This report describes the development of PALM PLUS, a single set of Malawi-adapted, integrated symptom- and sign-based primary care guidelines for adults, and an innovative training program for HCWs. PALM PLUS is not designed to replace national disease-specific guidelines, but rather to assist nurses and clinicians at health centers to integrate and apply existing guidelines and protocols more effectively. While we were successful in developing an integrated guideline tool to assist front-line HCWs, we are not yet able to say whether PALM PLUS is improving health outcomes. Clinical outcomes are being evaluated in a cluster-randomized-controlled trial (cRCT) of PALM PLUS guidelines in 30 rural health centers in a single district in Malawi, with approximately 200 HCWs in each arm of the trial.
Health system capacity and a shortage of trained workers have emerged as serious obstacles to achieving universal ART coverage [[7–11]]. The goal of ensuring equitable access to quality healthcare is further frustrated by the difficulty of retaining staff in rural areas of low- and middle-income countries (LMICs) . Overcoming these obstacles requires innovative strategies to optimize the use of existing staff, and interventions to train and retain staff . However, few studies have compared different interventions , and there is little evidence that is of direct use to the policymakers crafting health systems interventions .
Current models of HCW resource needs often look at HIV/AIDS care in isolation [7, 22], without considering the need to provide other care, such as for co-morbidities or non-HIV primary care. The risks of a vertical approach to health services are known , however the push for rapid scale-up and decentralization of HIV/AIDS services, the lack of integration with primary care, and the potential for additional disease-specific vertical programs  makes integration at the primary care level even more pressing. Some integration of health training has occurred [[25–27]], yet integration of clinical services is often ineffective, incomplete or non-existent, especially with respect to HIV/AIDS care and women's reproductive health [[28–31]]. A recent study from Malawi found that 81% of HIV positive mothers enrolled in a Prevention of Mother to Child Transmission (PMTCT) program were lost to follow-up by the six-month post-natal visit , suggesting an urgent need for better integration of pre and post-natal maternal health services. A review of 25 countries with the highest HIV prevalence rates found that nearly all reported low national programme performance in controlling HIV-related TB, and called for closer integration of TB and HIV programmes . Similar arguments have been made for malaria and HIV . Yet few interventions designed to achieve integration of clinical care for front-line HCWs have been carefully evaluated.
Understanding the impact of integrated interventions on HCWs is therefore required. Several studies from Malawi suggest the importance of training and supervision to health provider retention [[34–36]]. Focus group discussions with HCWs in rural health centers identified opportunities for training and career progression, and weaknesses in clinical and district-level supervision as key factors affecting job retention . Poor supervisory support and inadequate training resources for their clinical environment correlated with the likelihood of leaving the job and/or plans to leave the job within the next 12 months . A comprehensive literature  review of health staff recruitment and retention in LMICs found that training and continuing education opportunities and management support affected retention in remote rural areas, especially in Africa, while better salaries were a cause of staff mobility for only one-quarter of respondents in those countries. The same review suggests that policy options to improve recruitment and retention in remote rural areas included improving training for rural practice and better clinical tools to improve working conditions .
PALM PLUS and PALSA PLUS seek to address these issues through the implementation and evaluation of a targeted intervention to optimize the clinical effectiveness of frontline healthcare workers in rural health centers in addressing HIV/AIDS, TB and priority primary care conditions. Recommendations in multiple national guidelines may be impossible to implement in small health centers due to lack of access to recommended tests or treatments, or even the guidelines themselves, at the primary health centre level. Traditional in-service training is often also disease-specific . Disease-specific guidelines and training may be appropriate at specialized clinics in larger centers, but they provide limited support to front-line nurses and clinicians in primary care health centers. Clinical integration has begun to occur in Malawi, such as for TB and HIV/AIDS, but more comprehensive adult integrated guidelines and tools to assist the nurse or clinician in the consulting room have yet to be developed. Furthermore, the PALM PLUS guideline provides for greater empowerment of HCWs at the local level.
Our methodology was adapted from the original method to develop PALSA guidelines , and was similar to the process for adapting the Integrated Management of Childhood Illness (IMCI) guidelines, described by WHO some years ago . Other approaches have been described, such as proposed by the ADAPTE group , which include explicit and systematic search for and grading of available evidence. Consistent application of such an approach, while laudable, would have been extremely resource-intensive and impracticable in our context given that we were developing an integrated guideline covering a large number of conditions. Secondly, a large component of the ADAPTE methodology includes deciding which guidelines to draw upon. When adapting a guideline for a public health setting, this process is replaced by what we did -- source all relevant national and local guidelines, review for consistency, identify 'red-flags' (areas where we may be concerned about the evidence-basis for the recommendations and recommendations that do not account for local resource constraints) -- and work with Ministry partners to find solutions that are consistent with evidence, but can also be feasibly implemented. These differences speak to the underlying intention of our process versus conventional guideline development processes, where the motivation is to review how a condition is diagnosed and treated. Our motivation is rather to bring existing national guidelines together into a cohesive simplified easy-to-use tool that renders them implementable by variably skilled health workers working in constrained services. Our process, while less rigorous in terms of rating guidelines and recommendations in terms of the evidence they draw on, places more emphasis on ensuring the adaptation is compliant with country policies.
Our success at developing the PALM PLUS guideline and training program for Malawi's specific context suggests that it is possible to adapt it for use in other resource-poor settings. Qualitative evaluation of HCW perceptions of the PALM PLUS guideline and training, being carried out as part of cRCT implementation trial, will provide important data and lessons from this experience and may provide guidance for future adaptations. To date, these lessons include having at least one partner organization which takes a primary responsibility for leading the process (a non-governmental organization in the case of PALM PLUS though this could also be a governmental body), support from the original team, having staff dedicated to guideline development, involvement of ministry and key opinion leaders from early on in the process, and working to resolve any conflicts with other program priorities. The on-site intermittent training utilized in PALM PLUS reduced the cost and complexity of the training program by limiting the need for transportation and allowed for training to be scheduled when convenient for both trainers and trainees. Staff did not need not leave their patients, colleagues, or families behind for days or weeks in order to undergo training, unlike in off-site training programs. However, some HCWs may perceive this as a disadvantage, since off-site training can be seen as a break from the daily grind of care delivery in remote health centers, and the per diems that usually accompany off-site training may be of substantial value to staff. This question is being formally assessed in a qualitative evaluation being carried out as part of the PALM PLUS implementation. We did not include patients in the development of the guidelines, because to do so in Malawi presented tremendous challenges including language, limited healthcare fluency among representative patients, and cultural hierarchical barriers limiting opportunities for patients to challenge the views of HCWs. However, we are evaluating the effect of the PALM PLUS guidelines in healthcare-worker/patient clinical interactions through direct observation as part of our cluster randomized trial.
There were costs associated with the development of PALM PLUS, and there would be costs with their adaptation for other countries, but such costs may need to be seen as an integral part of a commitment to expand access to ART, such has been done by Malawi. A formal costing of the development of the PALM PLUS guidelines is part of our evaluation. While health system resources and structures in South Africa and Malawi are substantially different, the broadly similar disease burden combined with their geographic proximity may allow for easier adaptation. This may result in easier adaptation within the African continent than to other developing countries elsewhere.