Edinburgh Research Explorer Setting research priorities for global respiratory medicine within the National Institute for Health Research (NIHR) Global Health Research Unit in Respiratory Health (RESPIRE)

RESPIRE’s undertake applied respiratory health in both communicable and non-communicable disorders, which the potential to improve the respiratory health of children and adults in Bangladesh, India, Malaysia and Pakistan [1]. RESPIRE’s working ethos is to work bottom-up to undertake research priorities that have been identified by RESPIRE inves tigators in our partner low- and middle-income countries (LMICs) [1]. To this end, we conducted an internal process of identifying research priorities within the RESPIRE collaboration using the Child Health The highest-ranked research ideas included proposals to measure the burden of chronic obstructive pulmonary disease (COPD) in the four RESPIRE partner countries (ie, Bangladesh, India, Malaysia and Pakistan), along with the development of a protocol for a COPD prevalence study; also, HE-AL-ASTHMA studies, a cross-sectional study to determine the prevalence of limited health literacy followed by a quasi-experimental feasibility study of a pictorial action plan for asthma in Malaysia; then, implementation of pulse oximetry as a point-of-care diagnostic in health facilities that provide integrated management of childhood illness; assessing the feasibility of using a well-established tele-consultation facility (Micro-Health Centre – MHC) in management of COPD and asthma in a resource constrained remote rural area; and studying the perception of under-five pneumonia among caregivers of selected rural and semi urban communities of Pakistan through a mixed methodolo-gy-based study. The outcomes of this process were then used, in conjunction with other approach-es, to develop the list of the research projects led by RESPIRE investigators that would receive financial support from the RESPIRE budget.

T he NIHR Global Health Research Unit in Respiratory Health (henceforth 'RESPIRE') is a research and academic capacity development initiative funded by the UK Government through its National Institute of Health Research (NIHR). RESPIRE' s focus is to undertake applied respiratory health in both communicable and non-communicable disorders, which has the potential to improve the respiratory health of children and adults in Bangladesh, India, Malaysia and Pakistan [1]. RESPIRE' s working ethos is to work bottom-up to undertake research priorities that have been identified by RESPIRE investigators in our partner low-and middle-income countries (LMICs) [1]. To this end, we conducted an internal process of identifying research priorities within the RESPIRE collaboration using the Child Health and Nutrition Research Initiative's (CHNRI) method [2]. The outcomes of this process were then used, in conjunction with other approaches, to develop the list of the research projects led by RESPIRE investigators that would receive financial support from the RESPIRE budget [2].

THE CHNRI EXERCISE TO IDENTIFY RESEARCH PRIORITIES WITHIN RESPIRE
We first asked investigators from the RESPIRE collaboration to submit their ideas for research projects and PhD projects, expecting that those ideas would mainly be generated within Bangladesh, India, Malaysia and Pakistan, in line with their local needs. We then conducted an exercise in assessment of all submitted ideas using the CHNRI method, described briefly in the Box 1 [3][4][5][6][7][8][9][10][11]. A total of 35 investigators who were named on the RESPIRE research grant proposal submitted 37 research ideas. We then prepared those ideas for scoring and invited both the 35 proposers and a further 30 members from their academic teams to take part in the process of scoring ideas. The scoring was based on 10 criteria, which were agreed in advance by the RESPIRE team during their first meeting in Edinburgh in June 2017. These criteria were: 1. Answerability: is this research question likely to be answered by this research using the proposed methods and approaches?
2. Feasibility: is this research question likely to lead to deliverable outcomes over the time scale of this project?
3. Effectiveness: is this research question likely to lead to interventions that will effectively reduce disease burden, change provision of care, change policy or practice?
The NIHR Global Health Research Unit in Respiratory Health ('RESPIRE') is a research and academic capacity development initiative funded by the UK Government through its National Institute of Health Research (NIHR). We conducted an internal process of identifying research priorities within the RESPIRE collaboration using the Child Health and Nutrition Research Initiative's (CHNRI) method. A total of 35 RESPIRE investigators submitted 37 research ideas. The scoring of those ideas was then performed by 26 (out of 65 invited) RESPIRE researchers based on 10 pre-agreed criteria: answerability, feasibility, effectiveness, applicability, affordability, potential for cross-country scalability, burden size, equity, safety and sustainability.

Box 1. The CHNRI method for setting research priorities
The CHNRI method uses the principle of crowdsourcing to score proposed research ideas against a pre-defined set of criteria. This enables funders and policymakers to view the strengths, the weaknesses and relative ranking of each proposed research idea, based on submitted opinions of a larger number of experts. This method uses a systematic, transparent, and democratic approach to priority setting. While it allows researchers to independently generate and score research questions (RQs), it also involves funders, policymakers, and other stakeholders at an early stage of the process, ensuring their ownership in the outcomes. The CHNRI method has thus far been implemented in about 100 studies led by multilateral organisations (eg, World Health Organization, United Nations International Children's Emergency Fund (UNICEF)), national governments (eg, China, India, Iran, South Africa), and funders (eg, The Bill and Melinda Gates Foundation) to set research priorities in areas ranging from the reduction of global child mortality, dementia, or disability to the efficient execution of national health plans. The recognised advantages of this method include its systematic nature, transparency and replicability, clearly defined context and criteria, involvement of the funders, stakeholders and policy makers, a structured way of obtaining information, informative and intuitive quantitative outputs, studying the level of agreement over each proposed research idea, and independent scoring of many experts, thus limiting the influence of individuals on the rest of the group [2][3][4][5][6][7][8][9][10][11].
The highest-ranked research ideas included proposals to measure the burden of chronic obstructive pulmonary disease (COPD) in the four RESPIRE partner countries (ie, Bangladesh, India, Malaysia and Pakistan), along with the development of a protocol for a COPD prevalence study; also, HE-AL-ASTHMA studies, a cross-sectional study to determine the prevalence of limited health literacy followed by a quasi-experimental feasibility study of a pictorial action plan for asthma in Malaysia; then, implementation of pulse oximetry as a point-of-care diagnostic in health facilities that provide integrated management of childhood illness; assessing the feasibility of using a well-established tele-consultation facility (Micro-Health Centre -MHC) in management of COPD and asthma in a resource constrained remote rural area; and studying the perception of under-five pneumonia among caregivers of selected rural and semi urban communities of Pakistan through a mixed methodology-based study. The outcomes of this process were then used, in conjunction with other approaches, to develop the list of the research projects led by RESPIRE investigators that would receive financial support from the RESPIRE budget. 10. Sustainability: would this research question create data or resources that will lead to opportunities for further, sustainable funding?
The contextual background to guide the scoring was defined in terms of space, time, population of interest and disease burden of interest, as is standard practice in the CHNRI process [3,4,11]. Space was defined as the four partner countries (ie, Bangladesh, India, Malaysia and Pakistan), the time as the interval between now and year 2025, the population of interest was defined as respiratory disease sufferers, and the disease burden of interest was defined as "all respiratory diseases" within the defined space, time and population.
The scorers were also instructed to think beyond the endpoints of research questions and to keep in mind their broader scope and relevance. Assessing some of the proposed research questions was only possible if thinking was shifted from purely assessing the likelihood that the proposed research would achieve their endpoints, to what those endpoints could mean in the wider context and how these could help improve the overall current situation in terms of risk avoidance and intervention coverage at the level of the four partner countries in 2018.
A total of 26 RESPIRE researchers (among the 65 invited) returned their scores by the stated deadline and this allowed us to conduct the analysis of their input, which was performed in line with the guidelines for implementation of the CHNRI process [11]. The scorers were asked to assess each proposed research idea according to the 10 questions posed above as "yes" (coded as 1 point), "no" (0 points), "not sure" (0.5 points) or "don't know" (input left blank). The received scores allowed computation of "research priority scores" for each criterion and the overall priority score, the latter being used for the final ranking of the proposed research questions (Table 1) [3,4,11]. In addition, the "average expert agreement" (AEA) was computed for each proposed research question to demonstrate the level of controversy related to each proposed research question among the scorers who took part in the CHNRI exercise [3,4,11].     Table 1 provides the final rankings for the 37 proposed research questions that relate to global respiratory health and are of interest to the RESPIRE initiative. Research priority scores for these ideas are shown for each of the 10 priority-setting criteria and they are self-explanatory, as well as the overall research priority score, which is the basis for the final ranking.

OUTCOMES
Research ideas with the highest scores included proposals to measure the burden of COPD in the four partner countries, along with the development of a protocol for a COPD prevalence study; also, HEAL-ASTHMA studies, a cross-sectional study to determine the prevalence of limited health literacy followed by a quasi-experimental feasibility study of a pictorial action plan for asthma in Malaysia; then, implementation of pulse oximetry as a point-of-care diagnostic in health facilities that provide integrated management of childhood illness; assessing the feasibility of using a well-established tele-consultation facility (Micro-Health Centre -MHC) in management of COPD and Asthma in a resource constrained remote rural area; and studying the perception of under-five pneumonia among caregivers of selected rural and semi urban communities of Pakistan through a mixed-methods study.
Most importantly, all projects received an overall research priority score greater than 70.0, while a typical range of scores in CHNRI exercises is between 27.0-91.0 [4]. This means that the 26 RESPIRE investigators felt that all 35 proposed research ideas were promising, had few apparent shortcomings and were worth doing. However, there were rather strong internal differences in opinions related to 8 out of the 37 proposed projects, as captured by AEA<50.0 (Table 1), which meant that some of the proposed ideas were probably more controversial than others.
There are also some specific concerns related to some proposed research questions in relation to particular criteria. The criteria around which most concerns were raised, as captured by the criterion-specific research score <70.0, were sustainability (in 9 cases), scalability (in 7 cases) and effectiveness (in 3 cases; see Table 1).