Home --> Health Equity Lens --> Data, Research and Evaluation --> Data to Identify and Understand Health Inequities
|
Data to Identify and Understand Health Inequities
|
|
IN THIS SECTION
- Data Use - Innovative Types of Data - Strengthening Existing Data Capacity - Successful Use of Data for Health Equity - Limitations of Data Collection and Analysis |
It is important to note that health status data is not necessarily available for all population groups.Health status data is largely lacking for members of the Lesbian, Gay, Transgender, Bisexual, and Questioning (LGTBQ) population. A 2011 report from the Institute of Medicine (IOM), titled The Health of Lesbian, Gay, Bisexual, and Transgender People: Building a Foundation for Better Understanding, highlights the need for targeted data collection and research regarding the health status of this population. (IOM, 2011) |
Data Use
Both types of data are necessary to describe baseline status and to monitor changes over time as well as to make comparisons by place.
- Baseline data helps practitioners, policy makers, and community residents identify priorities and ensure that interventions reflect the community’s needs and resources. Tracking changes over time allow for evaluations of interventions as well as ensure that they remain focused on key priorities and are accountable to stakeholders.
- Neighborhood-level data are needed to facilitate the identification of priorities and other kinds of decision-making in different geographical locations. This can be challenging because many existing data sources do not allow for neighborhood-level analysis and/or would require substantial resources to do so.
Innovative Types of Data
Looking at the upstream causes of health inequities allows stakeholders to focus on the most meaningful indicators and helps shift the focus from individual risk factors and behaviors to community health and the structures that underlie inequities. Therefore, data that describes SDOH are needed to highlight the ways in which unequal power and privilege influence the distribution of resources required for health (Knight, 2014). However, this much needed data on social and structural determinants are not readily collected or collected in systematic ways.
For instance, in addition to monitoring high school graduation rates, it is valuable to collect information and monitor changes in per capita spending on public education. Similarly, the availability of affordable housing is an important SDOH, but the level of racial segregation in a defined community is necessary to paint a more complete picture.
Data Collection Tools
Investments must be made in creative and novel approaches to data collection such as:
Analytical Models
Many models analyze the underlying causes and factors of health outcomes. Ex: Root Cause Mapping |
The use of these models in most case would involve partnering with other sectors that have existing data to support a broader understanding of SDOH and health inequities. Furthermore, community members should be engaged in identifying, collecting, and interpreting new kinds of data for health equity.
|
Recommended Resource
Practitioner Guide for Advancing Health Equity: The Centers for Disease Control and Prevention's (CDC) guide offers a systematic list of questions for practitioners in government and community-based organizations to reflect upon when building their capacity for identifying and understanding health inequities (CDC, 2013, p. 21)
Practitioner Guide for Advancing Health Equity - Understanding Health Inequities
1) Where are we now?
- What are our organization’s current practices for identifying and understanding health inequities?
- Can we clearly articulate health inequities related to the health issues we are trying to prevent and/or address? If so, list those health inequities.
- What process can we set up to get a full understanding of health inequities in our community?
- What type of information do we need to ensure we have a full understanding of health inequities in our community?
- Have we looked beyond basic health risk behaviors and standard outcome data to examine social, economic, and physical indicators that may contribute to or maintain health inequities?
- Have we examined community context and historical factors that may help our understanding of existing health inequities?
- What combination of data sources do we need to better understand experiences of populations affected by health inequities?
- What sources or partners may already have the data we need for assessing community environments or health behaviors?
- Where can we go to understand the historical context of health inequities in the community?
- How do we currently engage community members in our data collection and analysis process?
- What process can we put in place to routinely engage populations affected by health inequities in collecting and analyzing data?
- What can we do differently to improve or enhance our ability to identify and understand health inequities?
- What is our plan of action to implement those changes?
Strengthening Existing Data Capacity
Understanding health inequities and their determinants can be improved by collaborating across sectors that may already collect the kinds of data that are needed. Similarly, it may be possible to make greater use of existing data within public health surveillance systems or within health and human service agencies. This involves linking data systems in ways that provide a more comprehensive view of community health. Adding data from one database to another can be resource intensive, and may require addressing legal barriers in addition to overcoming technical barriers. It is critical to ensure the protection of privacy when working with individual- level data, particularly as the groups most affected by inequities may already experience disadvantages related to their identity.
NATIONAL EFFORTS
Several national databases can be used to understand health inequities and their causes at the local level. The Data Set Directory of Social Determinants of Health at the Local Level contains an extensive list of existing data sources across 12 dimensions of the social environment, including: economy, employment, education, political, environmental, housing, medical, governmental, public health, psychosocial, behavioral, and transportation (Hillemeier, Lynch, Harper & Casper, 2004). Within each dimension, the directory includes several important indicators and data sources to describe those indicators. For instance, the political dimension identifies voter registration and voting rates as important indicators of civic participation and offers a specific data table within the Census Bureau dataset as a source for those indicators. The behavioral dimension includes indicators commonly used in public health surveillance, such as smoking rates and levels of physical activity. However, it also includes indicators such as the average local price of cigarettes and physical education requirements in schools. These latter indicators speak to the social and structural characteristics of the environment, which allow public health practitioners and partners to better understand upstream root causes. For the full directory, visit: http://www.cdc.gov/dhdsp/docs/data_set_directory.pdf. Appendix C from the CDC Practitioner Guide for Advancing Health Equity (2013) contains additional examples of resources for identifying and understanding health inequities. |
DELAWARE FOCUS
In the state of Delaware, the potential for such linkages can be facilitated by initiatives such as the Master Client Index (MCI), which tracks unique clients in each of the programs within the Department of Health and Social Services (DHSS) and the Department of Services for Children, Youth and their Families (DSCYF) (see http://dhss.delaware.gov/dhss/dms/irm/files/mci_interfacing_requirements.pdf). Similarly, the Delaware Health Information Network (DHIN) is a statewide health information exchange among health care providers that facilitates an integrated data to improve patient outcomes and patient-provider relationships, while reducing service duplication and health care spending (see http://dhin.org). These, and other data-sharing initiatives, can provide the foundation for more concerted health equity oriented efforts. |
Existing data collection and surveillance activities can also be strengthened with respect to the collection of race, ethnicity, and language data. Although race, ethnicity, and language data is captured in databases such as vital statistics and health care records, it is not collected consistently through other surveys, programs, or databases. It is recommended that race, ethnicity, and language data be collected across sectors and collected by a variety of agencies including government, non-profit organizations, and academic institutions, among others.
Minnesota Department of Health and Department of Human Services Recommendations
A race, ethnicity, and language workgroup of the Minnesota Department of Health and the Minnesota Department of Human Services specifically recommends the following:
- More detailed categories of race and ethnicity data should be used so that the data are more useful in understanding health issues and needs for particular groups.
- State agencies and organizations that collect and use health data should be regularly engaged with diverse communities to promote full understanding of how race, ethnicity, language, and culture affect quality, access, and cost of health services.
- Data collected by state agencies and health care organizations should be as accessible to communities, as possible. The criteria and process for obtaining access to data should be provided to and discussed with the communities, and agencies should take steps to ensure that information about relevant datasets is easily available online.
- A workgroup (such as the one that developed these recommendations) should continue on an ongoing basis so communities, health care stakeholders, and government agencies can partner to improve data collection policies and practices and, using the data, eliminate health inequities.
- A uniform data “construct” should be developed so that all health data collected use the same categories for race, ethnicity, and language. The uniform construct should be used not just by state health agencies, but also by licensing boards, other governmental agencies, health plans, hospitals, clinics, non-profit agencies, quality and performance measurement programs, and others who collect, analyze, and report health data. In this way, disease burden, risk and protective factors, access to care, and quality of care can be measured and communicated for smaller populations within an overall population. The uniform construct should build on existing frameworks for data collection, to eliminate duplication of effort. The data construct should be flexible so categories can be changed as needed. A process should be developed for assessing changes in racial/ethnic populations in the state and determining when populations are of a sufficient size to be reported as a separate category.
- Programs that rely on survey data should consider over-sampling or mixed mode approaches to obtain larger numbers for communities of color (MDH/MDHS, 2011).
Data Tools for Health Equity
Capacity to address health inequities at the community level can be strengthened by using various tools that help describe public health issues and available resources at the community level.
Geographic Information System (GIS)
GIS data may be used in concert with health data to generate maps, which provide a powerful tool for visualizing health inequities at the community level. More specifically, maps can be used to analyze spatial patterns of health and illness in tandem with social inequities such as poverty and income, race/ethnicity, and environmental health hazards (MDH, 2014). Ultimately, GIS maps can distill otherwise complex information into easily understood images. Importantly, they can be used to promote policy change, particularly because they can focus attention on areas defined by political boundaries (e.g. congressional districts). Maps in Section 3 of the health equity guide, also presented on this page, were produced using GIS Health Impact Assessments (HIAs)
The use of Health Impact Assessments (HIAs) described in Section 6 of the guide, also presented here, require a different kind of analytical approach, research skills, and sources of data than traditionally used in public health. However, they also offer an important way of understanding existing health inequities and the changes (both positive and negative) that may result from proposed policy changes. Specifically, HIA's call for community-based approaches to data collection and analysis; are grounded in the principles of equity, inclusion and democracy; often rely on mixed data collection methods (i.e. quantitative and qualitative approaches); and make connections between health and social and environmental conditions and structures. There is also a strong focus on dissemination and utilization of the results of the analysis. An in-depth description of this tool and its application can be found here. Community Health Assessments (CHAs)
Another opportunity for addressing health equity data needs at the community level exists through the use of community health assessments conducted by non-profit hospitals. The Affordable Care Act (ACA) now requires tax-exempt hospitals to regularly (at least every three years) conduct community health needs assessments and develop plans to address those needs. The law strengthens the hospitals’ obligation to work with public health agencies and others in this regard. Therefore, public health practitioners can partner with hospital administrators to support their data collection efforts and encourage them to implement action plans that focus on SDOH and equity. |
“Without a clear understanding |
Successful Use of Data for Health Equity
The Case of Minnesota
In its report to the state legislature of Minnesota, the Minnesota Department of Health identified “four keys to the successful use of data for addressing health inequities” (MDH, 2014). These recommendations apply to the collection of new data, the improvement of existing data, and the use of tools such as GIS mapping and HIA. A detailed description of each recommendation and a sample application strategy is presented below.
In its report to the state legislature of Minnesota, the Minnesota Department of Health identified “four keys to the successful use of data for addressing health inequities” (MDH, 2014). These recommendations apply to the collection of new data, the improvement of existing data, and the use of tools such as GIS mapping and HIA. A detailed description of each recommendation and a sample application strategy is presented below.
1. Make the data useful in terms of analysis, interpretation, and application.
This suggests that many different kinds of techniques may be needed for the collection, analysis, and reporting of data related to health equity. The approaches that are used will depend on the purpose or intended use of the data.
2. Results must be disseminated effectively.
Practitioners must consider their audience when deciding how to share their findings to achieve maximum impact. For example, data meant to inform policy change will be of little use unless policy makers can understand and appreciate the information. Different and creative channels for dissemination should be considered, such as interactive platforms and websites, newsletters, emails, and community forums. A public access web portal with interactive capabilities, such as allowing users to select indicators and geographic locations, can be particularly useful. At the same time, this approach may require substantial ongoing investment of staff to manage the portal’s operation and financial support.
Sample Strategy - Community Dinners
In Delaware, efforts to effectively disseminate data to community members have occurred through community dinners. The community dinner model seeks to engage individuals in places within their community, such as a school or recreational meeting area, to make data and information easily accessible. Resources required to successfully implement a community dinner rely on partnerships. Often organizations contribute staff members’ time, funds to order food, and space to house the event. Community dinners are a favorite tool to gather stakeholders and community members together, and have been implemented across the state. Christiana Care Health System and the Sussex County Health Promotion Coalition have set the tone for hosting community dinners, having achieved success in discussing health-related topics with local residents.
3. Ensure community Involvement in data collection, analysis, and dissemination.
The community should help to determine what data are needed and how the findings should be used. This may require practitioners to help build the capacity of community members so they are equipped to engage in some of the more technical aspects of data collection and analysis. “Community involvement in monitoring health inequities will increase awareness, ensure health inequity data are responsive to the needs of communities, create a sense of ownership of the data, and facilitate a collaborative, equitable partnership in creating health equity policies, programs and practices” (MDH, 2014, p. 67).
Sample Strategy 1 - Photovoice
In Delaware, Christiana Care Health System employed Photovoice, which uses photography to communicate social issues, to engage Black youth in an analysis of the issues that shape their lives. As participants in this community-based participatory research project, the youth were regarded as co-researchers and assisted in developing the research question while holding autonomy in the research process. Results indicated that the youth saw violence and substance abuse/addiction as barriers to their personal success (Christiana Care Health System, 2014, p. 13). Photos representing safety, gun violence, teen pregnancy, and risky behaviors (such as gambling, tobacco use, and addiction to prescription and illicit drugs) were evidence of concerns for these youth (Christiana Care Health System, 2014, p. 13). By coupling these data with statistical reports and peer-reviewed research, the Photovoice approach provided validation of what is known in academia and represents a unique opportunity to view the social determinants of health through the lenses of those most vulnerable to their effects. The Photovoice approach exemplifies community engagement and quality data collection and analysis. |
Sample Strategy 2 - CommunityRx
Another innovative example that involved community members in data collection is the CommunityRx system in the Chicago area. With funding from the Center for Medicare and Medicaid Innovation, a group of partners began developing a system comprised of a continuously updated electronic database of community health resources that will be linked to the Electronic Health Records of local safety net providers. In real time, the system will process patient data and print out a “HealtheRx” for the patient, which includes referrals to community resources relevant to the patient’s health and social needs. To identify community resources for the database, new jobs were created for individuals residing in Chicago’s low-income communities. Many high school youth were employed to collect data on community health resources as part of the Urban Health Initiative’s MAPSCorps program. The CommunityRX project includes the creation of a new type of health worker, called Community Health Information Experts (CHIEs), who help patients use the system and engage community- based service providers in using its generated reports. For additional information, visit http://www.uchospitals.edu/news/2012/20120508-communityrx.html. |
4. Effective collection and use of data requires a skilled workforce.
This may involve recruiting new staff with research expertise, retraining existing staff, or simply supporting staff who possess the appropriate skills by providing the time, tools, and resources necessary to engage in surveillance, analysis, and dissemination of health equity data. Importantly, a workforce skilled in epidemiology is one that includes staff knowledgeable about health equity and SDOH, in addition to possessing analytical skills and research expertise. Mobilizing a skilled workforce toward an enhanced focus on qualitative methods and community-based participatory research is also warranted for a holistic description of the public health issue and potential interventions. Finally, a culture of continuous learning within state agencies and community-based organizations can support the successful use of health equity data (MDH, 2014, pp. 65- 67).
Limitations of Data Collection and Analysis
1 - Data Representation
Investments in data collection and analysis are wasted if the information is not shared in useful ways. Infographics, or images used to portray data, can be particularly effective in conveying information to the public and policymakers. Therefore, investments may be needed to allow for easy access to the data once collected (e.g. interactive websites) and to effectively communicate the findings. Figure 1 shows how the average life expectancy for babies born to mothers in New Orleans can vary by as much as 25 years across neighborhoods just a few miles apart. |
Fig 1. Metro Map: New Orleans, LA
|
2 - Data Analysis
Another data challenge, inherent in working at the community level, is the limitation of small sample sizes. Ideally, data are analyzed by neighborhood to provide the most comprehensive understanding of local needs, assets, and priorities. However, the more granular the level of data collection, the greater the challenge in reporting rates and other statistical measures, and interpreting changes over time. This is because small changes can appear large and be potentially misleading. For instance, if there are 10 cases of a disease one year and nine cases the following year, this could be interpreted as a 10 percent drop. A larger area might have 1,000 cases one year and 999 the following year, revealing a 0.1 percent drop. In both instances, there is one less person with the disease, but the reduction may or may not be relevant in the context of the population as a whole. When working with small numbers, it is difficult to know if a change is meaningful, or the result of random chance or other anomaly. Statisticians often aggregate data into larger geographic regions or over multiple years to address this challenge. However, such aggregation is less helpful when developing and evaluating place-based initiatives at the local level.
3 - Lack of Skilled Workforce
Another barrier, that is somewhat easier to overcome than others, is the lack of a skilled workforce. Surely, practitioners working in epidemiology and surveillance need strong analytical capabilities, including skills in statistics and quantitative analytics. However, it is also true that health equity work requires that practitioners be skilled in qualitative research methods. Similarly, there is a need for workers to think creatively about the kinds of data necessary to understand health inequities and describe them in ways that compel action. For example, storytelling approaches, such as Photovoice and media advocacy, are likely to leave a lasting impression on audience members.
4 - Community Engagement
Data collection, analysis, interpretation, and dissemination for health equity require meaningful community engagement and empowerment. It is often a challenge for public health practitioners and partners to dedicate the time and resources necessary to leverage and sustain community engagement. However, for data collection and analysis to impact change, the data must be easily understood and utilized by those most responsible for making change— community members, stakeholders, and policy makers. Therefore, it is in the best interest of public health practitioners and partners to engage and empower communities. By including community members, stakeholders, and policy makers in the data collection and analysis process, it is more likely that they will use the information to develop appropriate and effective interventions.
Another data challenge, inherent in working at the community level, is the limitation of small sample sizes. Ideally, data are analyzed by neighborhood to provide the most comprehensive understanding of local needs, assets, and priorities. However, the more granular the level of data collection, the greater the challenge in reporting rates and other statistical measures, and interpreting changes over time. This is because small changes can appear large and be potentially misleading. For instance, if there are 10 cases of a disease one year and nine cases the following year, this could be interpreted as a 10 percent drop. A larger area might have 1,000 cases one year and 999 the following year, revealing a 0.1 percent drop. In both instances, there is one less person with the disease, but the reduction may or may not be relevant in the context of the population as a whole. When working with small numbers, it is difficult to know if a change is meaningful, or the result of random chance or other anomaly. Statisticians often aggregate data into larger geographic regions or over multiple years to address this challenge. However, such aggregation is less helpful when developing and evaluating place-based initiatives at the local level.
3 - Lack of Skilled Workforce
Another barrier, that is somewhat easier to overcome than others, is the lack of a skilled workforce. Surely, practitioners working in epidemiology and surveillance need strong analytical capabilities, including skills in statistics and quantitative analytics. However, it is also true that health equity work requires that practitioners be skilled in qualitative research methods. Similarly, there is a need for workers to think creatively about the kinds of data necessary to understand health inequities and describe them in ways that compel action. For example, storytelling approaches, such as Photovoice and media advocacy, are likely to leave a lasting impression on audience members.
4 - Community Engagement
Data collection, analysis, interpretation, and dissemination for health equity require meaningful community engagement and empowerment. It is often a challenge for public health practitioners and partners to dedicate the time and resources necessary to leverage and sustain community engagement. However, for data collection and analysis to impact change, the data must be easily understood and utilized by those most responsible for making change— community members, stakeholders, and policy makers. Therefore, it is in the best interest of public health practitioners and partners to engage and empower communities. By including community members, stakeholders, and policy makers in the data collection and analysis process, it is more likely that they will use the information to develop appropriate and effective interventions.