Which tasks were shifted and to whom?

Among the 15 studies funded by the GACD, 8 incorporated task-shifting from physicians to other cadres of the health workforce to manage hypertension. These studies were conducted in Argentina,14 Canada,15 Colombia, Ghana,16 India,17 18 Kenya,19 Malaysia, South Africa20 and Tanzania.15 All studies were randomised controlled trials, with one being a randomised controlled feasibility study. The cadres of NPHWs included nurses, community health workers,4 accredited social health activists (ASHAs)21 and community health extension workers (CHW supervisors). The NPHWs had between 8 and 14 years of education, while the nurses were graduates from formal nursing programmes. In four studies, tasks were shifted from physicians to NPHWs and in four studies, tasks were shared between two different levels of NPHWs (table 1).

The tasks shifted to NPHWs included screening of individuals, referral to physicians for diagnosis and management, patient education for improving lifestyle, titrating the dose of medications, follow-up and patient reminders for adherence to medications and appointments. Screening was undertaken by NPHWs in Colombia, Malaysia, India and Ghana (table 1). Three studies involved the use of mobile applications with clinical decision support tools to enable NPHWs to screen, manage and refer patients to physicians. In the SmartHealth India study, the mobile device also enabled blood pressure (BP) readings to be automatically wirelessly uploaded and included a treatment advice page that provided the 10-year CVD risk of the participant as well as guidance for ASHAs on providing, lifestyle advice and recommendations for referral.17 In the LARK study (Linkage And Retention to hypertension care in rural Kenya), community health workers were trained in a ‘tailored behavioural communication strategy’ and their mobile application had audiovisual tools such as illustrations/video19 to enable CHWs to link patients with hypertension to the health system and provide appropriate support. Decision support for the HOPE-4 (Heart Outcomes Prevention and Evaluation Colombia and Malaysia) study included, included diagnosis, guidance on initiating and providing appropriate combinations of CVD medications (statin and BP lowering agents) and counselling. The other tasks shifted to non-physicians included assisting in filing medical records, prepacking medications and reminding patients about their appointments using text-messages, telephone calls or personal visits (South Africa).20 Some interventions included delivery of medications14 and ongoing monitoring of BP.14 18 In Tanzania, a programme of cost-sharing through a mobile money platform and a health insurance scheme for low-income earners was introduced with the aim of increasing patient compliance with refilling antihypertensive medications and reducing cost to patients (as drug access through the government-funded health centre was unreliable).15 In the HOPE-4 study, CHWs in Columbia and Malaysia were able to initiate appropriate CVD medications as well as follow up patients at regular intervals. Trained community nurses in Ghana used the WHO PEN tool to titrate medications.16

There were some differences in the sites of care across the studies. The study in Argentina incorporated home-based care to patients and their families and included delivery of medications.14 In Tanzania, a combined approach was used, with CHWs measuring BP in the community, and non-physicians in the clinic accessing these readings, via smartphone, when the patient visited the clinic. Both the studies in India (SMARThealth and Control of Hypertension In Rural India (CHIRI)) were community-based with referrals to doctors based at local primary health centres. All of the studies included an education component either delivered by the NPHW of via text messages. The education included details about hypertension, its risk factors and how to manage it using lifestyle changes and adherence to medications.

Training of non-physician health workers

The training programmes ranged between 2 and 7 days in length with refresher training offered at 6 months in two studies, at yearly intervals (two studies), while in three studies, there was no defined time point for retraining. In the CHIRI study (India), ASHAs were trained to facilitate community groups and educate individuals to self-manage their BP.18 In South Africa, lay health workers were trained to assist nurses in educating patients about their condition.20 In the HOPE-4 study, CHWs in Columbia and Malaysia were trained with a curriculum developed by the study team and now adopted by the WHO Hearts Implementation Package.22 This curriculum involves CHWs using their mobile application to help provide lifestyle counselling at regular intervals following screening. The WHO PEN decision support tool for CVD management was used by trained nurses in Ghana to provide appropriate lifestyle counselling.16 The training programme was evaluated in five of the eight studies using pretraining and post-training tests and an objective practical clinical examination. In the other three studies, a framework was used to evaluate the knowledge and skills gained during the training (table 2).

NPHWs were already part of the health system in all except one study, Nkateko Trial (South Africa), where they were recruited specifically for research purposes.20 NPHWs received a salary for their role in three studies, a performance-based incentive in two studies and in three studies a combination of a salary and an incentive. All the research studies had a training programme that was developed specifically for NPHWs (table 2).

Enablers and barriers of task-shifting

The main challenges faced by the researchers included health system-related barriers, such as inability of NPHWs to prescribe evidence-based medications,23 availability and affordability of medications and high workload of the staff and NPHWs. The varying capacity, skill-sets and confidence of NPHWs also provided some challenges in providing standardised training. When this occurred, research teams had to offer on the job training and close supervision to ensure high quality and fidelity of the health intervention offered. In South Africa (Ntateko trial), where tasks were shared between nurses and CHWs, nurses were found to give CHWs additional tasks which were beyond their job description and hence difficult to perform. In some studies, there was a role conflict with other senior NPHWs who viewed the new role of newly trained junior NPHWs as a threat to their jobs. Other challenges included high staff turnover24 and vacancies in government sponsored NPHW positions. Prioritising work was also a major problem for NPHWs, many of whom had many other routine demanding activities. This barrier was further exacerbated by the lack of remuneration for NPHWs for the new tasks assigned and transition of trained nurses to new roles/locations of employment within the government system. The lack of understanding of research posed another major challenge. As examples, there were challenges to maintaining confidentiality of study participants by NPHWs, while researchers in one study had to ensure that the health promotion intervention was not delivered to the control communities. Unavailability or inconsistent supply of antihypertensive and other medications was yet another health system related barrier which several studies faced during the roll out of the intervention (see table 3 for further details).

The common enablers for task-shifting across all the studies was the enthusiasm of NPHWs to learn new skill sets and take on a new role. The community accepted their new role and reported that they had better access to healthcare as interventions were delivered in the community rather than the health centres, largely because health centres were difficult to access.25 In studies which involved the use of technology, there was an initial period of anxiety regarding the use of technology, but later NPHWs felt confident and described the tablet as a tool to help facilitate communications and engage with patients regarding lifestyle modification for CVD prevention.26 When remuneration, standardised to the country’s practices, was provided to the NPHWs for the additional tasks, this was reported to be a major enabler in implementing the intervention.

Prior to implementing the intervention, several investigators conducted audits of the health system, qualitative research with stakeholders and feasibility studies to better understand the main challenges in rolling out the main study. In five of the studies, there was some form of a pilot study, mainly to pilot and train NPHWs. The pilot studies were reported to be critical in making minor, and in some cases substantive, changes to the study and/or intervention design. In projects where there was no pilot study, investigators reported that a pilot would have avoided significant challenges in implementation and success of the programme. In addition, detailed process evaluation studies are planned in six studies, aimed at better understanding the moderators and mediators of the intervention (see online supplementary file 1 for additional details).

One of the targets of the United Nation’s Sustainable Development Goals is to reduce NCDs by 30% by 2030.25 In order to achieve this goal, there is a need to address the shortage of health human resources (including physicians, nurses and midwives). The shortage in human resources is currently estimated at 17.3 million, with the largest being in South-East Asia (6.9 million) and Africa (4.2 million).24 In these regions, the experience of the health system in using NPHWs is limited to infectious diseases and involves either short-term or curative care. In this manuscript, we have highlighted the critical lessons learnt from the field about task-shifting for management of hypertension from eight LMICs. These studies will contribute towards building the evidence required to understand the role of task-shifting for chronic disease management.

Policy and regulatory environment

Task-shifting requires engagement with policy-makers, health professionals and community members.27 At the health system level, it is essential to establish clear roles with detailed job descriptions and adequate remuneration for the tasks performed.28 In some countries NPHWs have a large workload, and adding management of NCD may compromise their productivity and the quality of their current services.29 Having structured training curriculums with evaluation, on the job training and supportive supervision would improve competencies, thereby contributing to better health outcomes in the communities.30 The current models of reimbursement are tailored for short-term and mostly curative care such as incentives given to the directly observed treatment (DOTs) provider on the successful completion of an individual’s DOTs programme or remuneration paid after complete immunisation of a child. Attractive incentive packages that will help motivate and retain NPHWs in the work force need to be devised and evaluated. These incentives could either be financial or non-financial, including enabling pathways for career progression.

During the peak of the HIV/AIDS epidemic, several countries in the African continent adopted changes in their legislation allowing ART initiation to clinical officers, medical assistants or nurses even as early as 2004, before the WHO recommended guidelines around task-shifting.31 A similar strategy is now needed for CVD management, where evidence-based decision support aided by technology can guide NPHWs to screen individuals in the community, and refer and manage individuals who do not have access to physicians. In Colombia, for example, trained nurses are allowed to prescribe from a list of essential medicines, an approach which facilitated the implementation of the HOPE-4 intervention. In Ghana, while there is no clear policy that allows nurses to prescribe medicines for BP management, nurses in the Task Shifting and Blood Pressure Control trial (TAASH) were trained to follow the WHO CVD package. In the TAASH study, the ability of the nurse to prescribe BP lowering medicines was a crucial enabler. Evaluation of these programmes, including compliance to study processes, will provide important details about the safety and efficacy of this approach for managing hypertension.

While task-shifting was the main aspect of managing hypertension in all studies, the effectiveness of the model does depend on the health work force and is reliant on a number of health system-related factors which are beyond the control of a research study and dependent on country-specific regulations on task-shifting. In order to address the global work force crises, the WHO recommends task-shifting as a viable option, but without country-specific policies that enable task-shifting and adequate physician oversight, it will be challenging to translate research into practice.

Recommendations for future research involving task-shifting

There are many lessons learnt from the investigations undertaken with the GACD projects (online supplementary file 1), including identification of the steps researchers should take while designing a study that involves task-shifting. Since task-shifting is just one component of the larger health system, it is important to audit the health system to better understand the external factors that may impact the successful implementation of the intervention. For example, if one of the aims of the task-shifting interventions is to improve adherence to medicines, the audit will help researchers understand whether medicines or a pharmacist is available at the health centre. Supporting this approach, surveillance of 596 communities in 18 countries in the Prospective Urban Rural Epidemiology (PURE) study provided evidence that medicines for preventing cardiovascular disease were both unavailable and unaffordable in most LMICs surveyed.32

Another lesson learnt was to engage with key stakeholders early and often, and well before implementing the intervention to better understand the perceptions of the community and policy-makers, for example, can reduced costs and improved access to culturally appropriate care overcome potential concerns over safety? This is essential because these perspectives influence the implementation and long-term sustainability of these programmes into real world settings.23 33 Stakeholders can also help by providing valuable feedback on the design of interventions from the outset, reducing the number of changes that need to be made during implementation.

Pilot studies were found to be an essential step prior to large-scale implementation. This helped researchers understand some of the challenges and resolve them before rolling out the intervention.26 33 Since task-shifting for managing hypertension is a new concept for NPHWs in LMICs, all the researchers involved had to ensure that NPHWs had gained appropriate knowledge and skills to provide quality care to the communities. This was often monitored by a research supervisor and sometimes supported by technology such as clinical decision support tools and blue tooth enabled BP monitors.