Application of the Toyota Management System across the Group Health Cooperative System Webcast (2/26/08)

I come from a leadership heritage that says, “help others be successful,” so I jumped at the chance to bring two great business partners (Karl Hoover, Executive Director of Quality&Informatics, and Lee Fried, Senior LEAN Consultant) together to conduct a national event on Group Health Cooperative’s work implementing the Toyota Management System across the enterprise. It was fun, and Lee and Karl are always great to listen to (as well to work with).

I attached two slides to the right which are around my contribution to the work, which dramatically changed the way we develop and maintain a statewide electronic health record with linked personal health record. The work was a lot of fun and very challenging at times (which is normal for a LEAN transformation, many peaks and valleys). I like challenges like this in what you learn about yourself in the process. One thing I learned about myself is that I am not afraid to work with people who are smarter than me (Karl and Lee as an example). It’s an attribute I have always looked for in other leaders, and I am happy I developed it, too.

Enjoy.

Note: This post was previously published, but is being reposted today to tie together the special report, which is now published on this site in its entirety. To see all of the parts together, click on this link.

Welcome

Tyriece, Lisa, and Ted Eytan, MD, Belair-Edison Clinic, Baltimore Medical System

This is a compendium of personas in our health care delivery system. It’s designed to inform individuals and organizations interested in increasing patient and family involvement in all aspects of care, and has been specifically commissioned by the funders of PCHIT.

At the current time, it is estimated that between 3 and 10 % of Americans have access to their care teams through personal health records (PHR). Also at the current time, there are a host of organizations and individuals working to increase this percentage.

Besides being a compendium of organizational personas, Profiles is also an interim status report on the PCHIT initiative. As such, it has a PDCA format.

Plan

PCHIT starts at the level of the patient-physician relationship and moves outward

A way to visually consider where PCHIT is focusing its efforts is through adaptation of the “Sequence of HIT Adoption” model proposed by the Robert Wood Johnson Foundation (see Electronic Health Records Still Not Routine Part of Medial Practice, Robert Wood Johnson Foundation).

Our plan was to spend time in a cross section of health care, observing the patient-physician interaction. PCHIT was designed to look at factors present in different care environments at the level of the patient-physician interaction. This is based on the concept of Genchi-Genbutsu. This concept is so central to the design of this initiative that we are including an explanation of it here:

First of all, gen means actuality or reality. When we look at the word gen-ba, it means the actual place. In the terms of manufacturing, we can loosely translate this to mean where the work is done. Why is this meaningful? It is not until we understand the other gen words that this begins to make sense.

Second is gen-butsu. Butsu means, the condition of the thing. In terms of manufacturing and considering the word gen-ba, we ask ourselves, “what are the conditions of things in the workplace, where the work is actually done?” The things we are looking for? The condition of the design, the quality, the process, the people, the methods, the equipment, etc. When we think of genba and genbutsu, we are looking to see if the conditions of our standards are deviating in the workplace. This forms the basis for standardization of all aspects in the business.

Third is gen-jitsu. The actual situation. We are looking for facts so that we may understand the gap between reality and standard. We are not looking for what it should be, we know that. We are looking for actual situations, or the facts. This helps us begin to dig for the actual root cause.

If we only consider the standards we tend to sit in a meeting room wondering why the equipment, the people, the materials and processes don’t meet standards. The only way to truly know, is to go to the actual workplace, observe the actual conditions and collect the facts. This leads to true understanding of reality. Otherwise our solutions we invent in the meeting room are for problems that are not really happening in the workplace. This is the reason why problem solving begins with the saying, “go and see for yourself, in the workplace where the work is actually happening.”

We weighted our interest toward organizations with emerging health information technology initiatives, as opposed to mature, or no health information technology initiatives.

We weighted our interest toward organizations serving diverse populations, including under-insured, safety-net, and minority populations (inclusive of federal protected classes as well as gay, lesbian, bisexual, and transgender individuals).

We weighted our interest toward organizations situated geographically in areas served by our sponsors, including New York, California, and District of Columbia.

Kaiser Permanente, a sponsor of this effort, is used here as a benchmark, given its maturity in patient centered health informaiton technology. The same is true for Group Health Cooperative.

Do

Over the next several days, we’ll publish our “Persona” description for key stakeholders in the implementation of patient centered health information technology.

We chose to use the persona concept, established in user interface enginering - you can read a little bit about it here.

Comments are Welcome

Your additional experience, expressed by adding comments on each page of interest, are welcomed

In many, if not all, of the sites we visited, the question of disparate access to PCHIT was raised. The same question has been raised with regard to EHR’s as well. In its report, the Expert Consensus Panel (see Electronic Health Records Still Not Routine Part of Medial Practice, Robert Wood Johnson Foundation, 3:27):

(The Expert Consensus Panel) has identified racial and ethnic minority patients and low-income or publicly insured patients as the two highest priority patient populations

The PCHIT Initiative broadens this view of vulnerable populations to include those with documented disparities including but not limited to individuals who are lesbian, gay, bisexual, and transgender. An additional vulnerable population of interest are returning soldiers (see: Longitudinal Assessment of Mental Health Problems Among Active and Reserve Component Soldiers Returning From the Iraq War).

Available data about Internet access contradicts conventional wisdom

Charts: Click on any to see full size (Sources: Benchmarking Digital Inclusion, ITIF, and Estabrook L, Witt E, Rainie L. Information Searches that solve problems. Washington, DC: Pew Internet & American Life Project; 2007)

In a review of the literature related to Internet use among vulnerable populations, we discovered that commonly held beliefs about use and access are not true. Even at the lowest educational and income levels, Internet use approaches 60 %, where it was only 10-30 % in 2001.

The following studies shed additional light on this issue:

• Estabrook L, Witt E, Rainie L. Information Searches that solve problems. Washington, DC: Pew Internet & American Life Project; 2007. A random sample survey performed between June and September, 2007 shows gains in Internet access and explores type of access - “High” for broadband at home or work, “Low” for no access or Dial-up only
• McNeill HL, Puleo E, Bennett GG, Emmons MK. Exploring Social Contextual Correlates of Computer Ownership and Frequency of Use Among Urban, Low-Income, Public Housing Adult Residents. J Med Internet Res 2007;9:e35. - In Boston public housing, 51 % of adults own and 42 % of adults regularly use computers, and those reporting feeling “safe” in their neighborhoods have a greater likelihood (76 % higher association) of using a computer regularly.
• Beckjord BE, Finney Rutten JL, Squiers L, et al. Use of the Internet to Communicate with Health Care Providers in the United States: Estimates from the 2003 and 2005 Health Information National Trends Surveys (HINTS). J Med Internet Res 2007;9:e20 - “In 2003, Internet users with high levels of education were more likely to have communicated online with an health care provider, consistent with previous studies. That education was nonsignificant in 2005 may suggest that health care consumers’ level of education is less of a barrier to communicating online with health care providers as the prevalence of online patient-provider communication increases.” and “We did not observe associations between online communication with health care providers and characteristics such as race/ethnicity or annual income that have been documented in other studies as evidence of a “digital divide””
• Carroll AE, Zimmerman FJ, Rivara FP, Ebel BE, Christakis DA. Perceptions about computers and the internet in a pediatric clinic population. Ambul Pediatr 2005;5:122-6. - “Most families like using computers and feel comfortable using the Internet regardless of socioeconomic status. Fears about the digital divide’s impact on the attitudes of parents toward computers or their comfort using the Internet should not be seen as a barrier to developing Internet-based health interventions for a pediatric clinic population.”

A more sensitive indicator of patient access to electronic health records is likely to be online banking (see this post on that topic), because online banking requires confidence and convenience as well as access to be successful.

Online banking use and income level, from Online Shopping, Pew Internet & American Life Project, 2008

East Boston Community

Patient-centered HIT applications do not necessarily require use of a computer on the consumer’s end. For example, a mobile phone may be the most effective vehicle for certain populations, whether the information coming to them is in the form of an automated phone call (which can be delivered in multiple languages), a text message (such as for medication reminders), or a more sophisticated combination of audio, graphics and video. A variety of strategies are profiled in a recent report published by the Georgetown Health Policy Institute’s Center for Children and Families (see Health Information Technology: Innovative Applications for Medicaid).

Outside of patient access to computers or the Internet, there are opportunities

Some analysts shortchange vulnerable populations by suggesting that language barriers, the digital divide, or health literacy pose insurmountable obstacles to effective PHR adoption. Perhaps no population faces a greater panoply of barriers–including Spanish as primary language, health literacy, access to computers and the Internet, geographic challenges, and a lack of care continuity–than migrant farm workers. The tool, MiVia, has demonstrated that PHRs can be effective tools when appropriate accommodations are made, such as using community health workers to help facilitate PHR adoption.

As we consider patient-centered health information technology, the definition should be broadened beyond personal health records, to any technology that provides the benefits and impacts of patient access. These impacts accrue whenever the health system is accountable to those it serves, by providing them the information they generate about them, whether in paper, computer or smart card form.

Unresolved issues

• It is unclear how pervasive the conventional wisdom of the “digital divide” is, and if there are related factors that would bias toward inaction even if the data were better understood for populations studied (ethnicity, income, education)
• For populations that are less well studied (e.g. lesbian, gay, bisexual, transgender, returning soldiers), the impact of provision of access to PCHIT in safety net environments is also unknown. With limited funding available to study sexual minority populations, for example, disparities may only be exacerbated in an environment of HIT without PCHIT.

Countermeasures

In 2008, we are emphasizing safety net providers and vulnerable populations in PCHIT work. We are providing the technical assistance of a knowledgeable medical informaticist and patient empowerment advocate to demonstrate the impact of PCHIT in a vulnerable population. We would also like to spend some effort in packaging this data and presenting it in leadership forums. Ted Eytan did this recently for the District of Columbia Primary Care Association, where it was well received (see Presentation to DCPCA, December 18, 2007), as well as on a recent event at Urban Health Plan, in Bronx, New York (see: “We did it! Thanks Affinity Health Plan and Urban Health Plan!“)

Ways to Engage

In addition to working with health care and IT leadership on promoting PCHIT as part of HIT, it would be valuable to engage with patients themselves. In 2008, we are hoping to shadow a patient who is part of a vulnerable population as they manage chronic disease. This will most likely happen on our trip to Sonoma, California, in March, 2008.

Sites visited

Union Square Family Health Center, Somerville, MA

• Institute for Family Health (New York)
• East Boston Neighborhood Health Center (Massachusetts)
• Cambridge Health Alliance (Massachusetts)
• La Clinica de La Raza (California)
• District of Columbia Primary Care Association (District of Columbia)
• Lifelong Medical Care (California)
• Queens Health Network (New York)
• Unite HERE! (New York)
• Urban Health Plan (New York)
• Baltimore Medical System (Maryland)

An Emphasis of PCHIT

As others studying health information technology adoption have pointed out, the populations served by these providers are of concern, and are being emphasized in this initiative. There is a focused description of the populations we are interested in and disparities in this report (see Persona: Vulnerable population).

There are multiple ways of identifying safety net providers (see Electronic Health Records Still Not Routine Part of Medial Practice, Robert Wood Johnson Foundation). Our work focused primarily on community health centers, which are estimated to serve about one in eight uninsured patients. Within CHC’s themselves, about 75 percent of patients are uninsured or on Medicaid.

EHR Adoption Among U.S. Physicians and Health Centers, from Health Information Technology in the United States: The Information Base for Progress: Robert Wood Johnson Foundation; 2006

In terms of adoption of EHRs, there is comparability to physicians in general, with a trend toward lower adoption for physicians with a higher percentage of medicaid patients. There is no comparable data for PHR implementation, unfortunately.

Resilient and Creative

Each safety net provider we visited, sometimes on the same day we visited a non-safety net provider, showed a different light onto health care. This was especially true for a physician accustomed to leading in a multispecialty group with commercial contracts (Ted).

The expectation might have been extreme resource constraint without ability to innovate in health information technology. However, we found organizations that are as technologically savvvy as their non-safety net peers, in an environment where 92% of community health centers nationally do not have electronic health records. This group has access to novel ways of financing improvement (in particular, a limited supply of grants, depending on the region, see Remarks to the National Association of Community Health Centers, HRSA Administrator Elizabeth M. Duke) and many that we visited have state of the art electronic health record systems.

However, few of the organizations above have implemented a personal health record yet. Cambridge Health Alliance was in the process of implementing theirs when I visited in November and has begun a pilot since. Institute for Family Health is about to begin its pilot in New York. At the same time, in conversation with providers and in some cases, patients at these locations, there was a general welcoming of the concept of patient access to the electronic health record, even in a multi-lingual care environment.

Dr.’s Davis and Isles, Belair-Edison Clinic, Baltimore Medical System

We learned about the financing model of safety-net providers that makes them ideal care providers in the communities they serve. Specifically, rates of reimbursement for in-person visits may be higher for federally qualified health centers. This environment is changing though, based on health reform efforts underway in states like Massachusetts.

At the same time, current models focus care provision on a physical visit. At La Clinica de La Raza, for example, I attended a celebration of the achievement of a milestone in number of encounters in a particular month. The celebration was a testament to the hard work and will of staff to support the organization’s viability. At the same time, there will be challenges to introduce non-visit based care if the unit of reimbursement is the in person visit.

“Uninsured” does not equal “Uninformed”

An impressive and reassuring finding in observations was that patients receiving care in these environments are getting informed about their care via various means, including the Internet, and are open to connecting with their providers this way. As a patient at Berkeley Primary Care told me, “I want my doctors to meet me half way.” Seeking care among multiple providers in the community with disjointed communication between them is perceptible information gaps for patients who have reduced abilities to tolerate fragmented care.

These information gaps are being addressed by consulting with others in the community, or by becoming disempowered in the care relationship. We saw examples of both, which were displeasing to both physician and patient. In these situations, patients may not be discussing these feelings with their providers in the exam room unless asked, which I also observed.

A recent study measured oncologists’ recognition of empathetic opportunities and found response to these to be low (22 percent with “continuer” statements) (see Pollak KI, Arnold RM, Jeffreys AS, et al. Oncologist Communication About Emotion During Visits With Patients With Advanced Cancer. J Clin Oncol 2007;25:5748-52). This finding, among a group of terminally ill patients, may be as relevant for members of vulnerable populations who provide empathetic opportunities to their care system to “meet them half way” through improved interaction and information about their care.

EHR deployments are without PHR deployments

A concerning trend we noticed was the assumption that an EHR deployment in this context should not include patient access or patient-centric health information tool deployment.

More than one individual in safety net environments expressed the following sentiment in our travels: “PHRs won’t work for this population, because of inaccessibility to computers/the Internet.” However, the data behind this assessment was hard to come by. The impact of statements like this, made in some cases from vendors of EHRs supporting these organizations, is that PHR deployment is not included in implementation plans. This is the case even when it is in other organizations’ rollout for the same product.

Our concern is that this is a significant missed opportunity and may result in the hastening of an exacerbation of differential HIT adoption and ultimately health disparities.

Observations and discussion with support staff again showed that there is more potential internet use and uptake than commonly believed. At a recent discussion hosted by the District of Columbia Primary Care Association, one clinic administrator said, “Whenever I walk into the waiting room, there is always someone using the computer (referring to a community-wide program to make computers available in local clinics),” and “we’ll never know if people will use this if we don’t set it up.”

No EHR deployments are planned in some, impact on physician recruitment?

Several safety net providers we visited have no EHR deployments planned at all, and I (Ted) witness varying degrees of discomfort with this situation, based on previous use of EHRs by staff physicians. Those that had direct use of EHRs in their past appeared more eager to adopt the technology. Bina Patel, MD, at La Clinic de La Raza lamented that when she chose to move to California to practice in a CHC, she interviewed at 7 different organizations only to find that she would have to practice on paper at each of them. A situation like this has the potential to impact future recruitment of young physicians to these environments. There are California CHCs that are implementing EHRs, such as Redwood Community Health Coalition (see Network of Community Health Centers Utilizes Electronic Medical Records System, Patient Portal and Electronic Health eXchange to Improve Patient Care).

An EHR is not a prerequisite, though

We discovered that having an EHR is not a prerequisite for using Patient Centered Health Information Technology. Prior to rollout of its EHR, pharmacists at Whitman Walker Clinic in Washington, DC are using freely available web tools such as MedactionPlan.com to prepare visual medication regimens for their patients. It is therefore possible to begin using tools that inform and activate patients in their care, very economically, and at a level comparable to EHR-equipped institutions.

Queens Health Network in New York City also demonstrates this idea through the use of smart cards, that patients can carry to providers without EHR’s, but with an inexpensive card reader that plugs into any PC.

Unresolved Issues

• Impact of reimbursement model on forward movement
• Awareness of digital divide issues
• Place of PHR deployment alongside EHR deployment - disparities in implementation plans between safety net and non-safety net providers is of concern
• Impact of disparities in technology use on recruitment of physicians in these environments
• Opportunities to implement patient-accessible HIT outside of an EHR implementation

Countermeasures

Provider collaborating using a state of the art electronic health record, East Boston Neighborhood Health Center, Boston,MAk

We plan to continue a focus on these organizations in 2008. Our next site, Urban Health Plan, in New York City, has a functioning EHR and is planning to rollout an associated PHR. We are working to arrange co-visitation with its payer, to explore financing models “on the shop floor.” In addition, we are separately preparing information about digital disparities, and are actively engaging with safety-net organizations that are implementing EHR’s now (see DC Primary Care Association - Improving Access and Quality using health information technology) or are about to implement PHR’s (Institute for Family Health) in the interest of changing perceptions in this community of care organizations.

Ways to Engage

At the current time, there are several organizations with active EHR programs with an interest in PHR deployment, and we will continue to work with them (Institute for Family Health, Urban Health Plan, Cambridge Health Alliance). It seems most appropriate to spend time studying their experience and generalizing to other similar providers. Conversations with payers as part of this engagement would also be useful.