia

Abstract

Objective

The World Health Organization (WHO) recommends shifting tasks from physicians to lower cadres for the delivery of antiretroviral treatment (ART) for countries short of physicians. Our objective was to evaluate the effectiveness and acceptability of ART delivery by health officers and nurses in Ethiopia.

Methods

A retrospective cohort study to evaluate outcomes of ART services in 25 health centresstaffed with health officers and/or nurses and 30 hospitals staffed with physicians in 2009. Median CD4-cell counts, mortality, loss to follow-up and retention were the primary outcomes. Interviews and focus group discussions were conducted with people living with HIV/AIDS, AIDS programme managers and health care providers to identify the types and acceptability of the tasks conducted by the health officers, nurses and community health workers.

Results

Health officers and nurses were providing ART, including ART prescription, for non-severe cases. The management of severe cases was exclusively the task of physicians. Community health workers were involved in adherence counselling and defaulter tracing. The baseline median CD4-cell counts per micro-liter of blood were 117 (interquartiles [IQ] 64,188) and 119 (IQ 67,190) at health centres and hospitals respectively. After 24 months on ART, the median CD4-cell counts per micro-literof blood increased to 321 (IQ 242, 414) and 301 (IQ 217, 411) at health centres and hospitals respectively. Retention in care was higher in health centres (76%, 95% confidence interval [CI] [73%-79%]) than hospitals (67%, 95% CI [66%-68%]). This difference is mainly due to the higher loss to follow-up rate in hospitals (25% versus 13%). Mortality was higher in health centres than hospitals (11% versus 8%), but the difference is not statistically significant. Service delivery by non-physicians was accepted by patients, health care providers and programme managers. However, the absence of a regulatory framework for task shifting, the lack of extra remuneration for the additional roles assumed by nurses and health officers, and the high cost for training and mentorship were identified as weaknesses.

Conclusion

ART delivery in health centres, based on health officers and nurses is feasible, effective and acceptable in Ethiopia. However, issues related to regulation, remuneration and cost need to be addressed for the sustainable implementation of these delivery models.

Introduction

Many countries have made significant progress towards the goal of universal access to antiretroviral treatment (ART).¹ However, major barriers must still be overcome if universal access is to be realized.² One of the major barriers, particularly in sub-Saharan Africa, has been a shortage of staff.^3–6 Conventional ART delivery by physicians is recognized as an impediment to the rollout of ART in resource constrained settings.^7,8 It is, therefore, recommended that alternative delivery models based on the public health approach with a redistribution of tasks among available staff are used to reduce dependence on physicians and expand the health workforce.^7–11

The WHO launched the ‘treat, train and retain initiative’ with task shifting as a strategy to increase the pool of staff who can provide ART services.^12,13 This initiative is based on reviews of the experience with alternative models of health services in low-income countries.^14,15 The use of mid-and low-level cadres as substitutes for physicians improves access to and sustainability of health services in rural and peri-urban communities.^14,15

Many countries have been using new delivery models for delivering ART since the guidelines and recommendations for task shifting was launched in 2008¹³ and are able to expand access to ART services.^1,9,10 The few studies which had evaluated the acceptability and effectiveness of these new delivery models were based on a small number of health facilities and patients.^16,17 Hence, there is a need to evaluate the acceptability and effectiveness of new ART delivery models at a national level.

Our aim was to evaluate health officers' (HOs) and nurses' delivery of ART in Ethiopia which started in 2006.¹⁸ We hypothesized that outcomes of ART services in health centres (based on HOs and nurses) would be the same as outcomes in hospitals (based on physicians), and that service delivery by HOs, nurses and community health workers (CHWs) would be acceptable.

Methods

A retrospective cohort study was conducted in 2009 based on 30 hospitals and 25 health centres providing ART for more than 200 patients. A list of facilities that were providing ART services was used as a sampling frame from which at least 15% of the health facilities in each region of the country were selected, including large (more than 1000 patients on ART), medium (500–1000 patients on ART) and small (200–500 patients on ART) facilities.

A database of ART patients was introduced in September 2006. Data were collected from routine patient registries, a hybrid of electronic and paper based patient management system, for mortality, loss to follow-up and retention. The study included patients who started ART in the 55 health facilities between September 2006 and August 2008. The study period ended in March 2009 so that the minimum follow-up period was six months and the maximum was 24 months. Mortality, loss to follow-up, retention and median CD4-cell counts were primary outcomes, and were determined at six, 12 and 24 months after starting ART.

A Ministry of Health study, conducted in 2007–08, estimated that 37% of hospital patients and 53% of health centre patients reported lost to follow-up, after 24 months on ART, had died.¹⁹ We used these findings to adjust and estimate the mortality of patients reported as lost to follow-up after 24 months of ART in this study.

We also analysed the trend in median (and interquartile range [IQR]) of CD4-cell counts for patients. Samples of 150 and 200 patient charts at each health centre and hospital, respectively, were selected randomly. CD4-cell counts at baseline, six, 12 and 24 months of treatment were identified for each patient.

A qualitative study using semi-structured interviews and focus group discussions was conducted. Semi-structured interviews were conducted in 2009 and early 2010 in all 55 health facilities to identify the role of HOs, nurses and CHWs for ART delivery. The questionnaire was completed by the head of the HIV clinic.

Exit interviews were also conducted with up to 30 patients on ART in each of the 55 health facilities to assess their satisfaction with the service they received. A semi-structured questionnaire in local language was developed and used as an interview-guide after pretesting in two health centres and one hospital in Addis Ababa, Ethiopia. Patients were randomly selected for interview until no new relevant information was being obtained. The interview was conducted by a physician from the FHAPCO. All study participants voluntarily gave informed consent.

Eight focus groups including a total of 57 patients on ART (seven groups with seven patients, one group with eight patients), were held to assess their attitude towards the services provided by HOs, nurses and CHWs. We also had one focus group with 11 programme managers and another with 10 health care providers on the new ART delivery model. Participants were selected on the basis that they were thought to have the potential to provide rich, relevant and diverse information pertinent to task shifting and ART delivery. A topic list was developed on the basis of the experience of the researchers in the field of ART delivery. The focus groups met for an average of two hours in the FHAPCO's meeting hall. The discussions were only partially audio-recorded (because of a problem with the tape recorder) after consent was received from the participants.

All the focus groups were led by an experienced moderator (YA), and attended by an observer who took notes on the discussions. Thematic analysis of the field notes and transcripts of the audio-recordings was based on the topic list used during the interviews and focus group discussions.

Results

ART delivery models and its scale-up

The number of health facilities providing ART and the number of patients receiving ART increased between 2006 and 2009 (Table 1). The ART delivery model in health centres was based on HOs and nurses who were responsible for the delivery of ART-related tasks. Severe cases and treatment failures were managed at hospitals by physicians: HOs and nurses in health centres referred patients with severe manifestations to physicians at hospitals level. CHWs performed tasks such as adherence counselling, referral and linkage between health facilities and care and support institutions, and defaulter tracing (Table 2).

Table 1

Antiretroviral treatment (ART) scale-up in Ethiopia (2006-09)

Variables	January 2006	January 2007	January 2009
Health facilities providing ART:
Health centre	0	125(51%)	311 (70%)
Hospital	73(100%)	119 (49%)	130 (30%)
Total	73(100%)	244 (100%)	441 (100%)
Patients started on ART:
At health centres	0	7,934 (11%)	38,845 (21 %)
At hospitals	26,921 (100%)	64,193 (89%)	147,693 (79%)
Total	26,921 (100%)	72,127 (100%)	184,978 (100%)

Table 2

Antiretroviral treatment (ART) related tasks performed by different staff in Ethiopia (2009-10)

	Physicians	Health officers (HOs)	Nurses	Community health workers (CHWs)^*
WHO staging	X	X	X	—
Opportunistic infections (Ols) management	X	X	X	—
Tuberculosis (TB) screening	X	X	X	—
TB diagnosis	X	X	X	—
ART eligibility	X	X	X	—
ART initiation	X	X	X	—
ART follow-up	X	X	X	—
Management of mild cases	X	X	X	—
Management of severe cases^**	X	—	—	—
Management of treatment failures and second-line ART	X	—	—	—
Adherence counselling	X	X	X	X
Defaulter tracking	—	—	—	X
Referral and linkage	X	X	X	X

CHWs have various names in Ethiopia: peer educators, expert patients, case managers, kebele-oriented outreach workers, adherence supporters. The most common are case managers and adherence supporters. Case managers play a role of care coordination between health facilities and community while adherence counsellors are focusing on adherence counselling and defaulter tracing activities

Severe cases included those with advanced HIV diseases with severe and persistent Ols, severe ART side-effects, ART failures

ART outcomes at health centres and hospitals

Table 3 indicates that retention in care was similar at health centres and hospitals after six months on ART (82% v 75%), and after 12 months (79% v 74%), but was significantly higher after 24 months on ART (76% v 67%). This was mainly due to the difference in loss to follow-up (13% at health centres, 25% at hospitals).

Table 3

Outcomes of cohorts of patients treated at health centres and hospitals in Ethiopia (2009)

Variables	Health centre (n = 25)				Hospital (n = 30)
Cohort^* period (months)	0	6	12	24	0	6	12	24
Number of patients	6206	6197	4022	856	31929	31269	23039	6595
Number of deaths	—	384	314	96	—	1513	1402	523
Number of patients lost to follow-up	—	741	666	110	—	4935	4600	1653
Number of patients retained in care and treatment	—	5072	3042	650	—	24821	17037	4419
Mortality rate (%) (95%CI)	—	6 (5.4-6.6)	8 (7-8.8)	11 (9-13)	—	5 (4.7-5.2)	6 (5.7-6.3)	8 (7-9)
Lost to follow-up rate (%) (95%CI)	—	12 (11.2-12.8)	17 (15.8-18.2)	13 (11-15)	—	16 (15.8-16.2)	20 (19.4-20.5)	25 (24-26)
Mortality rate and lost to follow-up (%) (95%CI)	—	18 (16-20)	25 (23-27)	24 (20-28)	—	21 (20.5-21.3)	26 (25-27)	33 (32-35)
Retention rate (%) (95%CI)	—	82% (80-84)	75% (73-77)	76% (72-80)	—	79% (78.7-79.5)	74% (73-75)	67% (65-68)
Median CD4-cell count (IQs)	117 (64,188)	223 (173, 328)	254 (206, 370)	322 (242,414)	119 (67,190)	225 (150,326)	275 (185, 355)	301 (217,411)

The cohort enrolment started in September 2006 and ended in August 2008. The shortest and longest cohort periods were 6 and 24 months

CI = confidence interval; IQ = interquartile

Mortality was higher at health centres (11%) than at hospitals (8%), but this difference is not statistically significant. When we adjusted for the loss to follow-up based on the findings from a previous study conducted in Ethiopia, loss to follow-up and mortality were estimated to be 16% and 18% at hospitals, and 6% and 18% at health centres.

The proportions of patients in whom CD4-cell counts were documented were similar at health centres and hospitals: 89%, 54%, 51% and 51% of patients at baseline and after 6, 12 and 24 months on ART. The median CD4-cell counts among patients at health centres and hospitals were similar at all time points.

Attitude of stakeholders towards new ART delivery models

Focus group discussions with patients elucidated that nurse and HO services were generally well-accepted, and that it had reduced the waiting time. Patients revealed that they were more comfortable with nurses than with physicians because nurses were friendlier and more supportive. They emphasized that HOs and nurses spent more time with them discussing their medical problems and took sufficient time examining them.

Patients identified three additional benefits of being involved in ART delivery: their life experience helped them provide appropriate counselling; it helped combat stigma and discrimination in society; and it provided them with an opportunity for employment.

Focus groups with AIDS programme managers and health care providers indicated that the new ART delivery model was unanimously accepted as a timely solution in the Ethiopian context. They also indicated that HOs, nurses and CHWs can provide high quality services if they are well trained, mentored and supervised regularly. One of the programme managers said: ‘it wouldn't have been possible to increase access to ART without the engagement of HOs and nurses in ART-related service delivery'. A hospital physician said: ‘HOs and nurses are able to stage patients based on the WHO criteria and initiate them on ART accordingly’. He added that task shifting is the right choice at the right time when there are a lot of patients waiting for treatment while the treatment is available free of charge.

Enablers and barriers to the new ART delivery models

Four enabling conditions were identified by the programme managers, for the smooth implementation of the new ART delivery models: the relatively high number of nurses in the country; the adoption of the public health approach to ART scale-up by the National AIDS Program; the large scale training of nurses and HOs on ART by the National AIDS Program and partners like PEPFAR and WHO; and the mentorship both in hospitals and health centres.

They highlighted that there are many more nurses than physicians in the country providing an opportunity if nurses are trained properly. They added that the public health approach had been beneficial and training and mentorship had facilitated task-shifting and ART scale-up. Mentorship had been organized in such a way that hospital physicians visit health centres for consultations on a regular basis to check how nurses and HOs are performing their task, and also to provide support for difficult cases. HOs with two years experience of managing patients are also providing mentorship to nurses in other health centres. Nurses and HOs can also solicit support from physicians by phone.

Five barriers were also highlighted. First, a lack of a regulatory framework with no legislation that allows nurses to perform certain tasks such as ART prescription. Second, nurses complained that they were not remunerated for the additional clinical tasks they were performing. Third, CHWs were either voluntary workers or contract workers but none were permanent employees in the formal health system. Fourth, the high cost associated with training and mentorship was a concern. Fifth, and most importantly, all respondents expressed their concern that the workload of ART, both in training and service delivery, had diverted many staff away from the general health services.

Discussion

Main findings

After 24 months on ART, retention of patients in care is significantly higher at health centres than at hospitals, and the difference was mainly due to the higher loss to follow-up at hospitals. Median CD4-cell counts at health centres and hospitals are similar. Service delivery by HOs, nurses and CHWs is accepted by patients, service providers and programme managers.

ART programmes in many countries have already begun reassigning tasks to low- and mid-level health workers to overcome the shortage of physicians. This includes the involvement of nurses in the evaluation of patients eligibility for ART, ART prescription for uncomplicated cases and trained CHWs including patients, providing adherence counselling.^20,21 Task-shifting in Ethiopia concurs with the experiences of other countries which have instituted delegation of ART-related tasks.^14,22,23

The quality of care delivered by mid-and low-level health workers has long been a contentious issue. Our study showed that patient outcomes in terms of mortality were comparable with physicians' care after 24 months on ART. Patients at health centres and hospitals had similar baseline median CD4-cell counts. Moreover, the loss to follow-up rate was significantly lower at health centres (13%) than at hospitals (25%). As a result, overall retention was better at health centres (76% versus 67%). Similar results were reported from other sub-Saharan African countries where patient outcomes were similar in sites with and without physicians.^17,24–26

Patients on ART welcomed services provided by nurses and HOs with some patients preferring them to physicians. Similarly, reports in South Africa show that nurse-led ART care models were feasible and acceptable to patients.²⁷ Reports from countries such as the USA, the UK and Australia indicate that the quality of care and patients' satisfaction are similar in nurse- and physician-based services.²⁸ As a result, there is a growing interest in the use of physician assistants and nurse practitioners in these high-income countries.²⁹

Strengths and limitations

Our study has several strengths and weaknesses. On the positive side, it included many health facilities and patients on ART and used both processes and outcomes to evaluate the acceptability and effectiveness of task-shifting. The study was based on data from a programme setting; hence, it is likely to reflect the reality on the ground. The study also analysed both the enabling conditions and gaps that might affect the sustainable implementation of the new ART delivery models in Ethiopia.

The study has three limitations. First, the facilities were not selected randomly but on the basis of number of patients receiving treatment in the facility and regional representation. However, all regions of the country were represented, the facilities included in the study were composed of large, medium and small hospitals and health centres, and the majority of patients started on ART were getting services from these health facilities. Hence, the findings from this study can be generalized to other facilities in the country. The second limitation is that we did not investigate the outcome of patients who were lost to follow-up, but instead used the findings from a previous study to adjust our results. We believe that the findings from the previous nationwide study can be applied to our current study as it is a big nationwide study that represents all health facilities providing ART in the country. The third limitation is that the CD4-cell counts were not documented for all patients and those without CD4-cell counts might have differed in terms of baseline characteristics and outcomes.

Implications

This study has indicated that ART delivery in health centres, based on HOs and nurses, is effective and acceptable in Ethiopia. It is important that the country capitalizes on the enabling conditions and addresses the weaknesses, such as lack of appropriate regulatory mechanism, inadequate financial incentives and career structure for low- and mid-level health workers, and the need for training and mentorship.

We recommend the following for future research: a cost-effectiveness analysis of ART delivery models with different degrees of task-shifting, including drug dispensing by CHWs; and the effect of ART training and services on general health services. New ART delivery models should not be viewed as a panacea but as a part of an overall comprehensive strategy that includes the production, deployment and retention of health workers.

Footnotes

Acknowledgement

The study was made possible thanks to a grant from the framework agreement between the Belgian Directorate General for Development Cooperation and the Institute of Tropical Medicine, Antwerp, Belgium.

References

World Health Organization, UNAIDS, UNICEF. Towards Universal Access: Scaling up Priority HIV/AIDS Interventions in the Health Sector: Progress Report 2008. Geneva: WHO, 2009

World Health Organization, UNAIDS. Progress Report on Global Access to HIV Antiretroviral Therapy: A Report on ‘3 by 5’ and beyond. Geneva: WHO, 2006

Mbewu

. Changing history: Closing the gap in AIDS treatment and prevention. Bull WHO 2004; 82: 400–01

Kober

, Van Damme

. Scaling up access to antiretroviral in southern Africa: who will do the job? Lancet 2004; 364: 103–07

Chen

, Evans

, Anand

. Human resources for health: overcoming the crisis. Lancet 2004; 364: 1984–90

Van Damme

, Kober

, Laga

. The real challenges for scaling up ART in sub-Saharan Africa. AIDS 2006; 20: 653–56

Jaffar

, Govender

, Garrib

Antiretroviral treatment in resource poor settings: public health research priorities. Trop Med Int Health 2005; 10: 295–99

Van Damme

, Kober

, Kegels

. Scaling-up antiretroviral treatment in southern African countries with human resource shortage: how will health systems adapt? Soc Sci Med 2008; 66: 2108–21

World Health Organization. A public health approach for scaling up antiretroviral (ARV) treatment: a toolkit for program managers. Geneva: WHO, 2003

10.

World Health Organization. A public health approach to antiretroviral treatment: overcoming constraints. Geneva: WHO, 2003

11.

World Health Organization. Scaling up antiretroviral therapy in resource limited settings: treatment guidelines for a public health approach (2003 revision). Geneva: WHO, 2004

12.

World Health Organization. Treat-Train-Retain. Taking stock: Health worker shortages and the response to AIDS. Geneva: WHO, 2006

13.

World Health Organization/UNAIDS/PEPFAR. Task Shifting: Global Recommendations and Guidelines. Geneva: WHO, 2008 Available online at: http://www.who.int/healthsystems/task_shifting/en/index.html (last accessed 29 April 2010)

14.

Dovlo

. Using mid-level cadres as substitutes for internationally mobile health professionals in Africa. A desk review. Human Resources for Health 2004; 2: 1–12

15.

Chilopora

, Pereira

, Kamwendo

Postoperative outcome of Caesarean sections and other emergency obstetric surgery by clinical and medical officers in Malawi. Human Resources for Health 2007; 5: 17

16.

Coetzee

, Hildebrand

, Boulle

Outcomes after two years of providing antiretroviral treatment in Khayelitsha, South Africa. AIDS 2004; 18: 887–95

17.

Sherr

, Micek

, Gimbel

Quality of HIV care provided by non-physician clinicians and physicians in Mozambique: a retrospective cohort study. AIDS 2010; 24: S59–S66

18.

Assefa

, Kloos

. The public health approach to ART service scale-up in Ethiopia: the first two years of free ART, 2005-2007. Ethiop Med J 2008; 46: 401–6

19.

Federal HIV/AIDS Prevention and Control Office (Ministry of Health) of Ethiopia. Antiretroviral treatment scale-up in Ethiopia; Success and challenges. Addis Ababai: MoH, 2009

20.

Assefa

, Jerene

, Leulseged

Rapid Scale-up of Antiretroviral Treatment in Ethiopia: Successes and System-wide Effects. PLoS Med 2009; 6: e1000056

21.

World Health Organization. Scaling up HIV and AIDS care: service delivery and human resource perspectives. Geneva: WHO, 2004

22.

The President's Emergency Plan for AIDS Relief. Report on work force capacity and HIV/ AIDS. Report submitted by the office of the US Global AIDS Coordinator, US Department of State, 2006

23.

Vasan A, Kenya-Mugisha N, Seung KJ, et al. Agreement between physicians and non-physician clinicians in starting ART in rural Uganda. Human Resources for Health 2009; 7: 75

24.

Bedelu

, Ford

, Hilderbrand

, Reuter

. Implementing Antiretroviral Therapy in Rural Communities: The Lusikisiki Model of Decentralized HIV/AIDS care. JID 2007; 196: S464–S468

25.

Wood

, Fox

, Conrade

Nurse management is not inferior to doctor management of ARV patients: the CIPRA South Africa randomized trial. Available at: http://www.ias2009.org/pag/Abstracts.aspx?AID=3822 (last accessed 29 August 2009)

26.

Kaplan

, Bekker

, Zwane

Long term programmatic outcomes for adults and children at a primary health care antiretroviral clinic in South Africa. Available at: http://www.ias2009.org/pag/PosterExhibition.aspx (last Accessed 29 August 2009)

27.

Aiken

, Lake

, Semaan

Nurse practitioner managed care for persons with HIV infection. Image J Nurse Sch 1993; 25: 172–7

28.

Wilson

, Landon

, Hirschhorn

Quality of HIV care provided by nurse practitioners, physician assistants, and physicians. Ann Intern Med 2005; 143: 729–36

29.

Horrocks

, Anderson

, Salisbury

. Systematic review of whether nurse practitioners working in the primary care can provide equivalent care to doctors. BMJ 2002; 3245: 819–23