Task-shifting of HIV care to nurses: successes, but problems to watch out for

“A nurse-led service can deliver ART care as effectively as a doctor-driven one, and even improves quality of care, but this pragmatic trial did not result in increased access to ART,” said Lara Fairall of the University of Cape Town. She was describing the results of the STRETCH study, a cluster-based randomised controlled study of nurse-managed and initiated antiretroviral therapy (ART) presented at the sixth International AIDS Society conference (IAS 2011), held in Rome this week.

Another presentation in the session suggested non-physician health workers could be mobilised to respond to AIDS in prevention services – notably male circumcision.

Meanwhile, an observational cohort study presented by Dr Megan McGuire of Epicentre, a public health research unit of Médecins Sans Frontières, suggested that HIV-care provided by nurses in Malawi was superior to that provided by clinical officers. However, as McGuire herself pointed, the clinical officers primarily managed the sickest patients with the most advanced disease. This was not an entirely fair comparison.

All these reports recommended task-shifting HIV care to nurses at the primary healthcare level. But the session also highlighted the tension at the conference between those lobbying for a quick shift to task-shifting and decentralisation in order to reach goals for universal access to HIV treatment, versus those warning that task-shifting may not be an overnight solution to the crisis in human resources for health care. Rather, there are indicators suggesting programmes must be aware that there can be gaps and problems in the transition for doctor-based to task-shifted care.

Nurses can make the cut

Several large, randomised studies have reported that male circumcision significantly reduces the risk of HIV acquisition among heterosexual men, and the World Health Organization has recommended that services offering male circumcision should be incorporated into national HIV prevention activities in high-HIV-burden settings. As a result of news reports and prevention campaigns, demand for the procedure outstrips supply in some sub-Saharan African settings.

“The shortage of health professionals poses a critical challenge to the male circumcision scale-up,” according to Dr Kelly Curran of JHPIEGO in Baltimore.

However, programmes have identified a number of potential approaches to expanding the numbers of health workers who can offer the intervention.

“The shortage of health professionals poses a critical challenge to the male circumcision scale-up,” Dr Kelly Curran

In Kenya, procedures and policy were adapted to empower nurses to safely perform male circumcision. After two campaigns, the programme has delivered 268,000 male circumcisions within 2.5 years.

In Swaziland, a survey was performed to identify existing nurses who were under-utilised: in other words, registered but unemployed nurses (many of whom were Zimbabwean), those recently retired or recently graduated, and often working abroad (mainly in the UK). At least 259 were found who could be trained to do the procedure.

Some public critics have been concerned that HIV services for treatment – not to mention prevention interventions, such as male circumcision campaigns – might draw healthcare staff away from other critical health care services, but Dr Curran pointed out that the Swaziland approach might remedy that to an extent.

“My lesson learnt from our efforts in Swaziland is that the first place where we should go is the unemployed healthcare workers, as well as the recently retired yet still energetic. In many of the countries where we work, retirement age is as low as 55 – many of those nurses would happily still be employed. They may not want to do surgery all day on their feet but they could happily do post-operative reviews or staff recovery rooms,” said Dr Curran. “So I think the key here is really to look everywhere you can first before engaging nurses or other healthcare workers away from service provision of other critical health interventions.”

Programmatic data on nurses and clinical officer-based care in Malawi

Malawi has a severe shortage of doctors, and yet over the last several years has ramped up a massive HIV treatment and care programme with very few resources. For example, in 2010, there were roughly 14,000 HIV consultations, 700 program enrollments and 400 ART initiations.

This was achieved with a mixed model of care, shifting ART initiation and management to clinical officers and nurses.

Dr Megan McGuire presented the findings of a study comparing treatment outcomes of 10,112 adults patients receiving ART between 2007 and 2010, who were cared for either by nurses, clinical officers or both in a large HIV programme in rural Malawi.

However, nurses were only supposed to care for patients who were antiretroviral naïve and starting on first-line treatment, with WHO stage 1 or 2 disease, CD4 cell counts above 100 and BMIs over 17. In practice there were some exceptions, but in general, patients cared for by nurses had less severe HIV disease and wasting than those cared for by clinical officers.

The nurses provided the majority of care for 1901 patients and the clinical officers 3386, while 4825 patients received mixed care.

At baseline, BMI was under 18.5 in 15.4% of those cared for by nurses, 35.4% of those cared for by clinical officers, and 25.9% who received mixed care.

Similarly, median CD4 cell counts at baseline were 195, 147 and 182, respectively, for the nurse, clinical officer and mixed-care cohort. The difference in severely symptomatic disease was more pronounced, with 19.1, versus 58.9 and 33.1% with WHO stage III/IV disease for the nurse, clinical officer and mixed-care cohorts, respectively.

Follow-up was right-censored at the earliest of the following dates: death, transfer out, last visit or 24 months of follow-up. Not surprisingly, more of those receiving care from clinical officers (the sicker patients) were around 5 times likely to die (around 20% vs. less than 5% for both nurses and mixed care) within 2 years of follow-up. Similarly about 40% of the clinical officer cohort was lost to follow-up, versus around 10% for the nurses, and just over 5% for those receiving care for both.

Acknowledging that some of this may have been due to ill health at baseline, the analysis was restricted to less severely ill patients (n=3846) with BMI>18.5, WHO stage 1 or 2, and CD4>100. The aIRR (95% confidence interval) for mortality was 3.42 (2.60-4.48) for those in the clinic officer cohort and 0.63 (0.47-0.86) for those receiving mixed care (versus those getting nurse-provided care).

However, this was an analysis based upon the health of the patient at baseline – and a serious long-term illness developing after entry into care (such as TB) would have led to the patient being referred to (and having more visits) with a clinical officer. Tellingly, CD4 cell counts were somewhat poorer for the clinical officer group at 12 months, but there was not much difference at 24 months.

It is possible that nurse care was indeed better for retention and health outcomes, but this was not a randomised study and cannot be used to reach that conclusion.

STRETCH

The STRETCH study (Streamlining Tasks and Roles to Expand Treatment and Care for HIV) was randomised – but because it was performed in real-world conditions, at a time and place (the Free State Province in South Africa) where nurses were in the process of being trained up to provide ART, it may not serve as the final word for whether nurses can provide ART care just as well as doctors or other clinicians.

Thirty-one clinics were randomised: 16 operated the STRETCH protocol, alongside 15 control clinics. The study had two parts: one (n=6321 patients) was to show whether nurse-led clinics could maintain patients who had been stable on ART for at least six months as well as clinics with more doctor-based care or support – as measured by viral load. This they could do easily.

The second part of the study evaluated whether task-shifting to nurse-led clinics would reduce mortality in people on ‘waiting lists’ for treatment by improving access and initiating ART among those with CD4 cell counts below 350 (9252 patients).

This study was unable to show this. However, Fairall pointed out that the nurse-led clinics began initiating ART gradually.

The results were previously presented at the 5^th South African AIDS Conference, and reported in much more detail in HATIP 179.

Discussion

Audience members asked whether it was too much to ask nurses to take on the additional burden of ART initiation and management – and whether this was reflected by STRETCH’s inability to increase access to ART during the study period.

“We did monitor [the adverse consequences of task-shifting on the nurses] during the trial. And yes, nurses are very burnt out. So one has to be very cautious about how and when one can delegate the clinical responsibility. So that really is ... a cause for consideration,” said Fairall.

"I think we’ve seen nurses as really having extremely broad shoulders and able to take on more and more and more. That does have consequences." Lara Fairall

“But going back to this concept of task-shifting – I think we’ve seen nurses as really having extremely broad shoulders and able to take on more and more and more. That does have consequences. The nurses in our trial didn’t seem to mind the new duties but they were well supported by our provincial co-ordinator, with whom they had a direct personal relationship,” she added.

Indeed, nurses need encouragement and support – and nurses should not be expected to handle the most complicated cases. Part of the STRETCH programme involves training nurses when to recognise the need to refer patients for doctor-based management.

The need for more specialised clinical support is recognised in Malawi as well – the goal is to be less reliant on it, however.

Dr McGuire said that they were now looking at whether it was possible to space out clinic appointments further, so that patients visit the clinic once every six months, or even once a year if they are stable.

Another important aspect of shifting more of the burden of care to nurses involves shifting some of their existing duties to less highly trained health workers, such as expert patients.

A number of other studies at IAS 2011 reported on how lay personnel can dramatically improve outcomes such as initiation of ART and retention in care. Conversely, the failure to make certain that such support services are available at the primary healthcare level could be one reason behind some of the disappointing results on the decentralisation of ART services reported during other sessions of the conference. Both these issues will be addressed in subsequent articles on www.aidsmap.com.