Significance Statement

What is already known on this subject? Cross-sector collaboration has become increasingly important for public health practice. HIA is an effective tool for building collaborative relationships across sectors. The MCH workforce is in need of actionable frameworks for cross-sector collaboration to impact social determinants of health and equity.

What this study adds? HIA practice in the US has not seen extensive involvement of the MCH workforce, but it has regularly incorporated MCH-relevant content. By filling this gap and becoming more involved in HIA, MCH professionals have an opportunity to build cross-sector capacity and ensure relevance of MCH content.

Introduction

Researchers, practitioners, and policy-makers need to address social determinants of health to meaningfully reduce population-level disparities and improve maternal and child health (MCH). Social determinants of health are conditions in the environments where people are born, live, learn, work, play, worship, and age that affect a wide range of health, functioning, and quality-of-life outcomes and risks (U.S. Department of Health and Human Services Office of Disease Prevention and Health Promotion, 2018). Addressing disparities requires a commitment to health equity, which means striving for the highest possible standard of health for all people and giving special attention to needs of individuals at greatest risk based on social conditions (Braveman, 2014). The Maternal and Child Health Bureau of the Health Resources and Services Administration has emphasized a need for integrating perspectives of social determinants and equity into MCH practice (Fine et al., 2010). Meeting this need requires collaboration between public health and other sectors like education, housing, and transportation (DeSalvo et al., 2016; Koh et al., 2011; Mattessich & Rausch, 2014).

Research on training needs of the public health workforce, and the MCH workforce specifically, emphasizes policy engagement and collaboration with sectors outside traditional public health silos as areas for improvement (Bogaert et al., 2019; DeSalvo et al., 2017; Raskind et al., 2019; Sellers et al., 2015). As a result, the National MCH Workforce Development Center advances strategies to help Title V leaders and MCH practitioners build collaborative policy engagement capacity, using systems integration perspectives that emphasizes social determinants of health and equity (Clarke & Cilenti, 2018; Margolis et al., 2017). One of these strategies is implementing a Health in All Policies (HiAP) approach.

HiAP considers how to systematically integrate public health perspectives on social determinants and equity into decision-making across non-health sectors (Ståhl et al., 2006; Wernham & Teutsch, 2015). One of the more common HiAP approaches is health impact assessment (HIA) (Gase et al., 2013; Rudolph et al., 2013). The National Research Council defines HIA as:

A systematic process that uses an array of data sources and analytic methods and considers input from stakeholders to determine the potential effects of a proposed policy, plan, program, or project on the health of a population and the distribution of those effects within the population. HIA provides recommendations on monitoring and managing those effects (2011).

While variations exist in practice, conducting HIA typically involves the iterative six-step process illustrated in Table 1 and components of the eight Minimum Elements listed in Table 2 (Bhatia et al., 2014; Dannenberg, 2016a; National Research Council, 2011).

Table 1 Six steps of health impact assessment and associated outputs (National Research Council, 2011)
Full size table
Table 2 Eight minimum elements of health impact assessments (Bhatia et al., 2014)
Full size table

HIA is a practical tool for synthesizing evidence and stakeholder input to foster collaboration and inform decisions made outside the health sector (Cole & Fielding, 2007). Because individual HIAs are primarily oriented toward practice, they are under-documented in peer-reviewed research literature. HIA researchers launched a dedicated journal in 2016 to address this issue and to provide more opportunities for HIA practitioners to publish their results (Stone, 2016). Over the past decade, published research on HIA in the United States has mostly considered effectiveness of the process, while individual case studies remain elusive in traditional research databases (Cole et al., 2019; Dannenberg, 2016b; Sohn et al., 2018). For example, searching the key words “health impact assessment” in PubMed returns a large number of results, but most are not specific to the prospective process defined above.

A PubMed search for “health impact assessment” conducted by the authors in January 2020 returned 1,264 results. When sorted by relevance, the first 100 results included 19 published accounts of specific HIAs, of which five were conducted in the United States. In the MCH context, even fewer research publications regarding HIA are available: as of January 2020, the Maternal and Child Health Journal has published no articles describing “Health Impact Assessment” in their title or abstracts. To the authors’ knowledge, the current study is the first of its kind to document potential for alignment between MCH and HIA practice in the United States and to highlight the breadth of opportunities for MCH practitioners to engage in such HiAP initiatives as a strategy to address the social determinants of health.

Methods

Research Overview and Key Questions

This study leveraged a repository of HIAs conducted in the United States to answer these questions: (1) to what extent have MCH-focused stakeholders been involved in HIA practice? (2) How often and to what extent do HIAs examine MCH subpopulations? (3) What proportion of HIAs include Title V National Performance Measures (NPMs) or closely related metrics in their analyses? (4) Are HIAs that engage MCH-focused stakeholders more likely to include information on MCH populations and NPM-relevant content? Fig. 1 illustrates the interrelatedness of the four research questions. Collected data also characterize the breadth of non-health policies relevant to MCH outcomes and a range of opportunities for MCH stakeholders to inform them. This study does not include any clinical or patient data.

Fig. 1
figure 1

Interrelatedness of four research questions guiding the review of HIAs

Full size image

Data Source and Inclusion Criteria

The Health Impact Project, a collaboration of the Robert Wood Johnson Foundation and The Pew Charitable Trusts, hosts a web-based repository of information about HIAs conducted in the United States (Health Impact Project, 2017). HIA researchers commonly use this repository as a data source (Cole et al., 2019; Cowling et al., 2017; Dannenberg et al., 2019; Gase et al., 2017). Although uploading HIA information to the site is voluntary, this repository has been viewed as a largely complete picture of US practice since its inception in 2010.

As of September 2017, the repository listed 424 HIAs. The entry for each HIA includes hyperlinks and contextual data like project title, lead organization(s), decision-making level (local, regional, state, or federal), and target sector (e.g. housing, education, transportation, etc.). HIAs were excluded from the current analysis when project documentation was not available, either because the HIA was in progress or because available hyperlinks were no longer active and reviewers were unable to find publicly available documentation via web search based on HIA title and/or lead organization.

Key Variables

Researchers coded three primary variables for each HIA reviewed. First, they determined the involvement of MCH-focused stakeholders. MCH-focused stakeholders included organizations or individuals that focus on infants, children, youth, adolescents, mothers, pregnant individuals, and/or families. State or Local Health Departments were only included as MCH-focused stakeholders if available documentation clearly noted representation from an MCH-serving division, branch, or team (e.g. the Title V Office, Division of Child Health, etc.). Reviewers did not assume inclusion of these perspectives unless stated. This variable was coded ‘yes’ if documentation revealed an explicit role for an MCH-focused partner in the conduct of the HIA, ranging from roles on the HIA project team or advisory committee to participation in document review or data collection as part of the process. If available documentation mentioned MCH-focused stakeholders but was unclear about the extent of involvement, reviewers coded this variable as ‘partial.’ The ‘partial’ designation was also used when the reviewer was unsure about the extent to which a given stakeholder explicitly focused on MCH, for example local educational institutions were regularly placed in this category.

The second variable considers attention given to MCH populations within the HIA. MCH populations include infants, children, youth, adolescents, mothers, pregnant individuals, and/or families. Reviewers coded this variable as ‘yes’ if documentation noted explicit focus on one or more of these MCH populations as part of the HIA scope. This meant that the HIA not only identified an MCH population as a subpopulation of concern, but that the assessment and/or recommendations reflect elevated attention on that group. Reviewers assigned a ‘partial’ designation if the HIA mentioned MCH populations but did not clearly incorporate them as part of the assessment. This included instances where a report includes statistics on an MCH population but not as a critical element of the analysis (e.g. the childhood obesity rate was mentioned as part of existing conditions but not in the assessment of potential impacts).

To define the third key variable, inclusion of MCH content, reviewers determined whether each HIA included topics related to one or more of the Title V NPMs. These 15 measures were selected as the basis for defining "MCH content" in this review because national leaders have identified them as foundational to the work of public health MCH professionals, as measurable, as directly modifiable by states’ programmatic activities, and as meaningful when changed (Kogan et al., 2015). Specific guidance and search terms were developed for each NPM and are listed in Table 3, along with examples of reviewer notes from specific HIAs. Reviewers only used the ‘yes’ designation when the exact NPM was noted in the HIA. They used the ‘partial’ designation when an NPM topic was touched upon in the HIA, but without including the exact NPM. A goal of this review is to inspire MCH professionals toward more upstream cross-sector engagement and collaboration by documenting the breadth of opportunities where their perspectives could have been valuable in existing HIA practice. Because this retrospective accounting of opportunities was generated from content positioned primarily within non-MCH perspectives, the authors were intentionally inclusive in identifying aspects of a given HIA that could be considered MCH-relevant content, and thus, some instances may stretch the bounds of what is considered closely related to specific NPMs.

Table 3 Specific guidance for HIA reviewers used to characterize NPM inclusion
Full size table

Review and Extraction Process

Starting with information available in Health Impact Project’s repository, researchers followed a protocol created and pilot-tested specifically for this review. First, they used project hyperlinks to obtain publicly available HIA documentation. This documentation was then reviewed to determine designations for the three key variables described above. In addition to a scan of available documentation for each HIA, variable-specific search terms were employed to more efficiently identify relevant content. Reviewers extracted this relevant content and summarized how it supported designations of ‘yes’, ‘partial’, or ‘no’ for each variable. A pilot-test of the protocol on 20 randomly selected HIAs included independent review of each by three researchers, who agreed 85% of the time across all variable designations. No HIA in the pilot explicitly included any of the 15 NPMs, which led researchers to adjust the final review protocol for this variable to be more inclusive of related content within the ‘partial’ designation. Two co-authors then collaboratively reviewed the remaining HIAs and adjudicated disagreements through discussion as needed. When reviewers could not make a clear determination, the research team made a final coding determination based on collective interpretation of the available content in the context of the study objectives. Reviewers applied the protocol to available materials and coded responses into an Excel spreadsheet.

Analysis

Coded data were transferred to SPSS (IBM Corp., 2017), and frequency tables were produced for each research question. Upon initial review of the data, yes/partial/no responses were recoded to facilitate statistical analysis and hypothesis testing. For the MCH stakeholder and MCH population variables, ‘no’ and ‘partial’ responses were combined to allow dichotomous comparison to the ‘yes’ responses. Because so few HIAs were coded as ‘yes’ for including NPMs, responses for this variable were dichotomized by combining ‘yes’ and ‘partial’ designations for comparison with those HIAs designated as ‘no.’ Researchers also created a variable that indicated the total number of NPMs each HIA addressed, ranging from 0 to 15.

To test the hypothesis that HIAs with MCH stakeholder involvement are more likely to include a focus on MCH populations and NPM-related topics, a single dichotomous dependent variable was created that combined inclusion of MCH populations and NPM-relevant topics. These binary data were used to conduct Chi-Square and Simple Odds Ratios to test the hypothesis. Statistical significance was determined at alpha = 0.05.

Results

Figure 2 illustrates how the 424 HIAs contained in the Health Impact Project database were narrowed to an analysis sample of 350 HIAs.

Research Question 1: To what extent have MCH-focused stakeholders been involved in HIA practice?

Fig. 2
figure 2

Inclusion diagram

Full size image

Slightly less than one quarter (24%) of the 350 HIAs reviewed demonstrated clear inclusion of MCH stakeholders (Fig. 3). The only example of an HIA that clearly indicated inclusion of a Title V agency considered impacts of potential changes to paid sick leave policy in Vermont (Vermont Department of Health, 2015). In that HIA, led by the Vermont Department of Health, representatives from the Division of Maternal and Child Health were listed as stakeholders, along with the Vermont Commission on Women, several school nurses, and a local child care facility. Under the definition of stakeholders described above, the most common stakeholder groups were school or other education system representatives and child or family-focused advocacy organizations, noted in 42% and 39% of HIAs coded as including MCH stakeholders, respectively.

Research Question 2: How often do and to what extent do HIAs examine MCH subpopulations?

Fig. 3
figure 3

MCH stakeholder inclusion in HIAs

Full size image

Eighty-four percent of the 350 HIAs reviewed at least partially considered MCH populations according to publicly available materials. As illustrated in Fig. 4, this 84% was split almost equally between HIAs designated as having a clear focus on MCH subpopulations (148 HIAs) and HIAs that only partially included them (147 HIAs).

Research Question 3: What proportion of HIAs include Title V National Performance Measures (NPMs) or closely related metrics in their analyses?

Fig. 4
figure 4

MCH population inclusion in HIAs

Full size image

Eighty-nine percent of the 350 HIAs included information designated as closely related to or relevant for at least one NPM, and less than 2% included specific NPMs. The mean number of NPM-related topics included in an HIA was 2.62 (SD 1.80). The largest proportion of HIAs (30%) included two NPMs, and most (61%) included one to three topics (Fig. 5). All 15 NPMs were represented at least once across all HIAs reviewed (Fig. 6).

Fig. 5
figure 5

Frequency of HIAs by number of NPM-related topics included

Full size image
Fig. 6
figure 6

Percent of HIAs including each NPM-related topic

Full size image

Injury and physical activity were the most common NPM-related topics, included in 61% and 59% of the 350 HIAs reviewed, respectively. None of the 214 HIAs including information about injuries used the specific NPM. However, all of them referenced injuries and hospitalizations in a way determined by reviewers to be relevant for MCH populations or practice. For physical activity, only one of the 205 HIAs coded as relevant for NPM 8 specifically included NPM itself.

Research Question 4: Are HIAs that engage MCH-focused stakeholders more likely to include information on MCH populations and NPM-relevant content?

The MCH stakeholder, MCH population, and NPM inclusion variables described above were used to test the hypothesis that HIAs involving MCH stakeholders are more likely to address both MCH populations and relevant measures. MCH stakeholder inclusion was the independent variable and dichotomized into HIAs with clear inclusion of MCH stakeholders (designated by reviewers as ‘yes’; 24% of the sample, 84 HIAs) and ones with no clear inclusion of MCH stakeholders (designated as either ‘partial’ or ‘no’; 76% of the sample, 266 HIAs). The dependent variable was a combination of MCH population focus and inclusion of NPM-related topics, where an HIA had to satisfy both conditions to be coded as ‘yes’ (41% of the sample, 143 HIAs).

HIAs that clearly included MCH stakeholders had seven times the odds of including both a focus on MCH populations and at least one NPM-relevant topic compared to HIAs that did not clearly include MCH stakeholders (OR 6.98; 95% CI 3.99, 12.20). This association is statistically significant (Chi-square, p < 0.001). Table 4 displays the distribution across these two variables.

Table 4 Inclusion of MCH stakeholders by MCH population focus and NPM-related topics
Full size table

To demonstrate the breadth of non-health policies informed by HIA, Table 5 includes the distribution of HIAs across different target sectors. Table 6 summarizes ten HIAs that illustrate the range of potential opportunities for MCH engagement in this type of cross-sector collaboration. These examples were selected to reflect variation in the three key review variables and in two key contextual factors: target sector and decision-making level (local, regional, state, or federal). They were also chosen to reflect the geographic diversity in HIA practice in the US.

Table 5 Distribution of HIAs by sector
Full size table

Discussion

Nearly all HIAs reviewed (95%) considered an MCH population in some way or included information relevant to at least one NPM. In contrast, just over a third (36%) contained evidence of MCH stakeholder involvement. The current analysis demonstrates that MCH stakeholder involvement in an HIA is associated with greater odds of that HIA addressing MCH populations and relevant topics, suggesting an opportunity for MCH professionals and HIA practitioners to engage with one another in strategic efforts to inform decisions made outside the traditional public health sector. Aligning these efforts allows both MCH professionals and HIA practitioners to better advance a “Public Health 3.0” approach, wherein public health leaders partner across multiple sectors to address social, environmental, and economic conditions that affect health and health equity (DeSalvo et al., 2017).

MCH professionals have opportunities to advance health equity through cross-sector collaboration by actively pursuing involvement in HIA taking place in their respective jurisdictions and strategically considering it as a means for addressing issues relevant to their stakeholders and communities. Where HIAs are more common, there appears to be opportunity for intentionally integrating MCH perspectives into ongoing cross-sector collaborations addressing health determinants of significance for MCH outcomes. Where HIAs are less common, there may be opportunities for MCH leaders to initiate or strengthen those critical collaborations through the use of this specific tool. As practice continues to coalesce around advancing health equity, both in MCH efforts and through the HIA process, opportunities for collaboration will emerge and can be leveraged to advance population health outcomes for MCH populations. Collaborative cross-sector opportunities are not exclusive to HIA, and future research could expand upon the current review to further explore roles for MCH in other HiAP approaches.

HIA practitioners have opportunities to more actively seek engagement with MCH stakeholders, which could lead to broader applicability of HIA findings and recommendations. Incorporating more intentional MCH perspectives within HIA could be a mechanism to garner broader stakeholder and community engagement, which in turn can raise visibility of strategic approaches to population health and health equity. There could also be practical benefits in terms of access to and interpretation of MCH-relevant data that might otherwise be absent in a given assessment. Incorporating this perspective could strengthen the recommendations that are a primary output of the HIA process.

Although not coded as a key variable for this analysis, reviewers noted general MCH implications of recommendations in roughly two thirds of HIAs. If Title V or other MCH professionals are engaged in developing these recommendations, they can better ensure strategic relevance for MCH stakeholders and alignment with other MCH activities in a given jurisdiction. For example, with Title V involvement, an HIA could move past general discussions of an MCH-related issue to include NPM-specific data and strategies, resulting in direct advancement of relevant existing Title V priorities.

Intra-sector silos within public health may provide one explanation for underrepresentation of MCH stakeholders in HIA practice. When led by public health entities, HIAs are often housed within an environmental health or chronic disease unit; though in recent years, HIAs led by offices focused on evaluation, policy, or equity have become more common (J. Dills, personal communication, April 1, 2020). This assignment of HIA responsibility to specific public health content areas might raise barriers to supporting active involvement of MCH professionals from within an agency, among other bureaucratic challenges. Without active attempts to bridge these intra-sector silos and address these challenges, HIA work done by one arm of a public health agency can easily occur in isolation from the other. As MCH leaders become more engaged in upstream policy work, awareness of opportunities to leverage existing cross-sector experience from within their own agencies is an asset. HIAs potentially present this type of opportunity, as do HiAP approaches more broadly (Rudolph et al., 2013). A more thorough examination of variation in resources available to support HIA over time is beyond the scope of this review; however, future research could determine if greater availability of resources to support HIA is associated with greater likelihood of MCH inclusion in those HIAs.

In terms of connecting to sectors outside of public health, Table 5 shows the distribution of HIAs by target sector. HIAs aimed at informing decisions related to agriculture/food systems, education, and labor/employment have included MCH stakeholders more regularly than HIAs targeting other sectors, but all sectors have included MCH stakeholders to some degree. These data are included to demonstrate the breadth of sectors covered by HIAs generally and to note the sectors most common to HIAs that specifically include MCH stakeholders. This review does not consider the temporality of MCH involvement in HIA and the MCH-relevant content contained therein. Some decisions considered by HIAs may be more directly tied to MCH populations and have a readily apparent reason to include MCH practitioners, leading to more MCH-relevant content in those HIAs. For example, decisions made in the education sector are likely to influence children, so an HIA of those decisions may naturally include some level of MCH perspectives from the outset. As a collaborative approach, different perspectives inform HIAs throughout the process, with the screening and scoping stages as the most opportune time for robust MCH involvement that would potentially guide assessment and recommendations toward more MCH-relevant content.

Limitations

Three interrelated limitations should be considered in interpreting these findings. First, approximately one in six HIAs listed in the Health Impact Project database did not have the publicly available documentation needed to be included in this analysis. Further, the database is populated mostly by voluntary self-reports from the HIA field, and thus, an unknown number of HIAs in the U.S. are not included in the data. How information from these HIAs might influence study results is unclear. Second, thoroughness of available HIA documentation varies widely (Rhodus et al., 2013). HIA is both a process and a report. This review only judged aspects of the process based on how well the associated reports (and other available documentation) described them. This limitation leads to an incomplete picture of HIAs that may have had a more robust process than reflected in their final reports and could have led to underestimating MCH stakeholder involvement. Finally, because of the exploratory nature of this review, the research team erred on the side of inclusivity when coding HIA content. Regular discussions of the data as they were being collected and resulting iterations of the review protocol may have introduced increased subjectivity in comparison to a more traditional systematic review of research studies. Future research examining involvement of MCH stakeholders in HIA practice could address some of these limitations by including interviews with leaders of and participants in specific HIAs.

Conclusion

HIA is an established tool that supports prospective engagement by public health stakeholders in non-health decision-making contexts. This exploratory review indicates limited involvement of MCH-focused practitioners in HIA practice to date. The results reveal that despite this lack of engagement, a large portion of HIA practice considers populations and measures with clear relevance to MCH. When the MCH workforce is involved, HIAs are significantly more likely to include specific MCH content. Therefore, intentional alignment of MCH and HIA practice, along with capacity building to support use of HIA by MCH stakeholders, would potentially reinforce strategies seeking to inform decisions made in other sectors. As noted earlier, these decisions influence social determinants of health, the inequitable distribution of which underpin many health disparities. HIA is one method through which public health perspectives can be integrated into these influential decisions for population health. Limited involvement of MCH practitioners in HIA points to a missed opportunity to ensure that their particular perspectives on public health are actionably included in these cross-sector collaborations to address social determinants.

Table 6 Selected HIA examples
Full size table