Following the first two panel sessions, participants divided into two breakout groups for more in-depth discussion of data from several case examples of transportation projects. After reconvening in plenary session, breakout facilitators Michelle Proser, the director of research at the National Association of Community Health Centers, and Art Guzzetti of the American Public Transit Association, briefly described the presentations and discussions that took place in their groups. (Highlights are presented in Box 5-1.)
One of the breakout groups heard presentations of data from two projects in the Washington, DC, area. The first presentation, Linking Transportation, Health, and the Built Environment in Washington, DC, was given by Anneta Arno, the director of the Office of Health Equity in the Office of the Director of the District of Columbia Department of Health; Raka Choudhury, the citywide transportation planner at the Progressive Transportation Services Administration at the District (of Columbia) Department of Transportation (DDOT); and Steve Strauss, the deputy associate director of the Progressive Transportation Services Administration at DDOT. The second presentation, Efficient and Effective Provision of Mobility to Health Care, was given by Steve Yaffe, the transit services manager for Arlington County, Virginia.
Proser reported that, in addition to hearing about the different models and innovations, much of the time in his group was spent discussing the importance of data for knowing who the patients/riders are and for demonstrating the value of transportation to health care outcomes. Participants discussed how best to demonstrate the value in interagency collaboration, with an emphasis on demonstrating the value together. The intent is not transformation in just one sector, but transformation of communities to achieve increased vitality and efficiencies.
Some of the data challenges discussed included the general lack of data sharing across sectors and various barriers to data sharing, such as restrictions on sharing personal health information under the Health Insurance Portability and Accountability Act (HIPAA). It was also noted that data collection is inconsistent across different hospital systems, even within the same community. Although all not-for-profit hospital systems are conduct-
ing community needs assessments, they are not using the same measures or benchmarking the same issues and problems, making it harder to have cross-community collaboration on the findings.
The need to have data to better understand who the consumers are was also raised, Proser reported. Can data be used to better segment patients with transportation needs? For example, it was suggested that the increase in the number of patients who are using public transportation programs could be due to the increase in chronic disease.
To address the HIPAA barriers, there have been attempts to create central hubs for data sharing, collaboration, and coordination of services. Several participants emphasized the need for better data on the health care side as well as the need for greater consistency across the community health needs assessments. It was also observed that there are many different efforts to provide transportation for different groups (i.e., different disease categories), and many participants discussed the need to identify areas of intersection and reduce overlaps in order to create more efficiency.
Participants in the breakout group discussed costs, Proser reported, and also the importance of being good stewards of resources. Several participants also highlighted the need to reframe how the case is presented in order to effectively demonstrate the return on investment of getting patients to care. It is a challenging conversation, Proser acknowledged, because the utilization of primary preventive and chronic care is increasing, but there are long-term benefits, such as lower utilization of acute care and an increased vitality of communities. There is a cost associated with improved outcomes in the short term, but there are potential savings in the long term.
Population health improvement, a better quality of life, and improved well-being are all measures of success. It was noted, however, that social interactions and general well-being are not captured in clinical care. Cost savings were also discussed as a measure of success, especially cost savings that result from maximizing efficiencies through prevention. As more and more patients are dealing with preventable chronic illness (e.g., end-stage renal disease), providers of both health care and transportation are now seeing how the social barriers to health affect outcomes and costs.
In closing, Proser reiterated the power of data for demonstrating value and the need to work collaboratively across sectors. Data is advocacy, she said, and can help us to understand specific community needs and identify where to reinvest any savings generated upstream.
The other breakout group heard presentations of data from two projects in the Boston, Massachusetts, area. The first presentation, Smart Transit for Health Care, was given by Moumita Dasgupta, a principal investigator in the Department of Physics and Astronomy at Amherst College, and Sarah Anderson, the business development manager at Cambridge Systematics, Inc. The second presentation, Comparing Scheduled and Actual Transit Accessibility to Health Centers, was given by Anson Stewart, a Ph.D. candidate in the Interdepartmental Transportation Program at the Massachusetts Institute of Technology. Two main points emerged from the presentations, Guzzetti reported: the need to optimize appointment trip scheduling, and the importance of reliability.
The first presentation described how geocoded addresses revealed that low-income populations are lacking health care access relative to the rest of the population. The proposed solution was to schedule the necessary transportation at the same time that the patient calls to schedule the health appointment, so that both needs are addressed at once. It was noted that health centers have expressed a willingness to participate in this approach if it is simple for them (e.g., through using smart software). Several participants discussed the need to resolve questions and address challenges in advance. Issues with billing, eligibility rules, cost, time, and expertise are all solvable, Guzzetti said, but might require policy reforms.
The second presentation considered the issue of reliability. The research that was presented had found that two-thirds of bus riders reported arriving late for or missing an appointment at least once within the preceding year. Stewart explained that the research hypothesis was that there would be a positive correlation between unreliability and missed appointments. The expectation was that more people would be missing appointments at health care centers, for which general transit accessibility is often less reliable. The preliminary findings, however, were the opposite. Health care centers, which by a certain measure have more unreliable transit, actually had fewer people reporting missed appointments. One preliminary theory to explain that observation is that those health care clinics are building in “reliability buffer time” because they expect the trips are going to be unreliable. There may actually be wasted reliability buffer time built into every trip, Steward said, and he suggested that this would be important to quantify.
Another topic of discussion, Guzzetti reported, was the relative reliability of other transportation options (e.g., highways). Several participants in the breakout group also considered what the social response to this unreliability might be—for example, improving the capacity and reliability of transit.
In response to a question from a webcast participant, Ysela Llort, planning committee chair, prompted participants for comments on the engagement of payers in transportation issues. Meadows responded that, despite recent progress, there is still a disconnect between transportation and payers, especially relative to Medicaid. He said that Geisinger is looking at developing a concierge service, so that if a patient is turned down by the Medical Assistance Transportation Program in Pennsylvania (discussed in Chapter 3), that patient can call the concierge service and arrange a ride to his or her appointment. Meadows also mentioned recent interest from a provider in making a change to its bundled payment methodology, so that the cost of the transportation would be covered as part of the bundle. This is a step in the right direction as well, he said.
Stacy Elmer from Kaiser Permanente in Southern California said that Kaiser often considers transportation to be either medically necessary or something that is a covered benefit. She added that Kaiser in Southern California is working to better meet the needs of its members in terms of transportation and to better understand why patients are missing appointments.
Flora Castillo, the vice president of community and strategic engagement at UnitedHealthcare and a volunteer board member of New Jersey Transit, said that UnitedHealthcare is working to create innovative partnerships with the transportation sector to ensure that United is addressing the needs of members. Castillo referred participants to a recently released white paper by UnitedHealthcare in which a framework is proposed for measuring the quality of care delivered in a managed, long-term services and supports environment.1 Within those measures are several questions on transportation, she added. Castillo said that UnitedHealthcare is interested in conducting more pilot programs, and she encouraged participants to discuss ideas with her.
1 Available at https://www.uhccommunityandstate.com/content/dam/community-state/PDFs/NAB_LTSS_Whitepaper.pdf (accessed August 4, 2016).
A number of workshop participants discussed further the potential of software for trip and appointment planning, and it was suggested that one function of such software could be to immediately contact a patient if he or she does not show up for an appointment in order to find out what the issue is and to assist. Another function of the software could be to find transportation options for getting back home for patients who do make it to their appointments. A participant noted the potential of application programming interfaces and open-source systems using standardized data. An advantage of this collaborative approach is that an improvement made anywhere is shared with the community using that software.