“It’s Everyone’s Problem”: Institutionalising Multisectoral Action for Maternal Health in Meghalaya, India

Sampling

The study’s qualitative process tracking consisted of district, facility and community‑level interviews and meeting observations conducted in multiple rounds in a sample of facilities across the state. We jointly identified six districts to reflect geographic, demographic and health system diversity across Meghalaya: North Garo Hills, West Garo Hills, West Jaintia Hills, Ri Bhoi, East Khasi Hills and South West Khasi Hills (Figure 3). These six districts were chosen purposively to ensure geographic diversity and logistical convenience for repeated interviews. Within these, we selected a sample of 30 Primary Health Centres (PHCs; population coverage <20,000) and Community Health Centres (CHCs; population coverage <80,000) using stratified probability proportional to size based on population at the district level and institutional delivery. Over the course of the evaluation, based on inputs from the Government, we replaced seven sampled facilities that no longer fit study criteria (upgraded to a higher‑level facility or stopped conducting deliveries) or were not implementing Rescue Mission activities.

Figure 3

Data sources

Data sources included primary qualitative data, quantitative administrative data from the State’s MOTHER app and secondary information, such as guidelines and relevant reports. Qualitative data included multiple rounds of individual interviews across the different levels of governance, focus group discussions at the community‑level and observations of key Rescue Mission meetings and field visits by the Government of Meghalaya. We tracked monthly administrative data on key indicators (institutional delivery, antenatal care and maternal deaths).

Data collection

We conducted interviews with six types of participants at different levels of government over three rounds of data collection. Three investigators, one for each region who could speak respective local languages, focused on facilities. A Khasi‑speaking supervisor conducted interviews at the block and district level. They also conducted focus group discussions with other stakeholders and observed facility‑level meetings, community events and meetings and district‑level review meetings (Table 1). Over time, the team built a rapport with facility staff and block‑level functionaries which enabled repeat interviews as well updates on Rescue Mission activities over time. Regular feedback loops with the government highlighted priorities for learning, based on which we included or excluded stakeholders in subsequent rounds of data collection.

Ethical considerations

Ethical approval was granted by the Sigma Institutional Review Board (Reference No. 10068/IRB/22‑23, 5/11/2022). Trained investigators administered informed consent procedures and recorded audio only after permission from participants.

Analysis

After every round of data collection, recordings were translated and transcribed from Khasi, Garo and Pnar into English, the team’s common language, and prepared for coding and analysis. The research team held internal workshops to code and analyse data thematically. Observation notes by field investigators supplemented transcripts. We removed identifying information for respondents and used pseudonyms in this manuscript. Themes that emerged across three rounds of data collection included sense of purpose, decentralised and adaptive leadership, political supportability, ownership, health system factors, accountability mechanisms, community engagement and use of data. Themes were shared with the Government to understand priorities for subsequent rounds of data collection and to validate interim findings.

We developed an index of state capability approach adoption and categorised Medical Officers (MOs) (based on interviews) across five categories: problem‑solving approach; intersectoral collaboration; use of data for planning; community engagement and sense of purpose. We used a 3‑point scale (limited, average and exceptional) for scores between 5‑15. The intention of mapping facilities by these indices was to identify opportunities to strengthen inputs rather than to rank MOs on performance.

Maternal health outcomes

The institutional delivery rates steadily increased from 59% in September 2021 to 69% in June 2023 (Figure 4). The Garo region had an institutional delivery rate of 67%, similar to the Jaintia and Khasi regions (65%). Maternal deaths reported in the State registry reduced from 204 (from September 2021‑2022) to 121 during September 2022–2023 [17].

Figure 4

Multisectoral collaboration

Multisectoral collaboration between the Departments of Health & Family Welfare, Community & Rural Development and Social Welfare is institutionalised across governance levels and through community outreach activities. Specifically, Rescue Mission engages the Department of Social Welfare’s Integrated Child Development Scheme (ICDS), an early childhood development programme which aims to address health and nutrition needs of young children, pregnant and nursing mothers and the State Rural Livelihoods Mission (SRLM) that organises women’s Self‑Help Groups (SHGs). Officials in both departments reflected a commitment to improving maternal and child health. Amongst officials of the SRLM, there was a recognition of the clear link between livelihoods and social determinants of maternal health, while the Social Welfare department viewed their role as primarily focused on reducing anaemia and improving nutrition.

District, sector and facility‑level review meetings

Meetings across sectors were the central method of fostering collaboration. At the State‑ and district‑level, meetings first articulated responsibilities by department, reviewed progress on key health priorities, identified implementation and coordination challenges and developed tailored solutions. After a round of covering all districts, an all‑District Review Meeting was convened by State leadership, chaired by the leadership of any of the three departments. The Health Department typically steered these more often than other departments. District officials viewed these meetings as a platform for cross‑district learning as well as problem‑solving for identified challenges. We observed that meetings helped sustain a sense of purpose and motivation across departments; target‑setting and evaluation by each District/State Department promoted a sense of urgency.

At the facility level, reviews (Sector meetings) were intended to facilitate teamwork across the three departments, including facility‑level and frontline cadres updates and review of any maternal or child deaths. If any challenges were identified, the Sector Team would attempt to find solutions during the meeting. These meetings also offered a platform to raise facility‑level and department‑specific issues. In our observation, Sector Meeting action plans tended to be generic. For instance, an MO would direct ANMs to follow up on dropouts for immunization without specifying how many were to be followed up and specifically in what communities. In exceptional cases, facilities developed concrete action plans with accountability for specific items.

The regularity of meetings appeared to vary by facility, often due to scheduling conflicts between the three departments. As a result, some districts mandated specific days in the month for Sector meetings. Other districts further mandated meetings be held in villages with particularly poor health indicators such as immunisation and institutional deliveries. Many facilities combined Sector meetings with other scheduled meetings at the Block‑level. The strength of participation of all departments varied across geographies. Functionaries from non‑Health departments recognised the value of Sector meetings but found it challenging to attend them regularly. Despite variation across meetings, there appeared to be a consistent, and perceptible, increase in collaboration due to joint planning, review and monitoring during meetings.

Transit homes

Transit homes located within or near health facilities for high‑risk pregnant women were introduced to facilitate timely access to delivery care for women with poor road connectivity, managed by SHGs and local entrepreneurs (Box 1) [18]. Uptake varied across geographies. Women who had previously given birth at home found it challenging to leave their children and household responsibilities to stay away for a prolonged period. Women who used Transit Homes were largely from areas that had non‑motorable roads. In Blocks with hard‑to‑reach areas but motorable roads, families preferred to rely on vehicles than stay in transit homes.

Nearly 94 transit homes were being run by SHGs; stakeholders perceived a clear difference in transit homes run by women’s groups because they were able to better support and engage pregnant women. Health facility staff in contrast were unable to engage with them in a similar manner. Some invited traditional birth attendants – trusted by the community – to run transit homes as a potential solution.

Decentralised leadership

Most Rescue Mission activities are centred on the leadership of Medical Officers (MOs) at Primary Health Centres, who in turn collaborate with facility and block‑level teams in local problem‑solving.

Sense of purpose

A central goal of SCEP was to build a sense of purpose and motivation amongst frontline workers and providers, ultimately to transform public service delivery through decentralised, local leadership with the freedom to make decisions. Accordingly, Rescue Mission cultivated Medical Officers (MOs) as leaders of their facilities and surrounding areas, which would also free District level leaders to focus on broader monitoring and support.

Amongst interviewed MOs, nearly all appeared to draw their sense of purpose from serving their communities. They described experiencing deep satisfaction from being able to treat patients who had travelled great distances to access care with limited resources. Positive feedback from patients and the community’s trust in the MOs were highlighted as sources of encouragement.

MOs also reported sources of motivation that went beyond service provision, such as feeling invested in the overall development of their communities, improved performance in schools in their area or increased male participation for health through Village Health Councils.

On the other hand, several MOs we interviewed reported inadequate human resources as a primary source of demotivation. They reported feeling overworked, finding it difficult to give all their patients and programmes sufficient time. They did however recognise that their work required them to be available 24 hours a day, while engaging in non‑clinical work such as facility administration and monitoring. MOs also reported difficult living and working conditions, particularly poor facility infrastructure, electricity and water. While these issues were perceived to affect service provision, they recognised that these extended beyond the purview of the Health Department.

Some MOs felt demotivated to perform their work when they observed low strength of participation from non‑health departments, overall slow government processes and inadequate compensation. They highlighted requiring support and capacity strengthening to handle some of the managerial, financial, and administrative aspects of their jobs for which they felt unprepared.

Problem‑solving

MOs appeared to take independent decisions on task delegation, day‑to‑day functioning of the facility as well as minor infrastructure‑related issues. For larger purchases and actions, MOs typically deferred to the Block‑ or District‑level decision‑makers. Most MOs perceived not having the required finances or human resources to implement actions for tackling bigger issues. However, they did appreciate the flexibility in spending that the Chief Minister’s Safe Motherhood Scheme allowed; MOs devised innovative, context‑specific solutions to transportation through this flexibility (Box 2). Facility visits by State officials appeared to serve as a direct line of communication between MOs and the State. Visits by technical health consultants allowed for direct grievance redressal, whereas MOs were more reluctant to share problems with State‑ and District‑level authorities during Review meetings.

Use of data

The State government launched the MOTHER app in 2019 to improve monitoring of data on maternal and child health. The app is meant to integrate real‑time data from the field collected and inputted by frontline staff (ANMs, ASHAs) for doctors and health officials at the block‑, district‑ and state‑levels to track progress through a dashboard. ASHAs and ANMs reported that the MOTHER app was one amongst many apps and portals across health programmes that they were expected to update, resulting in a time‑consuming process. Lack of network connectivity was identified as a key barrier to timely uploading of data. Capacity to use these apps appeared mixed, with older ANMs and ASHAs facing difficulties. Manual record‑keeping appeared to supplement data recording on apps. Data pertaining to the Rescue Mission was also shared with functionaries in the Social Welfare and Rural Development departments at the Block and District levels. The district‑level leadership used data to identify problems, such as tracking trends, identifying problem areas and informing action plans.

MOs drew on data to monitor performance on key service indicators during review meetings, but the degree of analysis and depth of action plans varied. Some MOs reported challenges such as requests for reporting in multiple formats with tight deadlines and poor network connectivity, as well as not being aware about how Blocks and Districts utilised the data they reported.

Adopting the state capability enhancement approach

MO adoption of SCEP principles ranged from a score of 5 to 13, with a median score of 10 (Figure 5). The index suggested that most MOs adopted the basic principles, but did not go beyond in most cases. Most importantly, we noted the variation in MO approach did not correlate with the type of PHC, system performance (such as institutional delivery rate) or context. Rather, sense of purpose and problem‑solving capabilities emerged as intrinsic qualities.

Figure 5

Community mobilisation

The Rescue Mission implemented community mobilisation through the combined, multisectoral strength of frontline workers (SHGs, ASHAs and Anganwadi workers). SRLM officials recognised that the strong social capital of SHGs was a valuable contribution to improving maternal health. However, some noted diversion from their main activities in a context with human resource shortages and increased demands for data reporting. The Department of Social Welfare had already collaborated with Health on overlapping population groups of pregnant women and children, such as health check‑up camps.

The Rescue Mission introduced a new cadre of Community Gender & Health Activists (CGHAs) drawn from SHGs. With their different backgrounds, CGHAs were to focus on social determinants such as food, nutrition and sanitation to support the ongoing work of ASHAs. They capitalised on a closer rapport with community members, as they were drawn from SHGs, compared to ASHAs who were seen as functionaries of the Health Department. At the time of data collection, grassroots convergence appeared to be strong only in geographies where SHGs were active and already interacted with the heath system. CGHAs highlighted certain challenges to their work, such as long working hours, a lack of transportation to conduct field visits and walking long distances in hard‑to‑reach areas. Given the overlapping mandate between the two departments, CGHAs noted difficulties with coordination as they found themselves answerable to functionaries across two departments.

Village health councils

Village Health Councils (VHCs) were introduced in 2022 as community institutions to build community ownership for health and nutrition in Meghalaya. Meant to serve as a link between the health system and communities, they have replaced previous Village Health, Sanitation and Nutrition Committees (VHSNCs). VHCs comprise of elected rather than appointed members, including leadership from Meghalaya’s traditional institutions and village organization presidents from women’s SHGs.

At the time of data collection, VHCs were just being set up and therefore, relatively new. However, we noted that in communities where VHSNCs were active previously, VHCs continued similar activities and demonstrated community ownership. Some VHCs activated non‑functional Sub Health Centres and headmen emerged as active leaders who built on previous experience creating community‑owned assets through Village Employment Councils.

Reflections with government

The theory of change posited changes in both technical and adaptive capacity of the health system and specifically MOs. While technical capabilities in using data or implementing programmes could be strengthened directly through training, the adaptive capability of MOs emerged as intrinsic skills that were not built through external inputs. At the system level, a deepened ownership of intersectoral collaboration – a longstanding challenge in governance – emerged as a key achievement of continued efforts through joint meetings and collaboration at the front line. Discussions with government after our first round of findings led to recognition of the need for closer engagement with MOs directly through local sessions, tailored training and learning across districts—rather than relying on Review Meetings to strengthen problem‑solving capability. The government leadership articulated that the central challenge for the Rescue Mission was to intensify efforts at building the “soft architecture” to achieve success in implementation, rather than technocratic approaches to problem solving.

Transferability

Three elements of Rescue Mission offer lessons for other settings on multisectoral approaches towards health. One, political leadership was a central, non‑negotiable feature – which overcame a well‑known challenge that the “most important barriers to multisectoral action are political rather than technical” [23]. Political support for maternal health was actively nurtured by policymakers from the outset, exemplified by the widely promoted Chief Minister’s Safe Motherhood Scheme. Maternal health could attain political priority in Meghalaya because it represented a convergence of problems and policies at the national and state‑level, further supported by an international focus on MMR in the SDGs [21, 24, 25]. The signal sent from political leadership established that maternal health was “everyone’s problem” and not solely that of the health department. Two, administrative leadership demonstrated problem‑solving in practice, both during State‑level meetings and facility visits. The importance of seeing empathetic leadership in practice emerged in reflections from interviewees, particularly the power of personal visits to facilities and following through on commitments of hands‑on support.

Three, building the case for maternal health for non‑health departments contributed to ownership. This process required communicating clear linkages or finding a common area of interest between health and mandates which fell outside the purview of health but addressed social determinants of health nonetheless. The department of Social Welfare had a natural overlap with activities on early childhood development and nutrition. However, with the Rural Development department, percolation of linkages between good health and improved livelihoods – a key aim of their work – from the State to grassroots levels required time and effort, especially for frontline cadres who had increased work as a result of multisectoral efforts. To build consensus for multisectoral action for health, clarity of linkages is essential for the buy‑in of non‑health departments, which was relatively straightforward for maternal health but may require further efforts for different health issues [23, 26].

Conceptually, the government’s focus on adaptive, rather than only technical, expertise drew from the problem‑driven iterative approach (PDIA), which supports governance methods that promote: local solutions for local problems; problem‑driven positive deviance; active learning and experimentation; and scaling through diffusion through engaging champions [27]. While PDIA specific to health systems and governance is an emergent field, its principles—albeit with a different set of actors—resonate with a well‑established method in India to address maternal health through engaging women and mobilising communities through cycles of participatory learning and action [28]. The further evolution of SCEP in Meghalaya to building empathy through active, grassroots engagement echoes wider experience in India with early government‑promoted self‑help groups in Andhra Pradesh that engaged in extensive, bottom‑up village‑level mobilisation [29].

Strengths and limitations

The embedded nature of the study allowed for “insider” access to review meetings, audits and most importantly, a rapport with leadership to discuss findings and shift focus as needed. To our knowledge, this study was amongst the first to examine application of a state capability approach to the health system; future efforts can expand methodological approaches accordingly. The exploratory, adaptive nature of this study allowed us to intensify focus on MO motivation mid‑way, for example. Future research can build on this with more structured evaluation of the impact of the Human Development Leadership programme and to identify mechanisms that influence building purpose [8, 30]. Regular interviews and visits to facilities allowed for in‑depth insights from a sample of MOs, although supplementation with a quantitative survey across the state may have facilitated greater insight into variation across the state. Moreover, future research should integrate more extensive research with community institutions and service users to supplement the perspectives of providers and government officials.