Examples of data improvements that could benefit asthma-related quality initiative programs include: increased availability of decision support tools to routinely and efficiently assess asthma control within electronic medical record systems; more timely and established sharing of information (e.g., asthma-related hospitalizations, emergency department visits, or missed school) across multiple points of care or data systems (e.g., hospitals, clinics, pharmacies, schools, health insurance plans); and greater availability, awareness, or acceptance of HIPAA-compliant platforms for communications between community-based providers and health care providers.

CDC has resources on asthma for health care professionals: cdc.gov/asthma/healthcare.html

And along with that Dr. Hsu mentioned... all the usual suspects regarding infrastructure needed to make that happen - messaging and terminology standards that encompass the full scope of the asthma domain, data quality evaluations, procedures for managing the volume of data from environmental sensing, analytic methods to evaluate complex multi-level and time-and-space interactions.

I also agree completely with what has been said. I really like the above posts, particularly the comments that support infrastructure enabling data sharing across sectors. I think that would help us to track outcomes over time as well as to identify patterns in utilization that could tailor care at either the patient or population levels. I would also stress the importance of a broad conception of sectors, inclusive of those social service and public agencies that could help to mitigate known asthma triggers.

I agree with the comments-particularly around data sharing across sectors and uniformity of data. I think data from other arenas such as schools could also help deepen the picture. With more and more discussions related to data and outcomes measures-I wonder if stopping and looking at factors/agencies outside of hospital and health care settings could be addressed better in what is collected. There is so much innovation in predictive analysis and use of population data that may be overlooked or used in different ways-to better understand and address disparities, especially. The American Academy of Pediatrics position statement on quality measures also talks about the importance of process measures in pediatrics. Including structure, process, and outcomes measures-from all the worlds touched by children could provide great insight into pediatric asthma and what interventions (including prevention and system-level changes) make the greatest difference.

That's an important issue. One challenge is getting permission to conduct research within schools - even for just monitoring air quality within the building some investigators report having a hard time getting permission to conduct those observations. Another challenge to any sort of interventions, is that the number of school nurses seems to be constantly reduced, particularly with cuts to school funding, and increasing number of schools rely primarily on administrative staff to do assessments and administer medications.

I am surprised that such a non-intrusive study like monitoring air quality faces this obstacle. Why?
Is anyone aware of successful attempts to implement interventions at school? What currently exists, and why? Perhaps this can help us see how to gain traction in schools.

Mostly just musing here... I don't know for sure but suspect part of the issue may be uncertainty around consent, school administrators are uncertain if they might be violating HIPAA or similar regulations by allowing such monitoring. Maybe some community lack of understanding and "big brother" fears? The PI that I spoke with was able to ultimately succeed but said it took a long time and was a lot of hassle.

On the other hand - there was an interesting article just yesterday (journals.plos.org/plosone/arti...) that highlighted asthma as one of the top "health barriers to learning" - perhaps that sort of thing can raise awareness of why this is important and help with relieving some of the misunderstandings within school communities

I agree we may not have as many school nurses as we would like. There is research that school nurses have done that shows how they can improve health and academic outcomes of students with asthma (and other chronic conditions) when they serve as coordinators of care. School nurses are a great partner for research in the schools. I think traction can be gained in schools when the approach is one of collaboration and looks not just at health outcome measures-but address academic outcome measures (and other measures the child and family want). I wanted to share two successful school-based asthma management/prevention programs. One took a comprehensive approach in MO (evaluation was done by CDC). The links wouldn't upload: ncbi.nlm.nih.gov/pmc/articles/...
ncbi.nlm.nih.gov/pmc/articles/...

Another done in Colorado (see upload).

In the case of environmental monitoring in schools there is concern about how to gauge results with any standards and what kind of mitigation would be required. In many cases there are no standard measurement methods. EPA and others advocate best practices for school maintenance to maintain indoor air quality in schools.

Just as we've found associations between place-based socioeconomic factors and population-level disparities in asthma morbidity, so too have we found associations between patient- or parent-reported socioeconomic hardships and patient-level disparities in asthma morbidity. Financial and social hardships seem to play a significant role in explaining disparities. They may also represent potentially-modifiable factors that could push us to more equitable outcomes in meaningful ways.

Asthma control is an important, modifiable predictor of asthma outcomes; asthma control is affected by numerous factors (e.g., self-management skills, medication adherence, exposure to environmental asthma triggers). In the United States, 40% of children with asthma do not have control of this disease (source: cdc.gov/asthma/asthma_stats/un...). Additional information on factors affecting asthma disparities is available at ncbi.nlm.nih.gov/pubmed/275704....

CDC and other federal agencies are aligning efforts to address racial and ethnic disparities in childhood asthma, through the Coordinated Federal Action Plan to Reduce Racial and Ethnic Asthma Disparities (released in 2012). An important focus of this Action Plan is to prioritize actions that will help ensure greater availability of evidence-based, comprehensive asthma care to ethnic and racial minority children. Additional information on this Action Plan is available at ncbi.nlm.nih.gov/pubmed/281217....

I would lean toward saying yes, both metrics of success and metrics that predict disparities should be evaluated. Asthma, and asthma management, are so complex with multiple interrelated factors at all levels across the spectrum--individual, environmental, provider, health system, policy & regulations all seem to interact dynamically over time. We have a lot of work to do, to try to study and tease out these interactions; and to develop ways to manage and analyze the complex data streams that will be required to tease these out.

Fully agree with Dr. Sward comment, but it is important to note that we are talking about real research in data science here not just downloading a program and run some analysis. Any clinical practitioner probably will agree that sequence of events is important in managing asthma, but reliable temporal reasoning methods (computable and scalable ones) in biomedical sciences are a very novel area of research that receive very little or no funding because it is sort of an orphan field; too applied for NSF and to esoteric and non-clinical for NIH.

Key metrics for measuring success in community-outreach intervention programs can include: completion rate for the intervention; assessments of knowledge, skills, and self-efficacy gained by participants; validated tests of asthma control and asthma-related quality of life over the course of the intervention; and monitoring medication adherence and (if an individual’s asthma is affected by asthma triggers) actions taken to reduce exposure to asthma triggers. Measuring asthma exacerbation frequency and severity (e.g., asthma-related emergency department visits and hospitalizations) before and after program participation is also valuable, if programs have access to these data or the ability to conduct such follow-up.

More information on proposed metrics for asthma outcomes is available at ncbi.nlm.nih.gov/pubmed/223865....

Oh that's a really nice reference, thank you Dr Hsu! I like how it's organized into core, supplemental, and emerging measures, and that it includes consideration of mediators and process measures such as adherence. The PROMIS website ncbi.nlm.nih.gov/pubmed/270217... looks like there's been progress made on their pediatric asthma measures as well. With increasing access to data sets (such as the all payer claims database), it seems plausible to be able to estimate metrics like increasing use of rescue inhalers - and medication/inhaler sensors could make that even more objective just as activity trackers and physiologic sensors could provide evidence about activity restriction (same constructs as what's listed, just potentially different way to measure, right?). I don't see a ton of subjective/qualitative items listed, like satisfaction or perceptions about asthma control, parent/caregiver emotional burden, confidence managing asthma, but those sort of metrics may be worthwhile to consider as well.

See my comments above much better data system are needed. PRISMS is a good satrt, but much more work remains.

Thank you! I concur and would add a set of measures that care coordination and health data exchange across the community, including the person at the center. For example, frequency of exchange of the Asthma Action Plan; % of patients who have an updated Asthma Action Plan; % of patients who are able to share/exchange their Asthma action plan (e.g. children/parents and school nurse)

One community intervention is monitoring. Our clinical colleagues developed an electronic symptom diary (they call it eAsthma Tracker) that includes logic to detect trends in asthma control, before those symptoms get to the point of needing urgent care or hospitalization. If the system detects a pattern of low or worsening Asthma Control Test scores, the parent (or participant) gets a text message or email (the participants choose how they want notifications) notifying them of the possible trend, and recommending they consult with the care team. For some cohorts they have linked this so that a clinical case manager also gets a copy of the notification and can follow up with the families. They are in the process of writing up and publishing, but in short - they found statistically significant reductions in ED and urgent care visits.

And along similar lines of monitoring as a community intervention - one of the things we are in the process of, in the PRISMS program, is that we have put particulate (PM) sensors inside people's homes. Some of the families have access to view the data from those sensors, and they noticed patterns in their own behaviors. One family, for example, started vacuuming earlier in the day when the kids are in school instead of right before they come home, because the parents noticed how long it takes for the PM count to return to baseline after cleaning--this was a simple change in habits that, for this family, reduced the need for asthma rescue meds. One possible application for personal/home air quality monitoring might be to help families learn what are their specific triggers, and what simple lifestyle changes they can make to help those.

It seems like in the hospital we are focused mostly on controlling the acute aspects - and active interventions by clincians (meds, etc.). In the community it's more about monitoring, control, and more subtle nuanced symptoms; and interventions are more behavioral and self/family initiated. I would ask the clinicians on this panel - is that a fair assumption?

I think that's a fair characterization. That said, I think we are feeling increasingly challenged to consider both acute and chronic across settings. In Cincinnati, this involved a real frame shift. Historically, a hospitalization involved treatment of the acute exacerbation with less attention paid to triggers and chronic needs. Over the last few years, we've tried to bolster our focus on the chronic, providing meds-in-hand (both rescue and controller if indicated), environmental screening, connections to home health, social work, and the city housing enforcement agency, etc. These changes were occurring alongside enhanced chronic management in our primary care and sub-specialty clinics as well as through broader partnership with schools and other community agencies. As a result, we've seen a reduction in the asthma hospitalization rate among Medicaid-insured children across our county. I think the more we can integrate community and hospital (or clinical), the more likely we are at achieving success.

This recent paper in JAMA Pediatrics looks at a school based asthma prevention program that heavily utilized telemedicine to reduce ER visits. jamanetwork.com/journals/jamap...

Examples of effective community asthma interventions and hospital-driven interventions exist. Some differences include: (1) Hospital-driven interventions are often more likely to have access to data to focus the intervention and to assess effectiveness; (2) Community interventions might benefit from inherent trust and buy-in from the community, which could result in more program participation; (3) Staff background and skillsets can differ between hospital-driven interventions and community interventions (e.g., licensed health care professionals vs. trained community health workers or other lay educators); (4) Funding mechanisms for these interventions (e.g., billing health insurance for services vs. grant funding) can also differ, resulting in varying sustainability or stability.

Given these examples of how community interventions and hospital-driven interventions might have distinct strengths, hospital-community partnerships could be synergistic by harnessing complementary assets.

CDC has additional information on multisector partnerships to improve asthma control: cdc.gov/sixeighteen;
cdc.gov/sixeighteen/docs/6-18-...
cdc.gov/asthma/public_health.h...

A problem with community based interventions is that they are very difficult to rigorously evaluate because data from these interventions tend to be less well structured and consistent. Developing better data collection systems to critically evaluate these intervention ois in the critical path to provide better and less costly medicine.

Thanks for highlighting this study! As a quick summary, we found pretty wide disparities across neighborhoods within our county, differences in asthma admission rates that tended to follow socioeconomic lines. Although not surprising, it provided us with a lens through which to view our population health efforts. Neighborhood has become a central means through which we now engage with families and community agencies and collectively seek improvement.

Are there trends in different parts of the country (or world) that see an opposite or unexpected trend? While it seems intuitive to see asthma rates to be correlated with socioeconomic lines and neighborhoods, I wonder if this relationship is less tightly coupled in certain areas, and why it might differ between areas.

Most of the data that I've seen tends to run in this same direction. That said, I'd be very interested in learning from others who have seen different trends. Similarly, I'd be very interested in learning about those "off-tangent" or "positive outlier" neighborhoods or areas. That is, are there neighborhoods who have low asthma morbidity rates while simultaneously having socioeconomic risks? We have some evidence of this in Cincinnati which we are just now trying to explore - no conclusions yet, but such exploration may inform how we, collectively, approach place-based asthma morbidity.

I think our local asthma researchers have seen similar trends to what was reported. That positive outlier idea is really intriguing though.

Here in the Salt Lake valley, at least, there are distinct "micro climates" caused by geography + how the city had developed. You can get very different weather patterns and there's evidence these have correspondingly distinct air pollution patterns. I wonder if some of the "neighborhood effect" might correlate with those "micro climates" - how much that confounds looking at other social determinants in those neighborhoods.

I am interested to see the outcome of your study, Dr. Beck.

Dr. Sward's musing is very interesting and forces us to look at additional variables. While we may have been able to identify climates distinct between cities, looking at differences locally could prove to be a challenge. I am not sure how to define these variables. Have you thought about how they could be defined?Being able to understand this may become more important as urban areas grow and the "neighborhoods" become more complex.

It is an interesting questions. Having spoken to some of our environmental epidemiologists, we do not have quite the differences in geography that Salt Lake does. That said, we do live in a river valley which can trap allergens and pollutants. We also know that differential exposures to certain types of pollution, eg traffic-related, can have an affect on asthma morbidity. In Cincinnati, it seems that many of the neighborhoods have some element of double (or more) jeopardy - high rates of poverty as well as high rates of traffic (or rail) related pollution. It can be hard to disentangle as I think the evidence suggests that both are relevant and important.

According to the most recent national data on asthma from 2015, current asthma prevalence in children (aged <18 years) was higher among Puerto Rican Hispanics (13.9%) and non-Hispanic blacks (13.4%), than among non-Hispanic whites (7.4%). Children living in poverty also have a higher asthma prevalence. Furthermore, black children have a higher risk of mortality and a higher rate of emergency department visits for asthma than white or Hispanic children. Sources: cdc.gov/asthma/most_recent_dat... and ncbi.nlm.nih.gov/pubmed/281217....

You will be surprise on how much the environmental exposures are dependent of locality. Off course at the very high and very low exposures this does not matter, but in the intermediate regime is very challenging (cumbersome and costly) to accurate measure (or estimate) individual level exposures with high spatiotemporal resolution.

I am a nurse, and a biomedical informatics researcher - and a parent/grandparent of kids with asthma. One of the PIs for an NIH project developing an infrastructure for research on pediatric asthma, in which sensors are used for the research (PRISMS). Mobile health, sensors, and similar technologies, show a lot of promise in being able to contribute to diseases like pediatric asthma that are affected by the environment (including physical environment, as well as lifestyle, behaviors, and social determinants of health). One of the challenges is that these new technologies are emerging and evolving so rapidly, and there is no consensus or standard on how the data should be structured or communicated. So how do we converge on a way to support research rigor and reproducibility in this very fluid, flexible situation? Another challenge is questions around security and privacy - how do we protect and "deidentify" data that are intrinsically linked to space and time?

Introduction. I am a primary care pediatrician and my focus is on implementing community-based asthma interventions to reduce asthma disparities. Currently working on an NHLBI project to provide evidence-based interventions that connect the home, clinic, school and community for low-income minority children diagnosed with asthma. I also co-lead the Asthma Population Health Workgroup at CHOP. We are implementing a bundle approach for high risk children diagnosed with asthma.

Hello Dr. Sward, thanks for sharing! These are interesting questions. The challenge regarding security and privacy is difficult. This issue spans across all healthcare data and inhibits data sharing. I am curious if anyone else has an idea of how to address this problem.
You refer to there being no standard surrounding data structure and communication. What is an example of how data is currently structured that deals with pediatric asthma?

Thank you, Dr. Bryant-Stephens. I think you will have excellent insight into our discussion, and I look forward to hearing about your experiences with the NHLBI project and CHOP.
I am interested to know- what does the bundle approach for high risk children entail?

I am a informacion scientist interested in IT infrastructure to support research in complex medical problems. The challenge here is "how to build responsive modular architectures that are really customizable for different research designs". They key is to provide the relevant data at the right time at the right cost. This is the only way to develop a body of quantitative research that can provide objective and authoritative assessment of different interventional strategies for asthma control. But this is not easy... It will require a great deal of team work among medical, environmental, and data scientists.

Hi Antonia - what I was talking about in terms of lack of structure is particularly within the sensors and related tech realm. Take for example something as seemingly common as particulate matter levels - EPA reports PM2.5 and PM10, so there is a little bit of consensus around that - but once you get to looking at data streams from sensors you find a lot of heterogeneity, both in how the label or name the data streams, but also in how they are measured (raw counts, mass per unit of air, etc.). Some sensors will crudely estimate "large" vs "small" particles, which we assume correspond roughly to PM2.5 and PM10, but may not actually match those. And then we get into the turf of "ultrafine" particles etc. - where are cutoffs for those, how are they measured and reported... And that's just for the things that are relatively common to look at! On the plus side I see a LOT of work in this area in the past year or two, so I'm optimistic.

Hi everyone! My name is Andy Beck, and I'm a pediatrician at Cincinnati Children's Hospital. I see patients in our primary care center as well as on the inpatient units. I have long been interested in public health, health disparities, the social determinants of health, and community engagement. I got particularly interested in asthma when, as a resident, I was involved in the care of an acutely ill child. She presented to our trauma bay in extremis and subsequently died from her asthma exacerbation. She lived down the block from our hospital. It highlighted to me how much distance we must still cover to collaboratively meet the needs of our patients and populations. Currently, my work focuses on the intersection of place and health. In particular, I work on our hospital's Community Health Initiative which, in part, aims to narrow disparities in the days children spend hospitalized across our neighborhoods (and across conditions). Asthma continues to be among our top drivers of inpatient bed-days (and disparities in inpatient bed-days).

Hello! I am Sarah Brundidge, the Director of Lung Health for the American Lung Association, Legislative Co-Chair for the WI Society for Respiratory Care, I continue to practice as a RT in my "free time", and I am an AE-C. A large part of my role with the ALA is to facilitate and lead asthma quality improvement initiatives for hospitals and clinics throughout the Midwest. I consistently see a lack of time spent on patient education and goal setting, and a lack of correct, consistent messages to the patient. When I first begin working with a hospital system I have to do provider education on the NHLBI (EPR-3) Guidelines. Oftentimes the healthcare team is unfamiliar with them, don't know how to use the severity, control, and treatment charts, and feel that they take way their ability to be "creative" practitioners.
I also do environmental home assessments for patients with asthma and COPD. We are able to provide trigger remediation supplies and self-management education for the families.

The data that may be relevant for epidemiological as well as interventional studies for pediatric asthma is quite diverse, including socio-economic, environmental and clinical data. Except for studies in very well defined cohorts that collect very specif data for a particular study (a costly approach), most of the studies attempt to make secondary use of exiting data. Here is the problem because data may have very different resolution, semantics, and quality its aggregation is costly and time consuming. In contrast with say baking data, which has very simple semantics, biomedical data has extremely complex semantics and state of the art computer science methods are not sufficient for automatic harmonization.

I am a Pediatician and Medical Director of American Well, the largest video-based telemedicine company in the US. I’m always looking for ways that new technologies can tackle the triple aim of increasing quality, decreasing cost, and increasing access.

I am the Director of Research for the National Association of School Nurses. Asthma is one of the most common chronic conditions of students in schools. I am late to the discussion-but happy to join! In school nursing we address pediatric asthma in many ways-addressing the disparities we see, as well as addressing the social determinants that play a large role in the management of asthma on a day to day basis. I agree with the comments made about linking and coordinating efforts -not just from hospital and community but from schools, housing, and any other aspect of children's lives that impact or could influence the management of asthma.

I agree that there are no standards surrounding data structure and communication-but great potential. School nurses are beginning to collect uniform data that could be very useful in better understanding pediatric asthma-even in better understanding prevalence, disparities and pre-disposing factors. Approximately 98% of children attend schools and although we do not have school nurses in all schools as we would like, school nurses do have a pulse and understanding of pediatrics health that is not always utilized. There have been some great work of where hospitals and schools are able to share records electronically related to children's health-including asthma action plans. As technology continues to evolve I see even more possibilities so that all involved in the child's care (child, family, healthcare provider, school nurse, etc.) can work from a shared document and plan of care. Exciting times!

Hope that you are correct, but drift of data standards and procedural compatibility are very difficult challenges when you are looking for relatively low signals.