Task shifting — the rational delegation of health care tasks usually performed by more highly trained health personnel to those with less training — is being widely hailed as a major part of the solution to the health care worker shortage that threatens to cripple further expansion of HIV prevention, care and treatment as well as other essential health services in resource-limited settings.

This should come as no surprise to HATIP’s readership community, many of whom have pioneered the very task shifting activities that WHO and others are now recommending taking to national and/or global scale.

Over the last five years HATIP has described how many activities can be performed quite well by nurses, community health workers, people with HIV or other non-medical staff — such as the home-based AIDS care (HBAC) programme in Uganda, nurse-run clinics in Lesotho, or TB screening by voluntary testing and counselling staff. Many of these innovations were born out of necessity, often initiated by the community itself to facilitate access to essential services.

And according to many presentations at the 2008 HIV Implementers’ Meeting and AIDS 2008, task shifting not only improves access to care, it can be performed without sacrificing quality of care — it may even improve the quality of care —and can get high marks in patient satisfaction. It can also be a more efficient and cost-effective way to provide services — but in the end, there may be little choice, because it may realistically be the only way to provide services needed by people with HIV quickly enough.

“Task shifting, from an economic perspective, is division of labour and the division of labour comes with efficiencies because each healthcare worker is focusing on specific tasks that they do best,” said Dr Kate Tulenko of the World Bank’s Africa Health Workforce Program during a symposium on task shifting at AIDS 2008. “But this is not something that’s exclusive to HIV and AIDS; it’s a health reform that’s been long needed in many countries and finally the emergency of AIDS is pushing it through.”

During a health crisis, the fastest route to staffing the health services will be increasing the number of staff who don’t take long to train. For instance, it can take seven years to produce a doctor, but (depending on the country) a few less years to produce a clinical assistant, less time to produce a nurse, and even less time to produce a community-based worker who may be trained for a few months or less to perform one specialised task — such as HIV testing and counselling, default tracing, adherence support — within a team of other health workers.

So the question is not whether to task shift, but rather which tasks to shift, to whom, and how to do it. The outcomes of task shifting can vary greatly depending on training, support and a number of factors — and there are limits to the tasks that can be shifted. So as more countries begin to scale up task shifting, it is crucial that we set up programmes with appropriate training, guidelines and standards based on what is currently known about task shifting.

Treat Train Retain’s global recommendations and guidelines on task shifting

The new WHO guidelines on task shifting, developed in collaboration with UNAIDS and PEPFAR provide a good place to start.

“Good health service coverage depends on having an adequate number of people, adequately trained to provide services,” Dr Francesca Celletti of WHO’s Department of Human Resources for Health, said during a presentation on the task shifting recommendations at AIDS 2008.

“For people living with HIV, in 94% of the cases, the first contact with health systems is not with a doctor but with a nurse, or a social worker or a community health workers (according to a Partners in Health study in Haiti),” she said. “So the question becomes how to empower those cadres to provide good quality services and to make sure that the service users are safe and the health workers protected.”

The new guidelines grew out of a process to document existing practices of task shifting in several sectors, and to ascertain key lessons learned, asking several key questions about each example of task shifting.

What is the impact of task shifting on the coverage of services and on the quality of the services provided? Was task shifting cost-effective? And was it a practice that is acceptable to the service user?

Coverage: There were many examples where task shifting positively impacted on the coverage of the services. For instance, in Ethiopia, access to HIV testing and counselling jumped dramatically from 500,000 people tested to 1,600,000 in 2007 after community health workers began performing the services. Likewise, in one district in Tanzania, there was a dramatic increase in the use of oral contraceptives when trained community health workers were engaged by the Family Planning Clinic and attendance at the clinic increased nearly five times within the same period.

Quality of services: “We also need to move away from the debate of ‘task shifting as being equal to second class care,’ as task shifting has been widely applied in high income countries like the US, UK, France, New Zealand,” said Dr Celletti, As an example, she mentioned a study from the US where physician’s assistants performed similar to or better than ID physicians or HIV expert general practitioners when delivering ART to people with HIV (Wilson et al, Annals of Internal Medicine, 2005). And in Brazil, community health workers were trained to perform immunisation and provide oral rehydration therapy to children. This resulted in sudden increase in the uptake of these services, with reductions in malnutrition, hospitalization and a profound decrease in the infant mortality rate (Cesar  et al, Social Science and Medicine, 2000).

Cost-effectiveness: “WHO, in its routine monitoring of TB programmes, has shown evidence that a decentralised approach to TB management, provides equal or better outcomes than a doctor-based conventional approach,” said Dr Celletti. “But WHO data also show that a decentralised approach for TB management is cost-effective in Kenya, Malawi, Uganda and South Africa, with the same effectiveness of treatment outcomes at a cost that is at least 50% less”.

Several other speakers at same symposium addressed the issue of cost-effectiveness. It seemed to be an issue with which Dr Louise Ivers of Partners of Health was clearly uncomfortable when talking about task shifting approaches in Haiti.

“We didn’t choose a model of using community health workers because it was a cost-effective model. We didn’t do a cost-effectiveness analysis beforehand. We used that model of care because we knew that in rural, isolated settings this model is going to work. It was going to get services to people who otherwise did not have services and furthermore there were no doctors available to provide the services,” she said.

“In using community health workers and using a model of task shifting, we anticipated that there would be an increase in the total health service use. So task shifting to community health workers should be in no way seen, and is in no way seen, as a cost-cutting exercise. We pay community health workers and we expected that as we provided services, the demand for services would increase. In fact, the objective of our model was that the demand for services would increase,” she said (more on the Partners for Health Project below).

“It’s not a cost-saving measure from an entire health sector point of view, but from a per-patient-visit point of view, it definitely can be cost-effective, said Dr Tulenko. “Because in the task shifting team, you have healthcare workers such as frontline and midline healthcare workers whose salaries are lower than the professional cadres such as physicians and doctors. And because of this, your per-visit cost for a patient will be less. The overall wage bill will be greater because you actually have more employees.”

Satisfaction: Dr Celletti referred to one of Dr Iver’s studies of task shifting in the central plateau of Haiti as an example of high satisfaction with the services provided by community-based workers. In that study, 97% (67% extremely satisfied, 28% very satisfied) of the clients were satisfied by the services provided under a task shifting approach.

These examples are just the tip of the evidence base used to develop the task shifting recommendations (and there were many more reports on task shifting projects reported at the recent conferences, see more below).

The task shifting recommendations

There are 22 recommendations divided into 5 categories, which can be read in full at http://www.who.int/entity/healthsystems/TTR-TaskShifting.pdf.

A)   Adopting task shifting as a public health initiative

Countries should consider implementing or extending task shifting where access to HIV services and other essential services is constrained by staff shortages, in combination with other efforts to increase the workforce. All relevant stakeholders, including the service user (people with HIV) and professional associations, should be engaged in the dialogue from the beginning; and a national framework developed to harmonise and stabilise services across public and other sectors. These efforts should be based on accurate data on human resources for health based on current demographic data, information on available services, gaps in services and existing quality assurance mechanisms.

B)     Creating and enabling regulatory environment for implementation

Existing legislation may need to be revised for cadres to take on other tasks (such as giving nurses the right to prescribe ART) or to create new cadres within the health workforce. Essential/emergency regulatory revisions may need to be fast-tracked while long-term comprehensive reform is underway.

C)   Ensuring quality of care

Quality assurance mechanisms will need to be adapted or created to support task shifting that can monitor and improve the quality of the services provided. Roles and competency levels have to be clearly defined for each cadre (existing or new) taking on new tasks, and these standards should provide the criteria for recruiting, training and evaluation. Countries should systematically develop standardised competency-based training programmes to meet their needs and to provide accreditation that the health worker is equipped to perform their tasks. Training programmes and continuing education should be tied to standardised, national certification, registration and career progression mechanisms. All health workers should receive ongoing mentorship and supervision within their health teams — and those providing supervision should be competent and have supervisory skills. Finally, it must be possible to assess the performance and competency of each health worker in each cadre.

D)   Ensuring sustainability

Countries should consider providing health workers taking on new or increased responsibilities with incentives, financial or otherwise to encourage retention and enhance performance. Also, while short-term volunteer work can be helpful, a sustainable health programme cannot be based on volunteer workers. Trained health workers, including those who are community-based, who are providing essential services should receive adequate compensation. Task shifting plans should be appropriately costed and adequately financed in order to be sustainable.

E)    The organisation of clinical care services

Countries should choose to adopt, adapt or extend those task-shifting practices that best suit their situation and needs. Functional referral systems need to be in place to support decentralised services that may be task shifted, and health workers need to know the referral systems well and how to use them.

The remaining recommendations refer to the tasks, listed in a long annex in the guidelines, that can be safely and effectively performed by the four or five cadres of workers: 1) non-physician clinicians (clinical officers/medical assistants) 2) nurses/midwives 3) community health workers and nursing assistants, 4) people with HIV and 5) Other cadres, including pharmacists, technicians or records managers who could theoretically be engaged in taking on new tasks as well. 

Essentially, many tasks can be shifted down from the more highly trained cadre to the one just beneath it, until people with HIV are reached, who the guidelines - somewhat weakly - state can be “empowered to take responsibility for certain aspects of their own care” and potentially help support others, particularly in regards to “self-care and stigma and discrimination.”

However, the engagement of people with HIV can do much more than that — historically, they have been responsible for many of the innovations in service delivery that have led to new cadres of health workers, devising adherence support programmes, home-based care programmes, prevention campaigns, etc. Of course, these roles become formalised over time, with training and standards — so that commonly people with HIV become community health workers. But it would have been nice if the guidelines had captured that a bit better.

Defining tasks and how far they can be shifted — the experience in Haiti

As mentioned, the laundry list of tasks that could possibly be shifted are listed in a long annex in the guidelines, but many of these came from a list that Dr Ivers helped compile in Haiti.

There are about 8.5 million people living in Haiti but only 730 doctors and 1013 nurses working in the public sector. Partners in Health works in collaboration with the Ministry of Health and one of its focuses is the delivery of HIV at the primary health care/community level.

“Most of our staff are community health workers; we call them “accompagnateurs” in Creole or in French, meaning “the one who accompanies,” said Dr Ivers. “Accompagnateurs are community health workers from the area where we work, living in the communities where we work, neighbours of our patients and clients who bring medicines to people everyday.”

Incidentally, she added that there is something of a tradition of community health workers in Haiti. There are health agents called ‘Ajan fanm/Ajan sante’ dedicated to the care of women or trained to do vaccinations; there are counsellors drawn from the community and trained; many of the lab technicians are people from the community without a high degree of education, but who have been trained to assist in the labs. Similarly there are pharmacy technicians, x-ray technicians, social work assistants, data clerks and medical records clerks — many drawn from the community to provide services.

WHO commissioned Partners in Health to do a survey into task shifting, in which they asked many HIV service providers about the distribution of HIV care-related tasks and what tasks were usually exclusively performed by certain cadres. They came up with a list of about 140 HIV-related tasks, about 50% of them exclusively done by doctors, about 20% exclusively done by nurses and so on.

“Community health workers were not really featuring at all in the traditional model of HIV care,” she said. “However, when we surveyed our sites in Haiti, we found that our distribution of tasks had really shifted [so that most of the tasks were being] performed by doctors, nurses, community health workers and by other non-clinical staff.”

In fact, only 28% of the tasks were exclusive to doctors or nurses in Haiti.

However, the ART programme is nurse-centred in Haiti, and the tasks are shifted rationally by type, with the largest shift seen in the management of patients just prior to and after starting ART. Nurses are responsible for prescribing and managing patients. Only a few tasks are not done by nurses, including starting TB therapy in patients with HIV/TB coinfection or in smear-negative TB cases; the definitive management of some complicated side-effects and complicated opportunistic infections.

Community health workers handle much of the pre-ART care, and provide facilitated referrals if there are signs of opportunistic infections or side-effects of ART.

“Community health workers provided a support system in the community,” she said. “They referred patients who had side-effects from drugs, they supervised therapy of drugs, they were an advocate for patients coming to the clinic to say ‘My patient is not well.’ Their referrals are taken seriously, they consult with the physicians about their patients, they consult with the nurses formally about their patients; and in those referrals, in consultations, they are respected. They are advocates for their patients. They will come and tell us that the roof is leaking, this person has no food in the house and really far more beyond just delivering medications to the patient everyday.”

How did this affect care in the programme? Although the evidence doesn’t come from a randomised controlled study, it is easy to interpret. In over 3000 people on ART, some started as long as 10 years ago, only 2% have needed to change to second line treatment, and despite significant political instability.

 “This is because the community health workers living in the community, they’re the neighbours, they still go to work everyday regardless of what was happening politically in the country," she said.

In addition, shifting tasks to nurses transformed clinics that were previously non-functional. Before task shifting, these clinics received on average only 10-50 visits per day. There were frequent stock-outs, no ART, absent staff, and HIV testing was only available at one stand-alone site that performed less than 40 tests per year.

Since introducing the nurse-led, community worker-supported model, each clinic receives an average of 200-300 visits per day. There are 50,000 voluntary HIV tests per year, over 8,500 people with HIV being followed and over 3,000 people on ART.

And as already mentioned above, patient satisfaction is very high.

Dr Ivers also stressed that the approach can be scaled up in other settings. Since 2005, Partners in Health began expanding this model to eastern Rwanda. In a short period that programme was able to enrol over 2500 people living with HIV on ART. They trained and hired over 800 villagers as community health workers and had nearly 100,000 visits in 2006, just one year after they started the programme.

MSF calls for scaling up task shifting

In a pre-conference satellite at AIDS 2008, Médecins sans Frontières, the organisation that pioneered delivery of antiretroviral therapy in some of the poorest communities in the world, repeated calls from the World Health Organization for more comprehensive attempts to shift tasks that could be done by community health workers and volunteers away from doctors and nurses.

But task-shifting is still happening inconsistently and has yet to realise its full potential.

If task-shifting is essential in achieving universal access and bringing down treatment waiting lists, Malawi’s Thyolo district illustrates how it could contribute. Thyolo, like the rest of Malawi, has around 1.3 doctors per 100,000 people, compared with the WHO minimum recommendation of 20 doctors per 100,000, and is even shorter of nurses – 17 per 100,000 compared with 28 per 100,000 across Malawi as a whole. WHO recommends a minimum of 100 per 100,000.

Marielle Bemelmans, MSF head of mission in Malawi, described what happened when task-shifting was implemented in Thyolo.

Voluntary testing and counselling was handed over to lay counsellors and the number of testing sites was expanded, leading to an increase in HIV tests from 15,000 in 2003 to 78,000 in 2007.

It has been estimated that somewhere between 9,000 and 12,000 needed antiretroviral treatment in Thyolo district by 2008. By June 2008 13,702 were on treatment – universal access in a rural area with a severe shortage of health care workers. How was this achieved?

Antiretroviral therapy inititiation was handed over to medical assistants, clinical officers and nurses after training, with follow-up by lay counsellors. Drug dispensing was shifted to patient attendants and the number of pre-treatment initiation visits was reduced, as were follow-up visits for stable patients. People with HIV, working as volunteers, have been involved through their support groups in pre-packaging of drugs, patient registration and referral.

Indeed, the only stages of the care pathway now being handled by nurses are clinical screening and staging, the assessment of opportunistic infections and nutritional status, and the initiation of cotrimoxazole prophylaxis. Volunteer usage – currently 5 – 8 people per clinic day – means that the district has needed to employ only 7 extra nurses to accommodate the growing patient caseload, not 27 as originally projected.

New starts on treatment have more than doubled since 2005, with around 40% of patients now being initiated at local health centres. Comparison of outcomes between patients who started treatment at a hospital or a local health centre showed no significant overall difference in retention in care, but higher death rates were seen in health centres and higher loss to follow-up rates in hospital patients.

“It would have taken five years to achieve universal access in the district without the community network,” said Marielle Bemelmans.

Task shifting in Kenya

Another example of task shifting is the recent report from Kenya, summarised in more detail in an aidsmap news report from AIDS 2008, which showed that a system of `express care` could be developed that devolved much of the burden of care of patients on ART to nurses. The model led to better outcomes for patients, the investigators reported.

AMPATH developed its model to deal to deal with two categories of patients utilising clinic resources most intensively - those with CD4 counts below 100 cells/mm³, who tended to be ill more often and have a higher risk of loss to follow-up, and those on ART who were clinically stable.

Express care reduced loss to follow-up, and reduced the risk of death by nearly 50%, demonstrating that task shifting does not have to result in poorer outcomes. By shifting tasks, it may be possible to improve the quality of care for the most vulnerable as well as increasing the volume of patients who can be treated.

Conditions for success of task shifting

Dr Wim Van Damme of the Royal School of Tropical Medicine, Antwerp, told the meeting that his research and analysis shows there is a series of conditions for success – but lessons learned are being ignored in many settings.

Firstly, selection and motivation: are volunteers motivated and committed? Evidence from Malawi, where the government is attempting to recruit 4,000 Health Surveillance Assistants to work at community level suggests that rigorous selection is being sacrificed in the attempt to meet recruitment targets.

Secondly, initial training is key in ensuring quality. In Ethiopia, where community health workers are being recruited as part of a national task-shifting initiative, Health Extension Workers receive one year of training. In Malawi Health Surveillance Assistants are trained for up to ten weeks.

Another condition for success, he said, is the need for standardised protocols and realistic job descriptions. This condition is being met everywhere he has looked.

“Supervision, support and supply of materials to do the job are also important, but support and supervision are very problematic in most treatment programmes. Paid staff don’t have time to supervise volunteers due to their heavy workload, so there’s no way to check that volunteers and community health care workers are following protocols. For example, in Ethiopia, referral to hospital by Health Extension Workers has not ben happening routinely,” he said.

Dr Van Damme also noted career structure and remuneration to be an important condition for success and he identified a number of conditions for scale-up of task shifting:

• Political support and a favourable regulatory framework that allows new cadres to take on tasks of existing health care workers: this is very much the case in Ethiopia and Malawi, but there is a lack of guidance from government in Uganda, leading TASO to establish its own regulatory framework to manage task-shifting in its operations, which provide home-based care in many localities.
• Alignment with broader health systems-strengthening activities such as expanding the doctor and nurse supply by investment in further education and a national plan for generating a long-term increase in the supply of doctors and nurses: once again, happening in Ethiopia and Malawi, but lacking in Uganda.
• Flexibility and dynamism: it remains to be seen whether task-shifting initiatives will permit the need to accommodate rapid changes, such as the move from voluntary counselling and testing to opt-out, provider-initiated testing in many settings over the past two years.

He also noted that task-shifting presents two opportunities for health systems:

• Using the real-life experience of patients, such as people living with HIV.
• Developing chronic care models that focus on adherence and retention in care: these are applicable across many diseases, not just HIV, and here Uganda is leading the way.

Remaining barriers to task shifting

Despite the ground swell of support for task shifting among the HIV community, there are a number of significant hurdles to scaling up task shifting in many settings. For instance, in South Africa, the South African National AIDS Council and a wide range of stakeholders are calling for changes in regulations and legislation that are barriers to task shifting and universal access. For instance, in South Africa lay counsellors are still not permitted to provide HIV tests (even using rapid test kits), and nurse are not yet permitted to initiate people on ART.

At a symposium on task shifting at the HIV Implementers' Meeting, there was a discussion about how professional associations are fighting task shifting in some countries.

“There are many misunderstandings about task shifting,” said one audience member. “Having global guidelines is not enough. There is a need for country-based endorsement and participation of local professional associations to accept the WHO guidelines. Another problem is that the link between the vocations and institutions and professional health services is very weak. And to certificate, to endorse, to promote the application of task shifting - there is a need for improving the link between the vocation and the institutions and health services. Some professional associations are more committed to the global health work force approach but others remain attached, protecting the turf of the professions.”

Others were concerned about the over-burdening of nurses, and that not enough was being done to protect the health of community-based workers particularly those with HIV (a future HATIP issue will address this topic).

Finally, there is a possibility that expectations of task shifting could be too high.

“With task shifting you can improve the efficiency of the work force but you cannot generate the kind of numbers that we need from a comprehensive primary healthcare approach - outside of this room, outside of this conference - we have a debate ongoing about funding for AIDS, funding for health systems and funding for global integrated primary healthcare. And the workforce to deal with this IS NOT there!,” said Dr Jos Perriens, Coordinator of the Systems Strengthening and HIV (SSH) unit in the HIV/AIDS Department at WHO, who moderated the discussion.

“Task shifting is not panacea, it is just one solution to the complex problems that are today causing the human resource crisis. It should be taken up along with other strategies such as the production of more healthcare workers and retention of the existing ones,” said Dr Celletti.

However, it is important that the HIV and TB communities engaging in task shifting perform ongoing operational research and document any quality improvement activities to help expand the evidence base of what does and doesn’t work — and why. Word should go out about task-shifting projects or methodologies that are working particularly well, and people should be advocates for taking their successful programmes up to scale.

References

Celletti F. WHO recommendations and guidelines on task shifting: The evidence, the content and the way forward for implementation. AIDS 2008, Mexico City, abstract MOSY0903.

Ivers LC et al. Task-shifting in HIV care: a nurse-centered, community-based model of care in rural Haiti. AIDS 2008, Mexico City, abstract WEAX0103.

Ivers L.  The community healthcare workers back on the scene for HIV and primary health care service delivery: The experience of Haiti and Rwanda. AIDS 2008, Mexico City, abstract MOSY0902.

Tulenko K. Task shifting: How much does it cost? Is it cost-effective? AIDS 2008, Mexico City, abstract MOSY0906.

Non-physician clinicians and nurses can take over many of the tasks in providing HIV care and treatment (including ART) in some resource-limited settings according to a number of reports at both the HIV Implementer’s Meeting in Kampala, Uganda and AIDS 2008 in Mexico City — and shifting tasks to them can free up doctors' time, and reduce the impact of HIV/AIDS programmes on the health system as a whole. Several of these studies also reported that the care was at least as good as that provided by doctors — with high client satisfaction.

However, for such task shifting to be sustainable, non-physician clinicians and nurses will need to be adequately compensated for their increased responsibilities, and some of their less usual tasks (paperwork, counselling, etc) will need to be shifted to less highly trained health workers. And such task shifting will not be the cure-all for most countries’ human resource crises — they will still need to train and engage more health care workers of all types.

The human resources for health crisis

There is a global shortage of medical professionals, such as doctors and nurses, that is especially severe in resource-limited settings. Doctors are in particularly high demand — too few are being produced, they take longest to train, and they are the most likely to accept positions for better pay in the private sector or move abroad to work in better resourced countries.

In many countries, doctors are the only cadre of health care workers allowed to prescribe antiretroviral therapy (ART), so the virtual absence of doctors in some areas (especially rural settings in poorer countries) makes it difficult or impossible to provide universal access to ART. Since it takes about seven years to produce new doctors, the only way in underserved areas to save potentially millions of lives is to empower cadres with less training to provide ART.

Non-physician clinicians

Non-physician clinicians go by a variety of names in different places, but “have established a history and effective presence in many countries,” according to Dr Seble Frehywot of George Washington University, who presented a survey at AIDS 2008 on the presence and use of non-physician clinicians in sub-Saharan Africa, with a special focus on how they can be employed in HIV programmes.

In the United States, non-physicians are called physician’s assistants or nurse practitioner, but in resource-limited settings, the most common nomenclature is clinical officer or health officer. Regardless of the name, these non-physician clinicians - healthcare providers with less post-secondary school training than a doctor but more than registered nurses - can to some extent do the work of a physician (sometimes without the supervision of a physician). They are trained in basic diagnosis and medical treatment to deliver a range of personal clinical health services.

Dr Frehywot said that their basic schooling takes one of two forms. Some countries take registered nurses or nurses with four-year degrees and add one or two years more training and then call them either nurse clinicians or nurse practitioners. Other countries take students right out of high school, give them three to four years of training, and an internship of 12 to 18 months. In the end, however, their function is essentially the same as those produced by the nurse model.

However, their curriculum is often defined by the local government, and tailored to the host country’s indigenous clinical needs. Because their training and certification is specific to each country, they cannot easily find positions abroad — and are much less susceptible to brain drain.

In addition, said Dr Frehywot, “we found, that when these clinicians are placed within the rural communities, they stay there.”

As for the cost for a work force comprised of non-physician clinicians, she said it was difficult to calculate in Africa because the data are not available. “What we have found is that mostly the cost ranges from US$1200 to US$4000 per year but this includes only their individual tuition and board.”

Non-physician clinicians are more commonly found in Anglophone Africa, although they do also exist in some French and Portuguese-speaking countries.

For instance, in Mozambique non-physician clinicians are called Tecnicos de Medicina.

“In Mozambique, we have three physicians to 100,000 patients,” said Dr Catherine McKinney of the US Centers for Disease Control speaking at the HIV Implementers' Meeting, “for a population of around 20 million.” She said that the Tecnicos de Medicina were introduced around the time of independence (33 years ago) but quickly took on increased responsibilities. “About 75% of the physicians left the country at the time of independence, so there was no upper echelon of health care, at that time, so clinical officers became the upper echelon,” she said.

This pattern has been repeated throughout sub-Saharan Africa.

Expanding the use of non-physician clinicians in HIV/AIDS

Most recently, Dr Frehewot said, “non-physicians have demonstrated an effective role in HIV and AIDS services,” with some countries such as Malawi, Ethiopia, Tanzania and Zambia building their ART strategies upon these cadres. In general, they are being used to perform HIV disease staging and diagnosis, to initiate preventive medicine for opportunistic infections, and to manage most of the uncomplicated clinical cases. Finally, they can initiate ART.

“But all this requires some kind of regulatory framework,” said Dr Frehywot. In other words, to shift the tasks from doctors to non-physician assistants, a number of changes must be made to the relevant laws/proclamations, regulations, policies, and guidelines in each country.

“There are a quite few things that the countries have to take into consideration,” Dr Frehywot said, including: the scope of practice and competencies for these healthcare workers; standards of care to be provided; the standard pre-service education and training that will be required, along with in-service training and certification; licensing registration and certification; supervision, mentoring and monitoring to make certain a high quality of care is provided; regulations about working conditions, how will they be recruited, deployed, and promoted, what they will be paid and other human resource issues; and finally, sub-national implementation, since many of these workers will be deployed in rural areas.

The same would hold true for shifting these tasks to nurses.

There are some barriers to increasing reliance on non-physician clinicians. “The cost is a still a little bit high and there are insufficient faculties and infrastructure for training these cadres. Finally, in some countries the medical profession shows resistance,” said Dr Frehywot, but she reiterated that non-physician clinicians can provide a high quality of care, and are a cadre of health workers that can be trained and deployed much more quickly than doctors, with a much higher rate of retention.

Tecnicos de Medicina provide high level of HIV service in Mozambique

There are currently only 600 Tecnicos de Medicina (TdMs) in Mozambique, but they prove a disproportionate part in clinical care. “These non-physician clinicians represent an effective strategy given their ease of deployment, longevity at post, shorter training time, more established training capacity, and lower cost for training and maintaining,” said Dr Kenneth Gimbel-Sherr of the University of Washington and Health Alliance International, at another presentation at AIDS 2008.

Dr Gimbel-Sherr said that evidence from US suggests mid-level non-physician clinicians perform as well as or better than MDs (including for HIV care) (Wilson), but that evidence from resource-limited settings was mostly anecdotal.

“This lack of evidence leads to continued reticence among policymakers to allow non-physicians to prescribe ART and provide continued care for patients on ART,” he said. “Much of this reticence centres around concerns about quality of care, and the lack of mentoring and ongoing support provided to NPCs – many of whom work in isolated health facilities.”

So Dr Gimbel-Sherr and colleagues performed a retrospective cohort analysis of outcomes among patients in the ART programme over a 3.5-year period (starting with the initiation of the national ART programme in July 2004) attending two HIV clinics along the Beira corridor in central Mozambique, the area with the highest and most mature epidemic in the country. The study compared outcomes in patients who were attended by TdMs and doctors.

The clinics both have a high patient load, with an average of over 900 new HIV-positive enrollees per month, an average of over 225 patients initiating ART per month, and nearly 5,000 clinical consultations per month across the clinics. There are more doctors (28) than TdMs (14) but the TDMs were more likely to be present in the clinics. So, out of the nearly 6,000 patients included in the analysis, most (69.4%) were treated by TdMs and only 30.6% by doctors. There were no significant differences in baseline characteristics (gender, baseline weight, CD4, clinical stage, etc).

Likewise, the outcomes seemed similar. Those patients seen by TdMs were more likely to be lost to follow-up, but less likely to have died, but these outcomes were again insignificant in the adjusted analysis. However, patients seen by TdMs were significantly more likely to have had a quarterly visit with a clinician for at least three out of four quarters after starting ART — probably because the TdMs were more likely to be present in the HIV clinics— and significantly more likely to have optimal adherence six months after starting ART initiation, as defined by pharmacy records.

The study has a number of considerable limitations. For instance, as one statistician in the audience pointed out, it doesn’t meet the standards of a randomised controlled clinical trial. However, it represents a start in documenting the quality of service that can be provided by supervised non-physician clinicians, which Dr Gimbel-Sherr concluded seems “equivalent to or slightly better than that provided by MDs.” However, he added that “results from large centralised sites with more supervision may not apply to smaller remote sites with less supervision.”

The information should be reassuring to Mozambique, which in 2006 embarked on a plan to produce thousands of TdMs to be deployed by 2010.

Shifting tasks to nurses in Rwanda achieves good results

However, it still takes quite a while to produce this new cadre of health workers, and for most countries to scale up ART programmes quickly, they have to turn to the health staffing resources they have — which is most commonly nurses.

Fortunately, “nurses can effectively perform when optimal conditions are met – and here we talk about good preparation, better tools, and consistent MD support,” said Dr Fabienne Shumbusho of Family Health International speaking at the HIV Implementers' Meeting in Kampala. She described a successful pilot programme in rural Rwanda in which nurses were trained and mentored by MDs to prescribe first-line ART for treatment-naïve adult patients at three primary health centres.

In most of Rwanda, the HIV programme is physician-centred. Doctors perform the physical exams, order CD4 count measurements, prescribe ART and provide follow-up of patients before and after starting ART. Nurses perform a supportive role, including requesting CD4 counting in some sites, and ordering refills of ART.

In the Family Health International pilot project, all the doctor’s primary tasks, except for the management of complex cases (including children) and the provision of mentoring and supervision, were shifted to the nurses who completed a course in HIV care and treatment. The pilot project, which ran from September 2005 to March 2008, included over one thousand patients, 435 of whom were started on ART by nurses.

The outcomes were very similar to what is observed in other settings in Rwanda where care is physician-centred with nearly 88% of alive patients on ART in March 2008, 6.7% dead and less than 3% lost to follow-up —with no cases of stopped treatment. At 18 months the probability of retention in care is 91%, and the median CD4 cell increase at six months was 140 cells, 166 cells at 12 months and 251 cells at 24 months.

The study also evaluated several aspects of nurse performance including:

• Was eligibility for ART correctly determined? Yes, the nurses didn’t put anyone on ART who wasn’t eligible, and only missed 4.2% of cases who should have been started on ART.
• Was the correct ART prescription made? Yes, with no patients put on drugs that were contra-indicated.
• Were ART patients adequately monitored through lab controls and clinical exams? Initially yes, though with longer follow-up, there was a decreasing amount of data.

“At the same time, we found that the level of completeness in this pilot compares favourably with data from a national evaluation conducted in 2004-2005 in Rwanda,” said Dr Shumbusho.

“There is room for improvement,” she concluded, but she recommended that the programme be scaled up nationwide. “We can achieve universal access to AIDS treatment by improving the skills of nurses to their full capacity and potential,” she said.

Task shifting to nurses frees up physician time and reduces impact of HIV programme on general health services

Indeed, a mathematical model presented at both the HIV Implementers’ Meeting and AIDS 2008, Joyce Chung, of the Center of Strategic HIV Operations Research and the Clinton Foundation HIV/AIDS Initiative found that if the nurse task shifting model was scaled up throughout Rwanda, it would dramatically increase the time that the country’s physicians have to tackle more complex cases.

The model included 976 patients from the FHI centres (only those patients enrolled by December last year). The model compared schedules and consultation times at one pilot clinic and one conventional clinic. Details of how the simulation model works can be found in Chung’s extra slides, in her powerpoint presentation on the AIDS 2008 site (use this link to download the presentation)

But overall, between September 2005 and December 2007, prescribing nurses expended a total of 942 hours, saving 737 hours of physician time. “This is equivalent to 6 months of a physician working 30 hours per week”, said Chung, “with one hour worked by a prescribing nurse saving approximately 47 minutes of physician time.”

This freed up time for more complex cases, such as treating children with HIV, and for work in non-HIV areas.

The latter finding is of critical importance because HIV programmes have been accused of stealing time away from general health services. By employing this task-shifting model however, even with ongoing scale-up of ART, Chung calculated that there would be increased physician capacity for the system as a whole.

“If we assume expansion to 59,000 people on treatment at the end of 2008, to deliver care in line with national guidelines would require around 9,300 hours of physician consultation time per month, or roughly 77 physicians providing direct patient care for 30 hours per week,” she said. “Since there are only about 150 physicians providing patient care in the public sector, this means that HIV care and treatment will absorb about 51% of the total physician capacity of the government of Rwanda by the end of this year.  But the application of task shifting nationwide over the course of the year would result in a 78% decrease in physician demand to just 17 physicians working 30 hours per week, or just 11% of the total public physician capacity. The amount of physician capacity available for complicated cases or non-HIV care is 183% higher with task shifting!”

Concerns and caveats

Despite the positive findings reported in the above studies some audience members at each conference expressed concerns that task shifting would be tantamount to second-class care. At the symposium in Mexico City, Dr Frehewot insisted this is not the case.

“If their pre-service training is really beefed up regarding psychology, pathology, medicine, microbiology, then they do give really good service. I can give an example: I’ve been living in the United States for about 15 to 17 years now. When I am sick — and I’m a medical doctor — I don’t like to treat myself. So I go to my family practitioner and most of the time, I see a nurse practitioner.  I’m a medical doctor but I’m seen by a nurse practitioner, which is excellent. Their training and everything is excellent. I have a five-year old daughter and when she has a cold or diarrhoea problem, and we go to a doctor’s office, most of the time we see a nurse practitioner or a physician assistant. We don’t see medical doctors — and this is in the United States,” she said.

Finally, another concern is that shifting these tasks to clinical officers and nurses would lead to over-burdened health workers — and some of their routine tasks “falling off the table.”

“That means, the work that nurses have done previously, is now not going to be done,” said one audience member from Malawi at the Implementers’ Meeting. “It’s really important to realise that task shifting — net gains from task shifting — can only happen if we manage to pull additional bodies into the system  —that’s shifting tasks to non-healthcare professionals who will then start to do HIV services.”

Chung emphasised that her study merely focused on clinician time saved by shifting tasks to nurses. But the model could similarly be used to assess the time saved to nurses by shifting tasks to community-based health workers trained to perform counselling and testing, recording and reporting and other tasks.

Dr Shumbusho stressed that nurses currently spend too much of their time doing paperwork that could be done by less highly trained staff. Thus shifting tasks to them doesn’t necessarily have to result in a heavier burden on them, and many nurses are only too happy to be doing more interesting work that they have, after all, been trained to do: “Nurse are already trained to do clinical examinations — we are not teaching them to do something they have never done.  We are just giving them the tools that will ease their tasks — and we are recognising what they are already doing,” she said.

References

Chung et al. Quantification of physician-time saved in a task shifting pilot program in Rwanda. XVII International AIDS Conference, Mexico City, abstract WEAB0205, 2008.

Gimbel-Sherr K et al. Task shifting to mid-level clinical health providers: an evaluation of quality of ART provided by tecnicos de medicina and physicians in Mozambique. XVII International AIDS Conference, Mexico City, abstract WEAX0105, 2008.

Frehywot S. Non-physician clinicians and HIV service delivery: Experiences from 45 countries. XVII International AIDS Conference, Mexico City, abstract MOSY0901, 2008.

Mullan F, Frehywot S. Non-physician clinicians in 47 sub-Saharan African countries. Lancet, 370(9605):2158-63, 2007.

Shumbusho F. Task shifting to achieve universal access to HIV care and treatment services in Rwanda : a nurse centered ART program at three rural health centres. HIV Implementers’ Meeting, abstract 621, 2008.

Wilson I et al.  Quality of HIV care provided by nurse practitioners, physician assistants, and physicians. Ann Intern Med 143;729-36, 2005.

Other resources

Public Sector Unions Fighting against AIDS (PSUFASA) hosts a resource map on HIV/AIDS responses in the public sector in Southern Africa on its website.

A comprehensive report on Malawi’s Public Health Sector can be found on the PSUFASA website.