A large, longitudinal study of EHR introduction in a not-for profit integrated health system shows a reduction in outpatient visits, an increase in virtual patient visits (either scheduled telephone visits or online through a patient portal), and maintenance of patient satisfaction. The study recently appeared in Health Affairs by Chen et al (here, but requires a subscription, medline reference here). Kaiser, an integrated health system, has a health insurance product, a hospital system, and a tightly integrated, staff-model (i.e. its docs are salaried) medical group. Their Hawaii region implemented a system starting in April 2004, which included an EHR with comprehensive doucmentation across care settings, secure patient-provider messaging and electronic interprovider messaging about care that is automatically incorporated into patient' records. Much of the tool is apparently the EPIC ambulatory care product. The Hawaii region was the first of their regions to go live in a staged roll-out.

The authors studied outpatient contacts (outpatient, urgnt care and ED, external referrals, scheduled telephone visits and patient-physician emails) in 2004 through 2007. During this time, the membership (that is the number of people whom they served) and physician staff numbers were stable (approximately 225,000 members, with 1.9 physicians per member).

Total office visits per member decreased 26.2%, but patient contacts (including office, telephone visits and email), increased 8.3 % (5.18 contacts per member to 5.61 ) between 2004 and 2007. The rate of referrals to external providers decreased, and the rate of ED and urgent care visits increased. Patient satisfaction scores and HEDIS scores were reported not to have changed.

In essence, they found that their patients seemed to substitute virtual visits (scheduled telephone and email contacts with their docs) for an office visit, and didn't seem to mind.

Several comments:

The use of virtual visits can save on bricks-and-mortar costs, and probably on staffing costs —one doesn't need to build new clinics or offices, to pay the utilities and other overhead, and to staff these clinics—to accomodate virtual visits. In addition, docs and other providers can accomodate these virtual visits when convenient. (For instance my brother-in-law, who is a Kaiser internist on the mainland, has ~one hour scheduled at the end of his day to answer patient email. My sister, who is a Kaiser beneficiary, reviews results of her lab tests via the patient portal, and isn't required to return to the office to review the results —she likes this). Both patient and doc need to adjust their workflow to take advantage of this new tool, but it may have cost and capital allocation implications if the findigs are confirmed.

It's worth noting that our current re-imbursement model, for most docs, actively discourages virutal visits, email interactions, or scheduled telephone visits. Substitution of vitrual visits can be a good thing, but note that in the Kaiser system of salaried physicians, there is no loss of income for substitution of a virtual visit for an in-person visit. The current payment methodology doesn't provide for re-imbursement of the time needed to provide a virtual visit. Until the re-imbursement model can be worked out, it's likely that virtual visits will have limited use, and only serve as a substitute for interactions which are not currently re-imbursed (e.g. calls for perscription refills, appointment scheduling, lab test follow-up, and the like).

Ominiously, ED and urgent care visits increased with introduction of the system. The authors provide no explanation —it may be that the change in physician behavior to virtual visits drove patients to the ED or urgent care department to have a personal interections. Or, the patients did not have their problems satisfactorially attended to with their virtual visit to their physician, and had to revert to the more expensive ED or the urgent care visit (where there is likely less continuity of care). These are observations worth confirming, as they may actually increase costs.

Finally, their finding that specialty referrals went down also bears further study. This may be a good thing or a bad thing. There are many explanations. Primary physicians may not have appreciated the complexity of their patient's problems during a virtual visit, and not referred the patient for specialty consultation. Contrariwise, the patients may have been more satisfied with the written information provided at the portal or in the email, and not requested a specialty opinion. Or there may be other reasons.

This is a nice beginning, but the absence of cost data, and the retrospective nature of the study should be considered when interpretting the results.

The abstract:

We examined the impact of implementing a comprehensive electronic health record (EHR) system on ambulatory care use in an integrated health care delivery system with more than 225,000 members. Between 2004 and 2007, the annual age/sex-adjusted total office visit rate decreased 26.2 percent, the adjusted primary care office visit rate decreased 25.3 percent, and the adjusted specialty care office visit rate decreased 21.5 percent. Scheduled telephone visits increased more than eightfold, and secure e-mail messaging, which began in late 2005, increased nearly sixfold by 2007. Introducing an EHR creates operational efficiencies by offering nontraditional, patient-centered ways of providing care.