September 6, 2007

P4P’s peek into the future of medical charting

The Leapfrog Group, a powerful group of major company and other large private and public healthcare purchasers known for supporting quality and cost-control incentives, promotes the results of this year’s Pay for Performance (P4P) and Consumer Incentive Survey of 75 “purchasers, government agencies, and health plans”. Conducted biannually since 2003, this year Leapfrog partnered for the first time with the P4P specialty software company Med-Vantage. The actual results are not online, only the press release, which describes several statistics: a nearly four-fold increase in the number of P4P programs to a current 148; a cost savings noted by about a third of respondants; and that provider performance was made publicly available by over 30% of respondants. 

An article posted earlier this week on the Web site of magazine Modern Healthcare Online commented on this press release: “Even as payers issue glowing news releases about the expansion of their programs’ scope and the millions of dollars in physician rewards they’re dispersing, government officials, research organizations and physicians are questioning the motivation and methodology of current pay-for-performance programs.” The above Leapfrog/Med-Vantage survey is contrasted with a report on P4P among commercial insurers released several weeks ago by PriceWaterhouse, which describes significant differences among these programs’ structures, performance metrics and rewards structures. 

The Modern Healthcare article quotes a surgeon/attorney who describes P4P as, “an attempt to control prices—but not for the benefit of the patient or the benefit of the doctor.” From elsewhere in the article: “Meanwhile, the physician-ranking programs [used by over 30% of respondants to the Leapfrog/Med-Vantage survey] under consideration for use by insurers Aetna, Cigna HealthCare and UnitedHealth Group have come under scrutiny by New York State Attorney General Andrew Cuomo, who expressed concern that the rankings may be used to steer patients toward less-expensive rather than higher quality providers.”

HIT is the elephant in the P4P room. According to the Modern Healthcare article, “In the end, the one industry segment that may truly be happy with pay-for-performance programs are health IT vendors who can persuade providers that their electronic systems are absolutely necessary for the data management that require such programs.” The article closes with a comment by a Pricewaterhouse director who believes that wider IT adoption would “‘pave the way for standardized measures,’ mak[ing] it easier to validate physician data and reduce the administrative costs of the programs.” The Leapfrog/Med-Vantage press release agrees: ”Advanced P4P programs are now developing tools to measure improvements in outcomes and eligibility for rewards directly from medical charts.”

August 16, 2007

Risky Business

The GRAIDS Trial: A Cluster Randomized Controlled Trial of Computer Decision Support for the Management of Familial Cancer Risk in Primary Care, published in the August 14, 2007 issue of the British Journal of Cancer, studied physicians’ use of family history risk assessment software in counselling patients. In this study, researchers in Australia and the UK used the software program, Genetic Risk Assessment on the Internet with Decision Support (GRAIDS). The research team randomized 45 group practices into two sets: one set was trained in the use of GRAIDS, which “links a user-friendly pedigree-drawing tool to patient-specific management advice regarding a family history of breast/ovarian and colorectal cancer, and provides additional numerical risk information about breast cancer.” The control group was trained in the current best practices for genetic referrals. Patients who were found to be at an increased risk in either set were referred to a regional genetics clinic. 

Outcomes were measured at practice, practitioner and patient levels. The trial found that the GRAIDS group both increased the number of genetic referrals, and ensured referrals more closely adhered to official referral guidelines. Additionally the GRAIDS group was able to identify more at-risk patients than the best practices group, and practitioners felt more confident in managing familial cancer. Positive patient outcomes were associated with GRAIDS usage as well; patients who were deemed at-risk in the GRAIDS group reported that they felt less anxious about their risk of cancer than compared to those patients in the best practices group. 

Genetics is increasingly important in clinical medicine, so it is no surprise that family history cancer screening software is being developed in countries throughout the world. For example, the study’s news release reports that Australian government is now funding a national electronic family history tool, based on GRAIDS software. In the United States, the CDC is developing its own software, Family Healthware, to help assess a patient’s genetic risk for coronary heart disease, stroke, diabetes, and colorectal, breast and ovarian cancers. The tool is currently being tested at three academic centers “to determine if personalized prevention messages tailored to familial risk will motivate people at risk to change lifestyle or screening behaviors.” Data collection is expected to be complete by the end of 2007. 

August 14, 2007

Requiem for an HIE Dream

A Health Affairs supplement from earlier this month offers five complementary post-mortems on one of the nation’s most pioneering and visible health information exchange (HIE) efforts: the Santa Barbara County Care Data Exchange or Santa Barbara Project (SBP), which ceased operation this past December after only eight years.

HIEs—advocated in the supplement’s final article by the CEO of the successful Indiana Health Information Exchange (IHIE) as “an essential strategy”—are local information systems designed to share healthcare data among physicians and others providing medical care to community patients. Examples of HIEs include Regional Health Information Organizations (RHIOs) (such as SBP and IHIE), and the CDC’s Public Health Information Network (PHIN). The current vision of the National Health Information Network (NHIN) is as an interconnected string of local or regional HIEs.

The first article, written by Professor Robert H. Miller, PhD (Health Economics), University of California, San Francisco,  tells What Happened to the SBP. The Project began in 1998 as an HIE demonstration experiment proposed by the technology company CareScience (whose CEO at the time was David Brailer, MD, who later became this nation’s first National Coordinator of Health Information Technology) and funded by the independent philanthropy California HealthCare Foundation. The article describes the details of SBP’s early evolution, slow progress, CareScience’s technical difficulties and business upheavels midway through, and SBP’s subsequent revamping and ultimate demise amid legal liability issues and insufficent post-grant funding. 

The rest of the first article, and the other four that follow, propose reasons for SBP’s demise and the lessons learned. ”The main underlying cause [for the Project’s slow progress, according to the first article,] was lack of a compelling value proposition for Santa Barbara [participating healthcare] organizations [and physicians].”  Almost none of the Project’s participants saw medical or financial value in the project, partly because physicians could already obtain electronic patient data using the less extensive and relatively closed Web portals already in place. SBP highlights the lack of efficiency and security inherent in peer-to-peer (Napster-type) design; the importance of active local governance; the need for an incremental series of small successes rather than, as CHCF staff describe in the third article, an “all-at-once” design; and how long-term planning results in a sustainable business model. SBP also demonstrated how HIEs need to be flexible, to change and grow in response to the community and to keep pace with the advances in technology and evolving needs of practitioners and their patients. Even today, and in spite of the advances in technology compared to SBP’s beginnings nearly a decade ago, HIEs are still very much a work in progress, still awaiting practical standards and privacy laws, software advancement and funding options in most communities. 

As Dr. Brailer closes in the second and easily best article of this strong series, From Santa Barbara To Washington: A Person’s and a Nation’s Journal Towards Portable Health Information, ”Thomas Kuhn described in The Structure of Scientific Revolution how the cumulative weight of research and experimentation–whether positive or negative–can hasten the collapse of an existing paradigm. We are in the earliest steps toward major upheaval in the obsolete paradigm of U.S. health care. Health IT is one of the prime forces of innovation and disruption. It will both hasten this change and soften the fall when change does occur. Projects like Santa Barbara—whether they ’succeed’ or ‘fail’—are part of a justified and relentless attack on the status quo of health care amid the unending hope for something better. Without these efforts, the old paradigm will continue, and we will have no chance for meaningful progress.”

August 2, 2007

Ordering CPOEs

Two studies in the current issue of Journal of the American Medical Informatics Association (JAMIA) highlight the current frustration with, and potential importance of, computerized physician (provider) order entry (CPOE) systems. The electronic entry of physician instructions for patient care has been touted by the government as a way to reduce medication errors and improve safety, although studies to date have shown mixed results. CPOE has the potential to change patient care, for better or worse, particularly when the ordering process is linked to clinical decision support software, clinical guidelines, and alerts.

Joan Ash, PhD, Department of Medical Informatics and Clinical Epidemiology at Oregon Health & Science University’s School of Medicine, and colleagues followed up their earlier work on unintended consequences of CPOE use at five selected hospitals, with this current study, The Extent and Importance of Unintended Consequences Related to Computerized Provider Order Entry. Here, they sought to measure the extent and importance of eight of these previously identified types of unintended consequences at every US hospitals with a CPOE system. A total of 176 hospitals completed the telephone survey, a response rate of 47%. Six of the eight types of unintended consequences previously identified by the five hospitals were also considered important by the majority of these 176 hospitals: workflow issues; communications issues; overdependence on technology; never-ending system demands; emotions (usually negative); and more/new work issues. The remaining two types—changes in power structure, and new kinds of errors—identified as important in the preliminary study were not regarded as important with the majority of these hospitals. The authors note an important limitation of their study: they queried individuals who “might have given answers biased in a positive way because they believe in CPOE’s benefits. This might help explain why the two categories of power shifts and new kinds of errors did not rank highly…” The authors’ recommend that hospital leaders charged with implementing CPOE consider each of the eight types of unintended adverse consequences carefully during their planning. “Implementation success depends on managing the unintended, as well as the intended, consequences of CPOE…Unless we make a concerted effort to avoid, manage, and/or overcome unintended consequences, the implementation of clinical information systems may lead to detrimental results.”

Evaluation of Outpatient Computerized Physician Medication Order Entry Systems: A Systematic Review provides a literature review of outpatient CPOE evaluation studies to determine if outcomes exist related to improved patient care. The authors identified 30 such studies, a relatively small number with few if any of them deemed to be well designed. The study’s main finding was that the available evidence does not demostrate that CPOE enhances safety or reduces medication costs, although they did find an increased adherence to clinical guidelines. The outpatient setting presents challenges to CPOE research because inpatient orders tend to be more easily collected and reviewed due to centralized hospital information systems.

As technology advances and the applied research aspects of medical informatics matures, we can hope for better, more clinically relevant studies on software and best-of-breed applications directed at evidence-based practices.

July 26, 2007

Phlaunting PHRs

A July 17th press release by Aetna underscores the current push for patients to use personal health records (PHRs), by government, private insurers, large employers—even Google

The promises of a patient-centric healthcare system where medical information is easily accessible by all and patients are full participants in their care, as exemplified by PHRs, is balanced by concerns over privacy of personal data, lack of medical record interoperability, potential disruption in the physician/patient relationship, confusion over definitions and goals of PHRs, and lack of evidence demonstrating their efficacy and improved quality care. For those interested in learning more about PHRs, the independent philanthropy California Healthcare Foundation published Perspectives on the Future of Personal Health Records last month. The report provides a forum for six computer-literate healthcare and legal professionals to share their knowledge, fears and visions on PHRs.  

July 19, 2007

Patient Access to an Electronic Health Record (EHR) With Secure Messaging: Impact on Primary Care Utilization” from the July issue of American Journal of Managed Care is self-described as “the largest study to date of the impact of access to secure patient–physician messaging on provider workload.” This retrospective study was conducted from 2002-2005 at the nonprofit managed care Kaiser Permanente Northwest (KPNW). It looked at nearly 4700 patients who were already active enrollees in HealthConnect Online service (which provides patient access to limited sections of their EHR and secure emailing) with 3200 of them also matched by age, sex, selected chronic conditions and primary care physician to a control group. They found a 6.7% drop in office visits (p<0.003), and almost 14% drop in phone calls (p<0.01) among HealthConnect Online participants as compared to controls. The authors conclude that, “Electronic messaging may provide a solution to pervasive efficiency and access issues for both patients and providers.” There is no mention of resultant changes in quality of care other than this taste: “KPNW collects data for the Health Employer Data and Information Set (HEDIS) as part of routine quality surveillance. The HEDIS reports for HbA1c testing did not vary to a statistically significant degree during the years under observation,” and there is no mention of broad patient satisfaction or changes in physician income. Modern Healthcare made this observation in their writeup: “Naturally, because Kaiser is an integrated delivery network, officials there can look upon technology driving a near-10% drop in office visits with far more equanimity than a fee-for-service physician group leader would.”

July 15, 2007

Quality Care by MDs, not EHRs

According to “Electronic Health Record [EHR] Use and the Quality of Ambulatory Care in the United States” from the July 9th issue of Archives of Internal Medicine, using EHRs in the out-patient setting does not appear to improve quality of care (as measured by compliance with 17 ambulatory care quality indicators developed at Stanford University). The retrospective study analyzed data drawn from the 2003 and 2004 National Ambulatory Medical Care Survey, an annual survey conducted by the CDC’s National Center of Health Statistics. During these two years, 18% of physician visits used EHRs and, when compared to physician visits without EHRs, the investigators found no statistical difference among the outcomes in 14 of 17 indicators.  
          Jeffrey Linder, MD, lead author and general internist at Harvard’s Brigham and Women’s Hospital, explained in an associated iHealthBeat article that, “…other studies have shown that [EHRs] are not much more than a replacement for the paper chart. In light of those findings, this is not that surprising…” Randall S. Stafford, MD, PhD, senior author and associate professor of medicine at the Stanford Prevention Research Center, offers his theories on why EHRs were not found to improve quality care: the study looked at older systems that probably lacked clinical decision support, and that “No matter how sophisticated the system, it can’t dictate a course of action to a physician…” For more on how the authors interpreted their study, see Stanford University’s press release. (Ed. note 07/19/07: For a different spin, see WSJ Health Blog’s Computerized Medical Files Not Much Better Than Paper, and the comments that follow.—ML)

Powered by WordPress