Transitioning patients from clinic to physician-managed care didn't work

Significant decrease in control after transition to physician

What do you do when your successful hospital-based anticoagulation clinic has more patients than available slots? Detroit Medical Center pharmacist Candice Garwood, PharmD, faced that situation about three years ago and tells Drug Formulary Review she thought she had come up with a workable solution—transition the most stable patients from the pharmacist-managed clinic to their physicians to continue their anticoagulation care.

"Out of our dire need to be able to care for more patients, we developed a proposal and presented it to the hospital P&T committee," Garwood tells DFR. "Our study to determine the success of that proposal came as a result of the change. The P&T committee approved the plan but wanted us to determine how well it had worked."

The plan was simple: Discharge patients after their warfarin therapy was stabilized to their physicians so the clinic pharmacist would have more time available for less stable patients.

"We thought it went well," Garwood says. "But when we analyzed it, it didn't seem like we were doing very well."

Garwood and her colleagues from Detroit Medical Center and the Eugene Applebaum College of Pharmacy and Health Sciences at Wayne State University published their findings in the January 2008 issue of Pharmacotherapy.1

Optimal warfarin dosing is expressed in terms of international normalized ratio (INR), a test measuring blood clotting time and comparing it to an average. According to recent research, patients with an INR that is in the therapeutic range more than 75% of the time have significantly fewer episodes of major and minor bleeding. "Patients' INRs must remain stable within their therapeutic range to minimize complications associated with anticoagulation therapy," Garwood says.

In a number of other studies, anticoagulation clinics managed by pharmacists have demonstrated improved care in patients receiving warfarin therapy compared with patients managed by their physicians.

Not only have pharmacist-managed clinics been associated with improved patient outcomes, including greater time in the therapeutic range, lower rates of significant and fatal bleeding, and lower rates of thromboembolic events, but they also have resulted in lower health care costs.

Thus, one study found that cost reductions associated with pharmacist-managed clinics included fewer emergency room visits, decreased hospitalizations, and an annual health care cost savings of $162,058 per 100 patients in 1998.

Identifying transition candidates

Garwood says the Detroit Medical Center transition program was proposed to allow new patients, and especially high-risk patients receiving bridge therapy with low-molecular weight heparin and warfarin therapy, to be seen in the clinic.

"Clinic personnel, pharmacy administration, and hospital administration determined that the best candidates for discharge from the anticoagulation clinic would be those who had achieved a high level of anticoagulation control," Garwood says, "as it was assumed that these same patients would be most likely to maintain good anticoagulation control once transitioned to physician-managed care."

The clinic's discharge policy for patients stabilized on anticoagulation therapy was presented to the Cardiovascular and Thrombosis subcommittee of the P&T committee. Patients were defined as clinically stable and eligible for transfer to physician-managed care if five of their last six INR values were within the targeted therapeutic range when the patient's therapeutic goal was 2.0-3.0 or 2.5-3.5, or having four of their last six INR values in the targeted therapeutic range for those with an individually narrowed therapeutic range (such as a range width of 0.5 INR unit). Patients were excluded if they discontinued warfarin within six months after discharge from the anticoagulation clinic or if they had resumed receiving medical care at one of the medical center's other anticoagulation clinics.

A total of 42 patients went through the process of stable transition to physician-managed care. Two of the 42 were subsequently excluded from the analysis because one had anticoagulation therapy stopped by his physician and the other was referred by his physician to a different pharmacist-managed anticoagulation clinic within the medical center. Records on the remaining 40 patients were retrospectively reviewed.

Anticoagulation control decreased

The analysis revealed a significant decrease in anticoagulation control in the patients who transitioned to physician care. Before the transition, 76% of INRs were in the target range. But after transition only 48% were in the target range. Garwood says they found the median time to first follow-up INR after transition was 41 days. They also found that the number of INRs above 4.5 and below 1.5 increased significantly after transition from the anticoagulation clinic.

Likewise, after transition there was an increased need for additional medical care related to either over- or under-anticoagulation. Before transition from the anticoagulation clinic, two anticoagulation-related emergency department visits for symptoms related to bleeding were reported for the same patient. After transition of stable patients, however, 13 cases of additional medical care were reported among seven patients. Seven of the cases required a physician office visit and six resulted in emergency room evaluation. Twelve of the events after transition were related to bleeding and one was thrombosis-related. Garwood says the frequency of INR assessment decreased after transition to physician-managed care.

Patient satisfaction surveys on pharmacist-managed care were completed by 39 of the 40 study patients, while 23 patients reported on physician-managed care. Garwood says statistically significant differences favoring pharmacist-managed anticoagulation were noted in all clinical care areas. There were no statistically significant differences in service-oriented questions.

"This study suggested better control of INR values and decreased need for anticoagulation-related medical care when warfarin management was performed by a pharmacist-managed anticoagulation clinic than by a physician in a usual care setting," Garwood says. "This low level of anticoagulation control was not unique to first follow-up, as it persisted through the six-month period of usual care management with only 36.5% of each patient's INRs in the therapeutic range. We believe that the stability of patients receiving anticoagulation therapy in our clinic is a function of patient participation in care and of clinical services provided by the pharmacists. Loss of anticoagulation control in our patients after transition to physician care supports our contention that the pharmacist's involvement in anticoagulation management has a significant impact on stability. Once patients were transitioned out of the clinic, our study reflected previously published findings, as anticoagulant control reverted to a decreased level consistent with other studies."

Garwood says study analysis revealed some additional interesting findings. First, there was considerable variability in frequency of INR assessment after transition to physician usual care. Although guidelines from the American College of Chest Physicians recommend an interval of INR monitoring no more than four weeks apart, even in patients with the most stable INRs, physician monitoring at Detroit Medical Center took place on average at intervals of more than four weeks. "In fact," she says, "time to first follow-up of INR after transition was a median of 41 days, which is well beyond the recommended interval of four weeks. Included in our findings is the discovery that three patients continued to receive warfarin after transition from the clinic without a single INR determination during the six-month transition period."

The average time between INR evaluations before transition was three weeks. Garwood says that can be explained by the fact that the clinic pharmacists preemptively will reassess anticoagulation based on nutritional and drug therapy changes to prevent deterioration of anticoagulation rather than reacting to deterioration. As such, she says, some patients will sometimes require more frequent visits to prevent deterioration in anticoagulation control, thus maintaining stability.

Practice environment may be key

Although the root cause for discrepancy with regard to INR reassessment between the pharmacist-managed anticoagulation and physician-managed care could not be determined from the study, Garwood tells Drug Formulary Review she has some ideas on the subject.

"First," she says, "the clinic saw people in person while the doctors often managed anticoagulation by phone. There may be decreased vigilance in INR monitoring by physicians who are overburdened with other patient concerns. Also, there may be less familiarity with the pharmacokinetic properties of warfarin and its dosing among physicians when compared with pharmacists specializing in anticoagulation management. Finally, the discharge policy may not have created a seamless transition of anticoagulation management back to the physicians."

The way the transition worked, physicians were notified of the policy by letter and then were contacted by fax or mail with information on transition care for each patient. Patients were instructed to contact their physician for their next appointment.

"Anticoagulation clinic patients routinely undergo extensive education on the importance of regular INR monitoring," Garwood says, "with education repeated at every clinic visit. But the study revealed a lack of patient self-directed care as evidenced by patients allowing decreased INR monitoring despite receiving extensive and repetitive education to the contrary." Satisfaction survey results showed that patients believe the clinic pharmacists had more experience managing anticoagulation control when compared with their physician. Patients also noted satisfaction with point-of-care testing, which allowed for immediate INR results and anticoagulation management.

Garwood tells DFR that because of the study results, the transitioning of patients to physician care was stopped and all patients are once again being treated in the clinic. "We knew the P&T committee would not look favorably at what was happening and so we resumed keeping the patients," she says.

However, during this interview, Garwood said another crisis was coming in the clinic and it might soon have to close to new patients unless another FTE clinic position could be added. However, adding a new position would create its own space and other issues.

Asked if she sees a way that anticoagulation clinics could successfully transition stable patients to physician care, Garwood suggests it might work better in a closed system. "At Detroit Medical Center," she says, "any doctor can refer patients to the clinic and thus transitioning patients often were returned to physicians who had no direct relationship with the clinic pharmacists. There were all sorts of problems, just in finding physician contact information, finding patients who had moved, etc. A system like this might work better if patients are coming from only one practice and are sent back to a doctor who sees them routinely and who has a relationship with the pharmacists. Not having a personal relationship with the doctors who referred patients raised the potential for a bad result. So much is dependent on the practice environment."

[Editor's note: For more information contact Dr. Garwood at or telephone (313) 966-7883.]


  1. Garwood CL, Dumo P, Baringhaus SN, et al. Quality of anticoagulation care in patients discharged from a pharmacist-managed anticoagulation clinic after stabilization of warfarin therapy. Pharmacotherapy 2008;28:20-26.