The Life Course Metrics Project

FAQ

The Life Course Metrics Project

Selected Indicators and Rationale

Q. Why doesn’t the indicator set include some mortality indicators used by the MCH community- for example, infant mortality or pregnancy-related death?

A.For the purposes of this project, state teams chose to include preventable measures that contributed to mortality rather than mortality indicators. In addition, mortality rates are small leading to statistical fluctuation, which in turn makes trends difficult to interpret. States made exceptions for indicators that they felt were indicative of overall community issues (homicide rate and suicide rate) or highlighted a significant mental health issue (suicide rate).

The discussion on including infant mortality centered on whether the indicator added something to the final set that was not already captured. Although team members thought it was a broader marker for community wellbeing and equity, there was not strong support that it added more than some already included components, like preterm birth and small for gestational age. When it came to a vote, there were not sufficient votes from the state teams to support its inclusion.

For pregnancy-related death, the discussion centered on the fact that although every state can calculate a maternal mortality ratio, those with active maternal mortality reviews will likely look worse than those that do not because their case finding is better. Other points that were discussed include:

• Team members felt that the measure was not standardized;there were comments about how the National Center for Health Statisticsis not using this measure until every state is using the new death certificate
• Others noted that maternal mortality is nearly always preventable
• For the life course perspective, there is something missing from pregnancy-related death as a measure, because loss of either parent could be detrimental, not just the mother
• The number of deaths in many states was too small in number for reliable routine monitoring.

Ultimately, states voted not to include the indicator.

With regard to unintentional injury prevalence and mortality, the teamsdid not include indicators in this category. The teams took the approach of focusing on intentional injury in the form of substantiated child maltreatment, intimate partner violence, bullying, suicide and homicide. We are examining whether injury prevention indicators should be added back to the set, but as of now we have not included any new indicators related to injury.

Q.Why doesn’t the indicator set include low birthweight or prenatal care? Theseare classic measures used widely in the MCH field and literature.

A. State teams originally consideredlow birthweight, but ultimately voted not to include it in the final indicator set. Low birthweight as a measure reflects preterm births and growth restriction. The contributors and risk factors for these two outcomes vary; therefore state teams preferred to further subdivide low birthweight into preterm birth and small for gestational age.

When discussing preterm birth versus low birthweight, team members felt that an indicator centered on gestational age would be preferred, particularly in light of CoIIN efforts, the ASTHO Healthy Babies Challenge, the March of Dimes campaigns and efforts, and improved ability to assess gestational age. Team members also debated whether low birthweight was still more reliable than a gestational age-based measure, but the quality of reported gestational age has improved over time to the point that most perinatal researchers prefer to use gestational age. Others noted that preterm birth is nested within low birthweight, leading the group to consider small and large for gestational age rather than low birthweight. Moreover, small for gestational age is a more sensitive measure of economic and other stressors during pregnancy. Team members noted that from a policy perspective, small for gestational age can be hard to explain when compared to low birthweight. Ultimately, the group voted to include preterm birth and asked the AMCHP staff to explore adding a small for gestational age indicator in place of low birthweight.

While exploring the addition of small for gestational age, some methodological issues arose. The teams agreed that the ideal methodology would be to use a national standard for small for gestational age that is based on the best clinical estimate of gestational age. However, there is no existing gestational age-based national standard to calculate percentiles using clinical estimate of gestational age; the most common methodologies use a national standard that estimates gestational age using the date of last menstrual period (LMP). To keep this indicator in the final set, as state teams asked, AMCHP staff framed the issue as follows: States that would like to use this indicator should create a standard internal state-based best estimate of gestational age. People who would like to calculate this indicator outside of a state and do cross-state comparisons or generate national estimates should use the existing national LMP-based standard. Partners in national organizations are encouraged to create a new national standard for future use.

Q. The indicator set does not appear to include children with special health care needs. Why?

A. Children with special health care needs was a critical group considered by the national expert panel and the state teams. There are approximately 10 indicators that directly reflect issues facing CSHCN; these indicators include:

1. CSHCN: Percent of children 0-17 with a special health care need (HRSA-MCHB definition) (NSCH)
2. EPSDT: Percent of Medicaid-enrolled children who received at least one initial or periodic screen in past calendar year (CMS – Annual Medicaid EPSDT Participation Report)
3. Early intervention: Proportion of children ages 0-3 who received EI services (IDEA 618 Child Count)
4. Medical home: Proportion of families who report their child (age 0-17) received services in a medical home (NSCH)
5. Delay in care: Inability or delay in obtaining necessary medical or dental care (NSCH)
6. Asthma ER utilization: Proportion of persons on Medicaid with asthma having one or more ED visits for asthma in a year (Medicaid Analytical eXtract files)
7. Depression among youth: Percent of 9th-12th graders who felt sad or hopeless almost every day for more than 2 weeks during the previous 12 months (YRBSS)
8. Preterm birth: Percent of live births born <37 weeks (NVSS)
9. Small for gestational age: Proportion of singleton live-born infants whose birthweight is at or below the 10th percentile for a given gestational age (NVSS)
10. MCH data capacity: Data capacity to support integrated childhood research (TVIS)

These indicators cut across the categories, and include aspects of care delivered to children and the capacity of the health care system to meet the needs of children and families (e.g. delay in care).Depression among youth is a powerful indicator bringing mental and behavioral health into this dialogue. Additionally, the narratives pertaining to SGA and preterm birth consider the public health prevention efforts and economic impact of these births but also look at the life course implications for the family and child, from a health and development standpoint. Finally the indicator focused on MCH data capacity has a number of implications for CSHCN, as this population group is a core area of focus of linkage and surveillance activities (e.g. newborn screening, birth defects surveillance).

Finally, CSHCN is a major population subgroup recommended for specific analysis across a number oflife course indicators. Instead of identifying them separately for indicators, the group preferred to recognize their importance across multiple indicators to provide a larger focus on children with special health care needs. Also, any indicator that relies on the National Survey for Children’s Health can be analyzed by CSHCN (further subdivided by type) and compared to non-CSHCN. The Michigan state team has begun preliminary analysis of these indicators and found distressing disparities.

Q. What indicators or factors identified throughout the life course did state teams want to consider in the final set but could not due to the lack of measurement tools? Communities may wish to advocate for their measurement or consider these indicators for special applications.

A.This is an important area of study for maternal and child health and better incorporating and measuring the life course perspective. However, this was not a focus of this state-led effort. The group did not spend time to directly answer this question. To answer it well, substantial research will be required that this group did not have either the time or resources to conduct. However, AMCHP staff did maintain a list of proposed but not accepted indicators as well as a list of ‘wish-list’ indicators that did not meet the data criteria, but for which further exploration is desired. These lists are available at

Q. Which indicators can be examined specifically for American Indians/Alaska Natives?

A.Although the group didn’t focus specifically on any subpopulation, health equity was a fundamental criterion for indicator selection. Any indicator that uses vital records data can be analyzed for American Indian/Alaska Native populations. Importantly, vital records follow established guidelines for the collection of race and ethnicity. Also indicators from the American Community Survey or U.S. Census can be examined by race/ethnicity.

Q. Did you consider measuring some elements, especially around equity, using qualitative instead of quantitative data?

A.The group recognized the value and the importance of qualitative methods and discussed the need for this type of work when using these indicators for various purposes.

Presentation and Categorization of the Indicators

Q. How did you come up with the 12 topical categories for the indicators?

A.The 12 topical categories were developed during the process of selecting the final indicators. The categories were needed to organize the selection and potential use of the indicators. Related indicators were grouped to form these categories. Of note, the topical categories were phrased using language that could potentially engage new and non-traditional partners and across disciplines, to facilitate communication about life course indicators.

Q. What are capacity measures, and how are they intended to be applied?

A. The concept of capacity is based on the idea that life course concepts require a level of community and organizational resources and readiness in order to realign public health approaches through a life course lens. This reorientation may require educational efforts to ensure communities and organizations understand what it means to operationalize a life course approach. Capacity indicators examine community or organizational resources and readiness to champion life course approaches and develop cross-and multi-sectoral partnerships to reach collective impact.

There were some challenges in creating capacity indicators. For example, the data and life course criteria used to assess each indicator were not necessarily supportive of indicators for capacity because they are fundamentally different from risks, outcomes, and services. However, both the National Expert Panel and the state team members felt strongly that a set of life course indicators should attempt to assess capacity. At times the state team members struggled with defining the difference between services and capacity indicators because there was some conceptual overlap. Because of the difficulties in teasing out what belonged to each category, the final set of indicators includes both of these types of indicators in a single domain. This is an area of opportunity for more conversation around what it means to have or build capacity for a life course approach within MCH.

The matrix Michigan is using to work with life course outcomes across the life course includes a section specific to community capacity. The impact, positive or negative, from social determinants of health occurs within communities where people live. As noted community capacity depends upon factors that must be identified and developed with different strategies than those that more typically are directly related to health prevention/intervention. This is a very critical area to further define and develop metrics consistency. Addressing disparities depends very greatly on the success of mitigating the negative impact of social determinants of health on those most at risk. Changing these health determinants so they are developed to be the supports they must be to assure equity for all community members was beyond the scope of this project. It is, however, a very necessary next piece of this work.

Q. The indicator set appears focused on exposure to risk and less so on resiliency measures or community transformation. Why?

A. Although the classic life course perspective includes resiliency factors in addition to risk factors, current public health measurement practices are primarily disease, risk, and services focused. From an epidemiology perspective, tracking disease prevalence and mortality has been the prevailing approach; most standard measures in epidemiology tend to be risk and services based. Some of this stems from practicality; since most of the population “survives,” tracking survival rates does not make as much sense as mortality rates. Also, true resiliency measures are not necessarily the opposite of risk measures; identifying factors that truly support or counterbalance risks is an area of the life course approach to MCH that still needs work. Additionally, the new measures will need to be captured through our current public health surveillance mechanisms. This gap in the field is reflected in what the team members felt were meaningful gaps in the indicator set.

There are a few examples of resiliency measures in the indicator set, including fourth grade proficiency, voter registration, receipt of immunizations and preventive care, which represent supportive measures for individuals and communities. This, too, is an area that will be assisted by further understanding of how to develop and measure community capacity, which is also often described as community resilience. The ability of an individual to be resilient is often dependent upon the availability, or not, of those very resources, relationships and opportunities in their community that make it possible for an individual to adapt, adjust and make changes to move forward toward improved health and wellness.

Implementation of the Indicator Set

Q. Why don’t the indicators have companion targets or goals associated with them?

A. From the start of the project, both the National Expert Panel and state team members struggled with the purpose of a set of life course indicators. Ultimately, the participants decided that the intention of putting out a set of life course indicators was NOT to create a new set of performance measures. To underscore this point, no targets, benchmarks, or goals will be released with the final set. States and communities may decide on their targets or goals for their purposes.

To aid state-level users with assessment, program planning, monitoring and evaluation, policy development, and engaging partners, AMCHP has provided a national estimate, using the most recent data publicly available, for as many indicators as possible. In addition, more than 30 of the indicators align closely or exactly with Healthy People 2020 (HP 2020) focus areas, and users of the indicators could refer back to the HP 2020 targets if desired. Finally, each indicator narrative includes a list of potential modifiers that can be used to examine the data in different ways to determine what the “best” group looks like and could use this group for target setting.

Q. Is there a way to score these indicators or to come up with a scorecard for how counties or states measure up to the indicators?

A.Such a life course composite measure would be beneficial. None were proposed for state teams to consider nor did the teams feel the current knowledge and experience with these measures as life course measures are sufficient to develop such a measure.

Q. Are all of the indicators intended to be measured at a state level? It looks like some are being measured at a local or national level.

A. The priority in terms of considering the data availability for a life course indicator was whether the data were available at the state level for use. However, some of the indicators, particularly those centered on capacity, have a national-level measurement. One example of such a measure is the percent of states that have the capacity specified; on the state level, the indicator is yes/no in terms of whether that capacity element is present at the state level.

Some indicators from Census data are framed in terms of household or county. In the narrative for these indicators, the authors provide information about how to calculate these for the state. In some cases, a state-level estimate will really only be informative when compared to estimates generated at the community level, as in the case of the dissimilarity index that indicates the degree of residential segregation. A state-level estimate does not give an accurate picture of how segregated some communities within the state actually are, but it provides a reasonable reference point for comparison.