What Technology is Needed to Improve Care: EHRs or Registries?

RS Head Shot 1

I spent a couple of days last week (June 10-11) at a conference co-sponsored by the HRSA Center for Quality and the NIH National Center on Minority Health and Health Disparities at which twenty of the highest-performing community health centers from the HRSA’s long-running Health Disparities Collaboratives (HDCs) described their accomplishments in improving primary care. They were stunning.

These health centers were not fresh-faced newcomers to the improvement collaborative du jour, but rather grizzled veterans who joined the HDCs ten years ago, have stuck with the method, and by now have embedded quality improvement deeply into their day-to-day operations and organizational culture.

If ever there could be a vivid demonstration of what is possible and right in the future of American primary care, this was it. Consider:

  • At CareSouth in South Carolina, under the leadership of Ann Lewis, the annual average cost of care for a diabetes patient is $343, compared with $1591 at comparable community providers in SC. Outcomes for these 3500 diabetes patients have improved steadily, with 50% of HbA1c’s less than 7.0. Appointments are available within 2 days. Staff turnover at CareSouth is 3%.
  • At FamilyCare in West Virginia, CEO Martha Carter has overseen a steady, radical improvement in diabetes care. Since 2002, the diabetes panel has increased from 1000 to 2700 while at the same time their average HbA1c has dropped from 7.8% to 7.1%. In the same period, the percent of pediatric patients with ER visits in the previous 6 months declined from 70% to 6%.

As team after team presented their stories, it became clear that these results were being achieved by safety net organizations on shoestring budgets, with highest-risk populations, despite the extra burdens of improvement work, data management and reporting. At La Clinica de la Raza in Oakland California, since 2007 100% of persistent asthmatics have been on controller medications. At Holyoke Health Center in Massachusetts the percent of diabetic patients with HbA1c’s over 10 has declined from 29% to 10% since 2000. At Hunts Point-Mott Haven in New York City, asthma hospitalization rates have decreased by 58% since 1997.

Together, these CHC leaders described a radically transformed model of primary care based on a fundamental change in attitude: instead of caring for patients ‘one at a time,’ they have accepted responsibility for a population of patients, based their criteria for quality on measures of the whole population, and persistently sought ways to improve those measures.

All of these high-performing health centers relied on information technology to achieve and measure results. But they referred to their systems as ‘registries’ – not as ‘electronic health records’ or ‘EMRs’.

EHRs versus Registries? – Lets Talk About Functions Instead

Despite the ever-present risk of oversimplifying a complex issue, and without worrying too much about what is really an EHR or a registry, let’s try to sidestep the controversy and focus instead on the software functionality that is needed to support the new model of primary care, as illustrated by these high-performing community health centers.

Today’s discussions of EHRs typically center on a relatively narrow range of functions that support the ‘one-at-a-time’ model of care: storing and displaying patient-specific clinical information at the point of care, reminding providers of the care needed by a patient during a visit, increasing the reliability and safety of common processes such as ordering tests and prescribing medications. When patient data can be readily shared among providers, these benefits are amplified across the continuum of care. These functions are important, but are only part what is needed to improve quality.

The registry functions described by high performing CHCs focus on using data to directly monitor and improve care for a population. We can distinguish two broad levels of registry functionality.

‘Base level’ registry functions are the starting point for practices working on quality improvement and are widely familiar to those participating in improvement collaboratives. They include:

  • Planning care for individual patients in advance of an office visit, based on evidence-based protocols
  • Providing ‘opportunistic’ care – that is, take every opportunity to provide care, even if it is not the reason for the visit
  • Tracking quality indicators  (for example, the percentage of diabetic patients with current eye exams) to evaluate effectiveness  of process improvements
  • Identifying gaps in performance to prioritize process improvements

The high performing CHCs who presented at the HRSA/NIH conference, many with over 10 years of experience in sustained care improvement, described uses of their registries that extend far beyond either EHRs or base-level registry functions. Advanced registry functions mentioned by these teams include the ability to

  • Stratify patients by severity in order to target planned care (for example, patients with DM with a recent increase/decrease in BMI)
  • Provide patient-specific outreach reminders to patients for needed care
  • Create individual patient care plans with goals, track patient progress against the plan, and identify subgroups of patients based plan progress
  • Identify subpopulations (by gender, age, geographic area) to differentiate process improvement interventions
  • Identify subpopulations with special needs to launch community-wide campaigns (e.g. high utilizers with no insurance and special needs)
  • Provide ubiquitous performance feedback to motivate providers and staff
  • Identify groups of ‘non-compliant’ patients to diagnose causes and target interventions such as promotora, case management, or group visits
  • Organize, coordinate and schedule ancillary and community-based services
  • Create and manage provider panels
  • Deliver transparent provider-specific feedback to encourage productive competition
  • Identify patients for clinical trials
  • Distribute a newsletter to all families in the practice that includes current performance data
  • Summarize utilization improvement and cost savings for negotiations with payers and networks, and to obtain grant funding

HITEP or Mis-Step?

As I listened to the conference presentations, I scribbled notes. I was pushing a deadline to respond to a Request For Comments from the Office of the National Coordinator for Health Information Technology (ONC) [http://edocket.access.gpo.gov/2009/pdf/E9-12419.pdf ] regarding a proposed ‘Health Information Technology Extension Program’ (HITEP). This program, fueled by $2B of stimulus funds, envisions a national network of ‘Health Information Technology Regional Extension Centers’ which will employ ‘extension agents’ to coach providers on the selection, implementation and ‘meaningful use’ [whatever that means] of certified EHRs.

Parsing the RFC during coffee breaks, I was struck by the disconnect between the ‘silver bullet’ overtones of the ONC’s vision of EHRs everywhere, and the dogged improvement work being carried out by the CHCs presenting their results to the audience of HHS personnel and health services researchers.

I was also struck by the fact that no ONC staff attended the meeting, although many agencies across HHS were represented. So I fear that the message about the software functions that are so ‘meaningful’ for quality improvement will go unheard as HITEP is put into play.

As I was leaving the conference I talked to a physician leader of one of the CHCs who was in charge of his center’s EHR adoption project. His center has been a long-standing user of registry software and has accomplished dramatic improvements in care. Their group adopted an EHR product based primarily on it customizability, and he has personally spent half of his time for the past two years reshaping the EHR software to support the registry functions that his practice has come to rely on.

More than ever, I am fearful that the ONC is pushing technology for technology’s sake, without thinking through the software functions that are actually needed by the providers who are pioneering the transformation of primary care.

ehrRichard Scoville, PhD, is an independent consultant specializing in healthcare quality improvement and performance measurement. He is an Adjunct Professor in the Department of Health Policy and Administration at the University of North Carolina at Chapel Hill where he teaches courses in healthcare quality improvement and informatics. He serves as an improvement advisor to the Institute for Healthcare Improvement, NICHQ, the Cincinnati Children’s Hospital Medical Center, and the Federal Health Resources and Services Administration on a range of collaborative improvement and systems design projects.

Categories: Uncategorized

Tagged as: ,

4 replies »

  1. I was pleased to see at least a fleeting reference to Promotoras, but would assert that they play a larger role in many CHCs and other provider organizations than just intervening with ‘non-compliant’ patients. Promotoras are part of a wider workforce of Community Health Workers (now recognized as an occupation by the USDOL at HRSA’s urging). CHWs are active in everything from outreach and population-based prevention to patient navigation, advocacy and chronic disease management. A provider coalition in Camden NJ is using a patient registry to identify frequent ER users with chronic conditions, who are then supported by a team of a CHW, a FNP and a social worker in bonding to a “medical home.” A group of 24 health organizations including the AMA, American Hospital Assn and American Nurses Assn wrote to Pres. Obama last month urging strategies to address disparities within healthcare reform; under their recommendation to expand the primary care workforce they explicitly list CHWs with other professions. Minnesota has recently started reimbursing for CHW services under FFS Medicaid. I could go on…

  2. I am the CEO of a 300+ physician IPA in CT. Many of our PCPs and medical specialists have been using a registry software program (DocSite) for a few years. They are using it to help manage patients with chronic illnesses (diabetes, cardiovascular disease, asthma) as well as to help keep their patients compliant on immunizations & preventive screenings, etc. Our cardiologists are also using it to help manage their warfarin (cumadin) clinic patients, as well. We are getting relevant labs uploaded automatically from Quest Diagnostics. My staff is entering hospital lab data. Almost none of these physicians have an EHR, yet they are achieving top process and outcome scores in all of our main P4P programs. We are driving big improvements in clinical quality and health outcomes without having to make huge investments in costly EHRs that can’t communicate with each other. Most of our physicians print off a paper Visit Planner (clinical superbill) which details chronic conditions, relevant lab results, evidence-based guidelines which details actual vs. goals and highlights when results are above goal or process needs to happen (labs, visits, etc.). Visit Planners are faxed to our office daily for centralized data input. This process ads very little new work to overburdened PCP offices and also is very cost effective. Cost to our docs is $60/month including centralized data entry. Almost all are achieving major ROI (700% to 1000%) in added revenue from P4P programs. Reporting for Medicare PQRI is a snap. Our collective quality metrics are continuing to improve as we bring ever more docs onto our system. This quality improvement program with low cost and real ROI has been a big draw for docs to join our IPA. We continue to try to work with web-based companies that offer real opportunities for clinical improvement, population management, data aggregation and interoperability at low costs. We are specifically not going with “big iron” EHRs based on client-servers and proprietary databases that can’t communicate, manage populations and cost a ton of money. Web-based applications like the DocSite registry and Practice Fusion web-based EHR offer much more value.

  3. Great write-up Richard. The community health centers have been doing a bang-up job for years!
    We and a few others have been offering chronic disease improvement registries with point-of-care decision support for many years. Throughout much of that time, it has been a market without a lot of commercial interest, although that has changed significantly this year. Timing appears to be everything. The market seems to be embracing the importance and simplicity of improving care in ways that work in the office. It’ll be great if the meaningful use definitions land on evidence-based care improvement methods.
    Improvement results similar to the examples Richard describes above in community clinics can be found in settings across the healthcare system. Examples from our users include – 1) HRSA critical access hospitals, 2) Large 300+ doctor PHOs (A1c in poor control from 26% to 12% in 4 months) 3) Rural regular primary care practices 4) Across IPAs (controller med use improvement doubled in 6 months, perfect care in asthma doubling in 12) 5) State health exchanges (vermont blueprint)
    These results often occur without the physician touching the computer directly. Instead, the registries with decision support emphasize getting the information into the care workflow, to the team and to the physician in either a paper or electronic office. The systems emphasize accomplishing the change (improved care) rather than prescribing the mechanism of office workflow (getting the doctor on the keyboard).
    The same systems allow users to successfully report commercial P4P and PQRI results – more than $4,000,000 back to providers in our customer base, lots more throughout the market.
    Heck, CMS in 2003 and the California Healthcare Foundation a year or so later pulled expert panels together to discuss the value of various chronic disease registries and improvement systems. The difference from then to now, is that the market itself is emphasizing the need to improve care as a critical driver. (Chronic Care Registries – population clinical data repositories with performance reporting and outreach functions, along with patient level, personalized planned care capabilities that fit within workflow — populated with structured data – from feeds, entered by the user or both).
    Thanks for writing up the details above, Richard.
    John Haughton

  4. Richard,
    The point you made earlier about them achieving stunning quality improvement is what I have been telling people that you do not need more money to improve healthcare. Sadly, you hear everyday in news that we do not have money or it will cost over a trillion dollar.
    There is enough money in the healthcare system that we can create a world class system, have all covered, and still save some money.
    People who say it will cost more are bean counters and they are scaling up the current crap that is going on rather than looking in a innovative way to offer solution.
    Last night, it even crossed that may be I should start a insurance company and show what I am saying….If the idea persists for a while, I might write it up and look for similar minded investors…