Specific considerations for research on the effectiveness of multisectoral collaboration: methods and lessons from 12 country case studies.

BACKGROUND: The success of the Sustainable Development Goals (SDGs) is predicated on multisectoral collaboration (MSC), and the COVID-19 pandemic makes it more urgent to learn how this can be done better. Complex challenges facing countries, such as COVID-19, cut across health, education, environment, financial and other sectors. Addressing these challenges requires the range of responsible sectors and intersecting services - across health, education, social and financial protection, economic development, law enforcement, among others - transform the way they work together towards shared goals. While the necessity of MSC is recognized, research is needed to understand how sectors collaborate, inform how to do so more efficiently, effectively and equitably, and ascertain similarities and differences across contexts. To answer these questions and inform practice, research to strengthen the evidence-base on MSC is critical. METHODS: This paper draws on a 12-country study series on MSC for health and sustainable development, in the context of the health and rights of women, children and adolescents. It is written by core members of the research coordination and country teams. Issues were analyzed during the study period through 'real-time' discussions and structured reporting, as well as through literature reviews and retrospective feedback and analysis at the end of the study. RESULTS: We identify four considerations that are unique to MSC research which will be of interest to other researchers, in the context of COVID-19 and beyond: 1) use theoretical frameworks to frame research questions as relevant to all sectors and to facilitate theoretical generalizability and evolution; 2) specifically incorporate sectoral analysis into MSC research methods; 3) develop a core set of research questions, using mixed methods and contextual adaptations as needed, with agreement on criteria for research rigor; and 4) identify shared indicators of success and failure across sectors to assess MSCs. CONCLUSION: In responding to COVID-19 it is evident that effective MSC is an urgent priority. It enables partners from diverse sectors to effectively convene to do more together than alone. Our findings have practical relevance for achieving this objective and contribute to the growing literature on partnerships and collaboration. We must seize the opportunity here to identify remaining knowledge gaps on how diverse sectors can work together efficiently and effectively in different settings to accelerate progress towards achieving shared goals.

Rural health research capacity building: an anchored solution.

Rural physicians face many challenges with providing rural health care, which often leads to innovative solutions. Despite their creativity with overcoming barriers, there is a lack of support for rural health research - an area of health care where research makes great impacts on small communities. Rural research capacity building (RRCB) is essential to support rural physicians so that they can conduct relevant research, but RRCB programs are sparse. Thus, our team at Memorial University of Newfoundland, Canada, has created an RRCB ecosystem through the 6for6 and Rural360 programs, which outline a pathway for rural physicians to make meaningful contributions to their communities through research. This article describes the RRCB ecosystem and explains how the 6for6 and Rural360 programs address the need for RRCB. Designed to train six rural physicians over six sessions per year, 6for6 fosters learning of research practices through a conceptual framework that envelops complexity science, systems thinking, and anchored instruction. The use of this framework allows the learning to be grounded in issues that are locally relevant for each participant and follows guiding principles that enable many types of learning. Rural360 continues the pathway by providing an in-house funding opportunity with an iterative review process that allows participants to continue developing their research skills and, ultimately, secure funding for their project. This anchored delivery model of RRCB programming is made possible through many support systems including staff, librarians, instructors, the university, and other stakeholders. It has successfully helped form communities of practice, promotes collaboration both between learners and with third parties, encourages self-organization with flexibility for learners outside of the in-house sessions, and ultimately drives social accountability in addressing local healthcare issues.

The Water@Birth Study: an exploratory study on the requirements of water for hand hygiene during labour and delivery in low-income countries.

BACKGROUND: Infections acquired during childbirth are one of the leading causes of maternal death; the majority of these deaths occur in low-income settings. Hand hygiene is one of the most effective ways of preventing infection but requires basic resources, such as running water, to be performed. Limited literature on water volume requirements for hand hygiene in healthcare facilities exists despite the importance of this information, particularly in resource-poor settings. AIM: To establish the volume of water required for hand hygiene during childbirth in low-income countries. METHODS: Data was collected in Aberdeen Maternity Hospital (AMH) and Felege Hiwot Referral Hospital, Ethiopia (FHRH), with an average of 14 and 16 deliveries per day respectively. Primary data on hand hygiene opportunities (HHOs) during childbirth were gathered using observational methods, and secondary data gathered from register and case-note reviews. The volume of water required for each HHO (H2O/HHO) was calculated by multiplying flow rate by hand washing time. Estimates of water requirements were derived by calculating the number of HHOs during childbirth and the H2O/HHO. Water requirement estimates from each facility were compared to each other as well as to WHO recommendations. Due to skewed data, Spearman's rho was utilised to explore the relationship between variables. RESULTS: Eleven deliveries were observed in AMH and 20 in FHRH. The number of HHOs was largely determined by the length of labour. Stringently following WHO recommendations lead to a significantly higher number of HHOs than was performed in clinical practice at both sites. Hand washing also occurred for a much shorter time than the WHO recommendation of 40-60 seconds, with an average of 24 seconds in AMH and 25 seconds in FHRH observed. The estimated number of HHOs at sites ranged from 5 to 16 per hour per delivery and water consumption from 21 to 159.6 litres per hour per delivery. Hand hygiene was estimated to require 8937.6 litres and 4838.4 litres per day or 638.4 litres and 302.4 litres per delivery for AMH and FHRH, respectively. CONCLUSIONS: Water requirements are variable due to the nature of childbirth but are not currently met in low-income countries. In terms of performance of hand hygiene, there is a large gap between clinical and recommended practice and thus room for improvement. The volume of water required for hand hygiene has significant implications for water requirements within maternity units, particularly in resource-poor settings. Further research on water requirements is merited to improve the targeting of limited resources.

From pipelines to pathways: the Memorial experience in educating doctors for rural generalist practice.

CONTEXT: This report describes the community context, concept and mission of The Faculty of Medicine at Memorial University of Newfoundland (Memorial), Canada, and its 'pathways to rural practice' approach, which includes influences at the pre-medical school, medical school experience, postgraduate residency training, and physician practice levels. Memorial's pathways to practice helped Memorial to fulfill its social accountability mandate to populate the province with highly skilled rural generalist practitioners. Programs/interventions/initiatives: The 'pathways to rural practice' include initiatives in four stages: (1) before admission to medical school; (2) during undergraduate medical training (medical degree (MD) program); (3) during postgraduate vocational residency training; and (4) after postgraduate vocational residency training. Memorial's Learners & Locations (L&L) database tracks students through these stages. The Aboriginal initiative - the MedQuest program and the admissions process that considers geographic or minority representation in terms of those selecting candidates and the candidates themselves - occurs before the student is admitted. Once a student starts Memorial's MD program, the student has ample opportunities to have rural-based experiences through pre-clerkship and clerkship, of which some take place exclusively outside of St. John's tertiary hospitals. Memorial's postgraduate (PG) Family Medicine (FM) residency (vocational) training program allows for deeper community integration and longer periods of training within the same community, which increases the likelihood of a physician choosing rural family medicine. After postgraduate training, rural physicians were given many opportunities for professional development as well as faculty development opportunities. Each of the programs and initiatives were assessed through geospatial rurality analysis of administrative data collected upon entry into and during the MD program and PG training (L&L). Among Memorial MD-graduating classes of 2011-2020, 56% spent the majority of their lives before their 18th birthday in a rural location and 44% in an urban location. As of September 2016, 23 Memorial MD students self-identified as Aboriginal, of which 2 (9%) were from an urban location and 20 (91%) were from rural locations. For Year 3 Family Medicine, graduating classes 2011 to 2019, 89% of placement weeks took place in rural communities and 8% took place in rural towns. For Memorial MD graduating classes 2011-2013 who completed Memorial Family Medicine vocational training residencies, (N=49), 100% completed some rural training. For these 49 residents (vocational trainees), the average amount of time spent in rural areas was 52 weeks out of a total average FM training time of 95 weeks. For Family Medicine residencies from July 2011 to October 2016, 29% of all placement weeks took place in rural communities and 21% of all placement weeks took place in rural towns. For 2016-2017 first-year residents, 53% of the first year training is completed in rural locations, reflecting an even greater rural experiential learning focus. LESSONS LEARNED: Memorial's pathways approach has allowed for the comprehensive training of rural generalists for Newfoundland and Labrador and the rest of Canada and may be applicable to other settings. More challenges remain, requiring ongoing collaboration with governments, medical associations, health authorities, communities, and their physicians to help achieve reliable and feasible healthcare delivery for those living in rural and remote areas.

Does rural generalist focused medical school and family medicine training make a difference? Memorial University of Newfoundland outcomes.

INTRODUCTION: Rural recruitment and retention of physicians is a global issue. The Faculty of Medicine at Memorial University of Newfoundland, Canada, was established as a rural-focused medical school with a social accountability mandate that aimed to meet the healthcare needs of a sparse population distributed over a large landmass as well as the needs of other rural and remote areas of Canada. This study aimed to assess whether Memorial medical degree (MD) and postgraduate (PG) programs were effective at producing physicians for their province and rural physicians for Canada compared with other Canadian medical schools. METHODS: This retrospective cohort study included medical school graduates who completed their PG training between 2004 and 2013 in Canada. Practice locations of study subjects were georeferenced and assigned to three geographic classes: Large Urban; Small City/Town; and Rural. Analyses were performed at two levels. (1) Provincial level analysis compared Memorial PG graduates practicing where they received their MD and/or PG training with other medical schools who are the only medical school in their province (n=4). (2) National-level analysis compared Memorial PG graduates practicing in rural Canada with all other Canadian medical schools (n=16). Descriptive and bivariate analyses were performed. RESULTS: Overall, 18 766 physicians practicing in Canada completed Canadian PG training (2004-2013), and of those, 8091 (43%) completed Family Medicine (FM) training. Of all physicians completing Canadian PG training, 1254 (7%) physicians were practicing rurally and of those, 1076 were family physicians. There were 379 Memorial PG graduates and of those, 208 (55%) completed FM training and 72 (19%) were practicing rurally, and of those practicing rurally, 56 were family physicians. At the national level, the percentage of all Memorial PG graduates (19.0%) and FM PG graduates (26.9%) practicing rurally was significantly better than the national average for PG (6.4%, p<0.000) and FM (12.9%, p<0.000). Among 391 physicians practicing in Newfoundland and Labrador (NL), 257 (65.7%) were Memorial PG graduates and 247 (63.2%) were Memorial MD graduates. Of the 163 FM graduates, 148 (90.8%) were Memorial FM graduates and 118 (72.4%) were Memorial MD graduates. Of the 68 in rural practice, 51 (75.0%) were Memorial PG graduates and 31 (45.6%) were Memorial MD graduates. Of the 41 FM graduates in rural practice, 39 (95.1%) were Memorial FM graduates and 22 (53.7%) were Memorial MD graduates. Two-sample proportion tests demonstrated Memorial University provided a larger proportion of its provincial physician resource supply than the other four single provincial medical schools, by medical school MD for FM (72.4% vs 44.3%, p<0.000) and for overall (63.2% vs 43.5% p<0.000), and by medical school PG for FM (90.8 % vs 72.0%, p<0.000). CONCLUSION: This study found Memorial University graduates were more likely to establish practice in rural areas compared with the national average for most program types as well as more likely to establish practice in NL compared with other single medical schools' graduates in their provinces. This study highlights the impact a comprehensive rural-focused social accountability approach can have at supplying the needs of a population both at the regional and rural national levels.

Diversity and divergence: the dynamic burden of poor maternal health.

Maternal health is a big issue and is central to sustainable development. Each year, about 210 million women become pregnant and about 140 million newborn babies are delivered-the sheer scale of maternal health alone makes maternal well being and survival vital concerns. In this Series paper, we adopt primarily a numerical lens to illuminate patterns and trends in outcomes, but recognise that understanding of poor maternal health also warrants other perspectives, such as human rights. Our use of the best available evidence highlights the dynamic burden of maternal health problems. Increased diversity in the magnitude and causes of maternal mortality and morbidity between and within populations presents a major challenge to policies and programmes aiming to match varying needs with diverse types of care across different settings. This diversity, in turn, contributes to a widening gap or differences in levels of maternal mortality, seen most acutely in vulnerable populations, predominantly in sub-Saharan Africa. Strong political and technical commitment to improve equity-sensitive information systems is required to monitor the gap in maternal mortality, and robust research is needed to elucidate major interactions between the broad range of health problems. Diversity and divergence are defining characteristics of poor maternal health in the 21st century. Progress on this issue will be an ultimate judge of sustainable development.

Quality maternity care for every woman, everywhere: a call to action.

To improve maternal health requires action to ensure quality maternal health care for all women and girls, and to guarantee access to care for those outside the system. In this paper, we highlight some of the most pressing issues in maternal health and ask: what steps can be taken in the next 5 years to catalyse action toward achieving the Sustainable Development Goal target of less than 70 maternal deaths per 100 000 livebirths by 2030, with no single country exceeding 140? What steps can be taken to ensure that high-quality maternal health care is prioritised for every woman and girl everywhere? We call on all stakeholders to work together in securing a healthy, prosperous future for all women. National and local governments must be supported by development partners, civil society, and the private sector in leading efforts to improve maternal-perinatal health. This effort means dedicating needed policies and resources, and sustaining implementation to address the many factors influencing maternal health-care provision and use. Five priority actions emerge for all partners: prioritise quality maternal health services that respond to the local specificities of need, and meet emerging challenges; promote equity through universal coverage of quality maternal health services, including for the most vulnerable women; increase the resilience and strength of health systems by optimising the health workforce, and improve facility capability; guarantee sustainable finances for maternal-perinatal health; and accelerate progress through evidence, advocacy, and accountability.

Who delivers where? The effect of obstetric risk on facility delivery in East Africa.

OBJECTIVES: Skilled attendance at birth is key for the survival of pregnant women. This study investigates whether women at increased risk of maternal and newborn complications in four East African countries are more likely to deliver in a health facility than those at lower risk. METHODS: Demographic and Health Survey data for Kenya 2014, Rwanda 2014-15, Tanzania 2015-16 and Uganda 2011 were used to study women with a live birth in the three years preceding the survey. A three-level obstetric risk index was created using known risk factors. Generalised linear Poisson regression was used to investigate the association between obstetric risk and facility delivery. RESULTS: We analysed data from 13 119 women across the four countries of whom 42-45% were considered at medium risk and 12-17% at high risk, and the remainder were at low risk. In Rwanda, 93% of all women delivered in facilities but this was lower (59-66%) in the other three countries. There was no association between a woman's obstetric risk level and her place of delivery in any country; greater wealth and more education were, however, independently strongly associated with facility delivery. CONCLUSIONS: In four East African countries, women at higher obstetric risk were not more likely to deliver in a facility than those with lower risk. This calls for a renewed focus on antenatal risk screening and improved communication on birth planning to ensure women with an increased chance of maternal and newborn complications are supported to deliver in facilities with skilled care.

Curriculum development of 6for6: Longitudinal research skills program for rural and remote family physicians.

PROBLEM ADDRESSED: To address barriers challenging the engagement of rural and remote family physicians (RRFPs) in research, Memorial University of Newfoundland in St John's has developed a longitudinal faculty development program (FDP) called 6for6. OBJECTIVE OF PROGRAM: To establish and evaluate a longitudinal FDP that promotes a foundation of research activity. Program description Informed by a needs assessment in phase 1, phase 2 saw the 6for6 curriculum designed, developed, and implemented to reflect the unique needs of RRFPs. Preliminary evaluations have been conducted and results will be presented after year 1 of the program. CONCLUSION: The 6for6 FDP has been positively received by participants, and it is evident that they will serve as champions of rural research capacity building. It is anticipated that by April 2017, 18 RRFPs will be equipped with the research and leadership skills required to foster research networks within and outside their communities.

Needs assessment for development of 6for6: Longitudinal research skills program tailored to rural and remote family physicians.

PROBLEM ADDRESSED: Rural and remote family physicians (RRFPs) face greater barriers to research engagement than their urban colleagues and have access to fewer faculty development programs (FDPs) to foster their research skills. OBJECTIVE OF PROGRAM: To identify and prioritize skills and services that RRFPs need to engage in research. PROGRAM DESCRIPTION: Memorial University of Newfoundland in St John's used a needs assessment as the foundation for developing an FDP for RRFPs. The assessment comprised a systematic literature review and environmental scan, key informant interviews (n = 10), a focus group with RRFPs (n = 15), expert group meetings (n = 2), and needs assessment surveys (n = 19). CONCLUSION: The assessment identified barriers to RRFPs engaging in research, priority considerations for the development of a research FDP for RRFPs, and research areas to be included in the program curriculum. This information was used to inform phases 2 and 3 of program development, which are further discussed in a companion article.

Effect of vaginal self-sampling on cervical cancer screening rates: a community-based study in Newfoundland.

BACKGROUND: Cervical cancer is highly preventable and treatable if detected early through regular screening. Women in the Canadian province of Newfoundland & Labrador have relatively low rates of cervical cancer screening, with rates of around 40 % between 2007 and 2009. Persistent infection with oncogenic human papillomavirus (HPV) is a necessary cause for the development of cervical cancer, and HPV testing, including self-sampling, has been suggested as an alternative method of cervical cancer screening that may alleviate some barriers to screening. Our objective was to determine whether offering self-collected HPV testing screening increased cervical cancer screening rates in rural communities. METHODS: During the 2-year study, three community-based cohorts were assigned to receive either i) a cervical cancer education campaign with the option of HPV testing; ii) an educational campaign alone; iii) or no intervention. Self-collection kits were offered to eligible women at family medicine clinics and community centres, and participants were surveyed to determine their acceptance of the HPV self-collection kit. Paired proportions testing for before-after studies was used to determine differences in screening rates from baseline, and Chi Square analysis of three dimensional 2 × 2 × 2 tables compared the change between communities. RESULTS: Cervical cancer screening increased by 15.2 % (p < 0.001) to 67.4 % in the community where self-collection was available, versus a 2.9 % increase (p = 0.07) in the community that received educational campaigns and 8.5 % in the community with no intervention (p = 0.193). The difference in change in rates was statistically significant between communities A and B (p < 0.001) but not between communities A and C (p = 0.193). The response rate was low, with only 9.5 % (168/1760) of eligible women opting to self-collect for HPV testing. Of the women who completed self-collection, 15.5 % (26) had not had a Pap smear in the last 3 years, and 88.7 % reported that they were somewhat or very satisfied with self-collection. CONCLUSIONS: Offering self-collected HPV testing increased the cervical cancer screening rate in a rural NL community. Women who completed self-collection had generally positive feelings about the experience. Offering HPV self-collection may increase screening compliance, particularly among women who do not present for routine Pap smears.

From joint thinking to joint action: a call to action on improving water, sanitation, and hygiene for maternal and newborn health.

Yael Velleman and colleagues argue for stronger integration between the water, sanitation, and hygiene (WASH) and maternal and newborn health sectors. Please see later in the article for the Editors' Summary.

Assessing the impact of a cleaning programme on environmental hygiene in labour and neonatal wards: an exploratory study in The Gambia.

BACKGROUND: Effective surface cleaning in hospitals is crucial to prevent the transmission of pathogens. However, hospitals in low- and middle-income countries face cleaning challenges due to limited resources and inadequate training. METHODS: We assessed the effectiveness of a modified TEACH CLEAN programme for trainers in reducing surface microbiological contamination in the newborn unit of a tertiary referral hospital in The Gambia. We utilised a quasi-experimental design and compared data against those from the labour ward. Direct observations of cleaning practices and key informant interviews were also conducted to clarify the programme's impact. RESULTS: Between July and September 2021 (pre-intervention) and October and December 2021 (post-intervention), weekly surface sampling was performed in the newborn unit and labour ward. The training package was delivered in October 2021, after which their surface microbiological contamination deteriorated in both clinical settings. While some cleaning standards improved, critical aspects such as using fresh cleaning cloths and the one-swipe method did not. Interviews with senior departmental and hospital management staff revealed ongoing challenges in the health system that hindered the ability to improve cleaning practices, including COVID-19, understaffing, disruptions to water supply and shortages of cleaning materials. CONCLUSIONS: Keeping a hospital clean is fundamental to good care, but training hospital cleaning staff in this low-income country neonatal unit failed to reduce surface contamination levels. Further qualitative investigation revealed multiple external factors that challenged any possible impact of the cleaning programme. Further work is needed to address barriers to hospital cleaning in low-income hospitals.

Towards elimination of maternal deaths: maternal deaths surveillance and response.

Current methods for estimating maternal mortality lack precision, and are not suitable for monitoring progress in the short run. In addition, national maternal mortality ratios (MMRs) alone do not provide useful information on where the greatest burden of mortality is located, who is concerned, what are the causes, and more importantly what sub-national variations occur. This paper discusses a maternal death surveillance and response (MDSR) system. MDSR systems are not yet established in most countries and have potential added value for policy making and accountability and can build on existing efforts to conduct maternal death reviews, verbal autopsies and confidential enquiries. Accountability at national and sub-national levels cannot rely on global, regional and national retrospective estimates periodically generated from academia or United Nations organizations but on routine counting, investigation, sub national data analysis, long term investments in vital registration and national health information systems. Establishing effective maternal death surveillance and response will help achieve MDG 5, improve quality of maternity care and eliminate maternal mortality (MMR ≤ 30 per 100,000 by 2030).

Correction: Postnatal infection surveillance by telephone in Dar es Salaam, Tanzania: An observational cohort study.

[This corrects the article DOI: 10.1371/journal.pone.0254131.].

Clean birth kits to improve birth practices: development and testing of a country level decision support tool.

BACKGROUND: Clean birth practices can prevent sepsis, one of the leading causes of both maternal and newborn mortality. Evidence suggests that clean birth kits (CBKs), as part of package that includes education, are associated with a reduction in newborn mortality, omphalitis, and puerperal sepsis. However, questions remain about how best to approach the introduction of CBKs in country. We set out to develop a practical decision support tool for programme managers of public health systems who are considering the potential role of CBKs in their strategy for care at birth. METHODS: Development and testing of the decision support tool was a three-stage process involving an international expert group and country level testing. Stage 1, the development of the tool was undertaken by the Birth Kit Working Group and involved a review of the evidence, a consensus meeting, drafting of the proposed tool and expert review. In Stage 2 the tool was tested with users through interviews (9) and a focus group, with federal and provincial level decision makers in Pakistan. In Stage 3 the findings from the country level testing were reviewed by the expert group. RESULTS: The decision support tool comprised three separate algorithms to guide the policy maker or programme manager through the specific steps required in making the country level decision about whether to use CBKs. The algorithms were supported by a series of questions (that could be administered by interview, focus group or questionnaire) to help the decision maker identify the information needed. The country level testing revealed that the decision support tool was easy to follow and helpful in making decisions about the potential role of CBKs. Minor modifications were made and the final algorithms are presented. CONCLUSION: Testing of the tool with users in Pakistan suggests that the tool facilitates discussion and aids decision making. However, testing in other countries is needed to determine whether these results can be replicated and to identify how the tool can be adapted to meet country specific needs.

(The shift to) online delivery of a rural faculty development program in research skills: lessons learned.

BACKGROUND: While rural physicians are the ideal candidates to investigate health and healthcare issues in rural communities, they often lack the required skills, competencies, and resources. As a result, research skills development programs are crucial to help ensure communities receive the quality of care they deserve. Memorial University of Newfoundland created a research skills development program called 6for6 to empower and enable rural physicians to research solutions to community-specific health needs. 6for6 program delivery was exclusively in-person until 2019. However, with limitations introduced due to the COVID-19 pandemic, organizations around the globe needed to respond quickly. As we work to return to a post-pandemic environment, program administrators and educators worldwide are unsure whether to retain or remove the changes made to programs to adapt to the pandemic restrictions. Therefore, this work addresses the impact of the online delivery model in two areas: 1) attainment of competencies (specifically research skills, knowledge, and attitudes); and 2) participant experiences, defined as the ease of attendance, the capacity to interact with team members and peers, and challenges or barriers associated with navigating program resources. METHODS: We compared the effect of an online delivery model pivoted to adapt pandemic restrictions with the original model (primarily face-to-face) on the acquisition of learning competencies and participant experience using a mixed-methods study. Various data collection methods, such as a pre-post program survey, post-program focus group, and structured observation, were utilized. RESULTS: From 2014 to 2021, 35 physicians attended the program (30 face-to-face and five online). The Wilcoxon-sign-rank test did not show any significant differences in the participants' median change of research competency scores who attended face-to-face and online learning, respectively: knowledge (32.6, 26.8), attitudes (3.8, 3.5), and skills (32.4, 20.0). Flexibility and accessibility were key aspects of participants' experiences during the online model. Comparison with previous years demonstrated no significant challenges with the virtual delivery model, yet participants struggled with mentorship challenges and learning-life balance. CONCLUSIONS: Although presenting some unique challenges, the online model did not negatively affect learner competencies. Likewise, it provided opportunities for rural physicians to attend learning sessions and interact with experts and peers while remaining in their communities.