- Email: [email protected]
Enhancing Pharmacists’ Recommendation Process in an Internal Medicine Resident Clinic
EXPERIENCE
Enhancing Pharmacists’ Recommendation Process in an Internal Medicine Resident Clinic Kimberly K. Daugherty and Karin Kangas
Received April 3, 2003, and in revised form June 3, 2003. Accepted for publication August 4, 2003.
ABSTRACT Objective: To determine the number and type of recommendations made by pharmacists in an internal medicine residency clinic, the percentage of recommendations that were addressed and acted upon by the residents, whether changes needed to be made to the recommendation process, and areas in which pharmacy services could be improved. Setting: Outpatient, urban, internal medicine clinic. Practice Description: The clinic operates Monday through Thursday afternoons. Each of these afternoons, six to eight medical residents evaluate patients. Pharmacists provide services on Monday and Thursdays. The pharmacists spend the mornings before patients’ visits reviewing charts and making recommendations for more appropriate medication use. Originally, pharmacists’ recommendations were listed on a form that was attached to the front of each patient’s chart but did not become a part of the medical record. At times, pharmacists made recommendations while a medical resident presented the patient’s case to the attending physician.
Kimberly K. Daugherty, PharmD, BCPS, is Assistant Professor of Pharmacy Practice, Ferris State University, Grand Rapids, Mich. Karin Kangas, PharmD, is Ambulatory Care Clinical Pharmacy Specialist, Spectrum Health, Grand Rapids, Mich. Correspondence: Kimberly K. Daugherty, PharmD, BCPS, 21 Michigan, NE, Suite 425, Grand Rapids, MI 49503. Fax: 616-391-3783. E-mail: [email protected] Disclosure: Kangas is employed by Spectrum Health HealthCare System, which is financially responsible for the internal medicine residency clinic used in this study. Daugherty works in the internal medicine clinic but receives no financial support from the health care system. The authors declare no other conflicts of interest or financial interests in any product or service mentioned in this article, including grants, employment, gifts, stock holdings, or honoraria.
Practice Innovation: Data for calendar year 2002 were analyzed retrospectively to identify a recommendation acceptance rate, areas of potential improvement, and actions that might enhance pharmacists’ effectiveness. Main Outcome Measures: Number of recommendations made, number of each type of recommendation made, percentage of overall recommendations addressed and acted upon by the residents, and percentage of each type of recommendation addressed and acted upon by the resident physician. Results: For 61 patients, 135 recommendations were made by pharmacists during 2002. Of these, 72 (53.3%) were therapeutic interventions and 63 (46.7%) were recommendations for laboratory monitoring. Overall, 66 (49.0%) recommendations were addressed and acted upon by the residents, while 69 (51.1%) were not addressed by the residents. Because of deficiencies identified in the communication process, a new form was developed that includes spaces for pharmacists’ recommendations and residents’ responses. Other potential future enhancements include providing more education for residents and implementing a pharmacist-run disease clinic. Conclusion: This retrospective analysis showed that pharmacists were able to identify many therapeutic and laboratory interventions, thus serving a useful role in patient care and physician education in an internal medicine residency clinic. Keywords: Pharmacy services, medical residents, internal medicine clinic, ambulatory care, pharmaceutical care. J Am Pharm Assoc. 2004;44:89–94. Vol. 44, No. 1
January/February 2004
www.japha.org
Journal of the American Pharmacists Association
89
EXPERIENCE
Acceptance of Pharmacists’ Recommendations
S
pending on prescription drugs has increased faster over the past 5 years than has spending on hospital and physician services.1 The number of new prescription medications approved each year, the increasing age of the U.S. population, use of direct-to-consumer advertising, and the aggressive marketing of medications to physicians by pharmaceutical manufacturers have all combined to help increase the spending on prescription medications.2,3 However, the prescribing of these medications is not always appropriate.1,3 Because of the need to use medications appropriately, many pharmacists have begun providing cognitive services. Cognitive services are defined as clinical activities intended to improve medication prescribing and use.1 Over the past 25 years, pharmacist-provided cognitive services have become a valuable component of many health care systems.4 Many authors have demonstrated the economic and clinical benefits of pharmacists’ services in a variety of clinic settings and for a range of disease states.1,3–5 The ambulatory care setting is optimal for the provision of pharmaceutical care and cognitive services because the pharmacist has access to the patient’s records and is able to review all of the patient’s medication requirements.7 The number of patient visits in the ambulatory care setting has increased over the past few years, but the number of primary care physicians has not.5 Because of the increased demand on their time, many primary care physicians find it difficult to stay abreast of current therapeutic options and to review complex drug regimens.2 One way of addressing this problem is to have pharmacists provide physician education and identify drug-related problems in the ambulatory care setting.2,6 As a result of the growing body of evidence indicating that pharmacists are effective in promoting appropriate medication use, pharmacists’ services were begun at one of the outpatient, urban, internal medicine residency clinics within a large nonuniversity
AT A GLANCE Synopsis: During 2002, pharmacists working two afternoons per week in an internal medicine clinic staffed by medical residents made 135 recommendations for changes in drug therapy or additional laboratory monitoring. About one half of these recommendations were acted upon by residents, somewhat lower than in some past published studies. After analyzing their experiences, authors of this article concluded that improvements were needed in the way they communicated with residents. A new form was developed that became a part of patients’ medical records and provided a place for residents to respond to pharmacists’ comments. Analysis: Integrating new pharmacy services into established practice models is challenging. By taking time to assess successes and analyze areas of improvement, pharmacists can increase the utility and value they bring to health care teams.
90
Journal of the American Pharmacists Association
affiliated medical center (Spectrum Health HealthCare System, Grand Rapids, Mich.).
Objectives The objectives of this project were to determine the number and types of recommendations made by the pharmacists in an internal medicine residency clinic, the percentage of recommendations that were addressed and acted upon by the residents, whether changes needed to be made to the recommendation process, and areas in which clinic pharmacists’ services could be improved in the future.
Practice Description and Pharmacists’ Role The internal medicine residency clinic that was the setting for this project schedules patients in the afternoons Monday through Thursday. An average of 40 adult patients receive medical care in the clinic each afternoon. These cases are usually complex, with patients presenting with multiple disease states (e.g., a patient with concomitant diabetes, hypertension, hyperlipidemia, and heart failure). Six to eight medical residents evaluate patients each afternoon. All residents must discuss the patient’s case with an attending physician before completing the patient’s visit. During January through December 2002, two pharmacists provided new clinical services, one on Mondays and other on Thursdays. Both provided drug information, patient counseling, and resident education. Pharmacists spent time each morning before patients’ visits reviewing charts and making recommendations for more appropriate medication use. The pharmacists’ recommendations were based on current therapeutic guidelines and medically accepted standards of care, but difficulties were encountered in establishing a system that resulted in improvements in patients’ drug regimens. At the beginning of this project, pharmacists were placing their recommendations on a yellow form that was attached to the front of each patient’s chart (see Figure 1). Because this document was used for communication only and was not approved by the health system, it did not become a permanent part of the chart. Residents typically did not save the form, and they therefore would frequently not remember to address the recommendations later. The pharmacists tried discussing the recommendations with the medical residents during patient presentations to the attending physician, but this was not always possible because of the number of residents in the clinic and the other pharmacy services that were being provided.
Practice Analysis The clinic pharmacists kept records of their recommendations during 2002, their first year in the clinic. Each record contained the patient’s name, medical record number, date of birth, and date of serwww.japha.org
January/February 2004
Vol. 44, No. 1
Acceptance of Pharmacists’ Recommendations
EXPERIENCE
Figure 1. Pharmacist Recommendation Form in Use During Time Period of Analysis Response:
Patient Name: Medical Record No.: Pharmacist Name:
vice; medical resident’s name; and the type of recommendation made. These records were used to create the database for this project. Inclusion criteria for this analysis included cases for which pharmacists made recommendations in 2002 and patients who were present for the appointment on which the pharmacy recommendation was to be made. Records for patients who did not keep or who canceled the appointment for which a pharmacist’s recommendation was made and records for patients whose charts were unavailable for review were excluded from the analysis. After receiving approval from the Spectrum Health Institutional Review Board and a waiver of informed consent, the pharmacists currently working in the clinic conducted a retrospective review in March 2003 of the charts for patients for whom pharmacy recommendations were made in 2002. The charts were used to determine whether a resident addressed and acted upon the pharmacist’s recommendation and at which patient visit the recommendation was acted upon, if at all. Only information contained in patients’ charts was used for this project. We analyzed the data collected for this retrospective analysis to determine the total number of recommendations made, the number of each type of recommendation made, the percentage of overall recommendations addressed and acted upon by the residents, and the percentage of each type of recommendation addressed and acted upon by the resident. A recommendation was considered addressed if the resident explained on the chart what the recommendation was. A recommendation was considered acted upon if the resident explained why it was or was not accepted (e.g., “labs will be done next week” or “aspirin cannot be used because of a history of bleeding”). The outcomes for this project included information on the different types of recommendations made and accepted by the residents, changes that needed to be made to the recommendation process, and areas in which clinic pharmacists’ services could be improved in the future.
Results Seventy-four patient charts were initially evaluated. Thirteen charts were excluded (11 patients did not keep or canceled their physician’s appointment, and charts for 2 patients were unavailable for review). Therefore, 61 charts were available for analysis. Vol. 44, No. 1
January/February 2004
www.japha.org
The records contained 135 pharmacists’ recommendations for patients—72 (53.3%) therapeutic interventions and 63 (46.7%) recommendations for laboratory testing. Overall, 66 (49.0%) recommendations were addressed by the residents, while 69 (51.1%) were never addressed. Of the 66 pharmacists’ recommendations addressed, 59 (43.7%) were acted upon by the resident at the patient visit immediately following the making of the recommendation, 4 (3.0%) were in the resident’s dictation to be discussed at the patient’s next appointment and were acted upon, and 3 (2.2%) were in the resident’s dictation to be discussed at the patient’s next appointment and were not acted upon. The 59 recommendations acted upon at the patient’s current visit included 28 (47.5%) therapeutic interventions and 31 (52.5%) orders for labortory monitoring. Most recommendations dictated for future discussion involved laboratory tests. Tables 1 and 2 show the results of the therapeutic and laboratory interventions, including details on the 69 recommendations that were not addressed.
Discussion and Practice Innovation The overall rate of residents addressing and acting upon the pharmacist recommendation in our experience, 49.0%, is lower than rates identified in other published reports. A study by Lee et al.3 revealed 92.8% acceptance of the pharmacists’ recommendations. In more than 30.0% of these cases, the patient showed improvement in the medical problem after a pharmacist recommendation was received. Several reasons may explain why the acceptance rate seen in our analysis was lower than that in Lee et al.’s study. The latter study took place in a Veterans Affairs Medical Center where pharmacists’ recommendations were made in written form but also included in the patient’s electronic chart. This allowed the providers to review the recommendation at their convenience. In our clinic, the recommendation form was not part of the patient’s permanent record, so the resident could not review pharmacists’ recommendations later. The decision to address and act upon a recommendation had to be made at that patient visit. In some instances, a patient was seen by a different resident who may not have been comfortable addressing the particular pharmacist recommendation because he or she Journal of the American Pharmacists Association
91
EXPERIENCE
Acceptance of Pharmacists’ Recommendations
Table 1. Results of Pharmacists’ Recommendations for Therapeutic Interventions
Type of Recommendation
Acted Upon at Current Visit No. (%)
Change dose
Dictated to be Discussed at Next Visit and Acted Upon No. (%)
Dictated to be Discussed at Next Visit and Not Acted Upon No. (%)
Never Addressed No. (%)
Total No. (%)
11
(39.3)
0 (0)
0 (0)
11
(25.6)
22
Discontinue drug
3
(10.7)
0 (0)
0 (0)
5
(11.6)
8
(30.6) (11.1)
Add medication
2
(7.1)
0 (0)
0 (0)
12
(27.9)
14
(19.4)
Inhaler training
4
(14.3)
1 (100)
0 (0)
5
(11.6)
10
(13.9)
Other
8
(28.9)
0 (0)
0 (0)
10
(23.3)
18
(25)
(39)
1 (1)
0 (0)
43
(60)
72
(100)
Total
28
Table 2. Results of Pharmacists’ Recommendations for Laboratory Monitoring
Recommended Test
Acted Upon at Current Visit No. (%)
Dictated to be Discussed at Next Visit and Acted Upon No. (%)
Dictated to be Discussed at Next Visit and Not Acted Upon No. (%)
Never Addressed No. (%)
Total No. (%)
Glycosylated hemoglobin (A1c)
7
(22.6)
1 (33.3)
1
(33.3)
4
(15.4)
13
(20.6)
Urine microalbumin
3
(9.7)
0 (0)
0
(0)
6
(23.1)
9
(14.3)
Lipid profiles
11
(35.5)
1 (33.3)
0
(0)
5
(19.2)
17
Liver function enzymes
6
(19.4)
0 (0)
1
(33.3)
4
(15.4)
11
(17.5)
Basic metabolic panel
2
(6.5)
0 (0)
1
(33.3)
6
(23.1)
9
(14.3)
Other Total
2 31
(6.5)
1 (33.3)
0
(0)
1
(49)
3 (5)
3
(5)
26
did not know the patient. Other times, the appointment was too short to allow addressing a particular recommendation. Also, Lee et al. evaluated acceptance by providers and did not state whether those providers were attending physicians, residents, or other health care providers.3 Our study evaluated acceptance among only resident physicians. In our experience, many attending physicians and other providers are more familiar with pharmacists’ services and more receptive to recommendations from pharmacists than are medical residents. In a study by Blakey and Hixson-Wallace5 in a geriatric ambulatory care clinic, physicians accepted 98.6% of pharmacists’ recommendations. This study showed a decrease in the number of drugs used per patient, positive or neutral clinical outcomes based on the recommendations, and a cost savings to the clinic. Again, several reasons may explain the much lower acceptance rate in our analysis. In the previous study, the pharmacists’ recommendations became a permanent part of the physician’s chart and the pharmacists also discussed the recommendations with the physicians before they decided whether to accept the recommendation. In our study, pharmacists’ recommendations were not included in patients’ charts and were not discussed on a regular basis with residents before they evaluated the recommendations. The second difference was that the pharmacists in the prior study documented the results of their discussions with physicians, noting whether recommendations were accepted.5 In our experience, documentation of whether recommendations were addressed was left up to the resident, so many of the recommendations may have been considered but not documented. Hanlon et al.6 compared pharmacist-written recommendations provided to medical residents with usual care. Recommendations 92
Journal of the American Pharmacists Association
(3.8) (41)
4 63
(27)
(6.3) (100)
were written for both the intervention and usual care groups, but only the intervention group actually received the written recommendation. The recommendations for the usual care group were only used for chart review to determine whether the recommended changes would have been made by the residents without the pharmacists’ recommendation. The intervention group’s acceptance rate was similar to that in our experience. Resident physicians enacted pharmacists’ recommendations about one half of the time in the intervention group, compared with about one fourth of the time in the control group, a significant difference. The physicians in this prior study were highly satisfied with the pharmacists’ input and thought their advice was helpful. All three of these previous studies show that regardless of the number of recommendations accepted, a pharmacist can provide useful education and improvements in patient care. After reviewing the results of the above studies, we identified three flaws in our recommendation process: (1) no permanent documentation of recommendations was included in patients’ charts for future review by residents; (2) the pharmacists were not able to discuss all their recommendations with residents before they made treatment decisions; (3) patients’ charts did not include documentation as to whether recommendations were addressed and acted upon by the residents or the reasons why they were or were not acted upon by the resident physician. To address the first and third flaws, we developed a new recommendation form (see Figure 2). This new form includes a block for pharmacists’ recommendations and also spaces for residents to document whether they accepted the recommendations and to provide the reasons the recommendations were or were not implemented. The second flaw was harder to address because of the nature of www.japha.org
January/February 2004
Vol. 44, No. 1
Acceptance of Pharmacists’ Recommendations
EXPERIENCE
Figure 2. Pharmacist Recommendation Form Developed Based on Analysis of Results and Other Studies
Pharmacy Recommendation Patient Name:
Date of Recommendation:
Medical Record No.:
Physician Seeing Patient:
Recommendations:
Response:
1.
1. Accepted: ❑ Yes ❑ No If no, why not:
2.
2. Accepted: ❑ Yes ❑ No If no, why not:
3.
3. Accepted: ❑ Yes ❑ No If no, why not:
the clinic. During clinic hours, there may be as many as eight residents on duty, and a pharmacist cannot personally speak with every resident about every recommendation made. When discussing the new recommendation forms with the residents, we encouraged them to speak with the pharmacist about the recommendations. Evaluation of the recommendations not addressed by medical residents revealed several areas in which pharmacists’ services could be improved in the future. One area was resident education on treatment guidelines for several common diseases, such as diabetes, asthma, and heart failure, which may help increase the acceptance of recommendations made for laboratory monitoring and dose changes. Another area was the development of pharmacist-run disease clinics in which patients’ drug therapy could be monitored between appointments with medical residents.
Limitations Our analysis did not involve testable hypotheses or comparison groups, precluding statistical analysis. Because our analysis was retrospective, the percentage of addressed and acted upon recommendations may have actually been higher than is reported because of the lack of documentation in patients’ charts. The retrospective nature of this analysis also did not allow for collection of data on the number of times pharmacists talked with residents and whether recommendations were implemented based on such discussions. Many recommendations may have been discussed when the resident presented the patient’s case, but the resident did not document this discussion in the chart. Finally, patient outcomes related to the recommendations could not be determined.
Vol. 44, No. 1
January/February 2004
www.japha.org
Conclusion Our retrospective analysis showed that in 2002, the addition of pharmacists within an internal medicine residency clinic proved useful in patient care and physician education. Pharmacists made recommendations for 72 therapeutic and 63 laboratory interventions. Medical residents addressed and acted upon 66 (49.0%) of pharmacists’ recommendations. Improvements in the recommendation forms, resident education, and implementation of a pharmacist-managed disease state clinic should further enhance patient care at this practice site.
References 1. Farris KB, Kumbera P, Halterman T, Fang G. Outcomes-based pharmacist reimbursement: reimbursing pharmacists for cognitive services. J Managed Care Pharm. 2002;8:383–93. 2. Zunker RJ, Carlson DL. Economics of using pharmacists as advisors to physicians in risk-sharing contracts. Am J Health Syst Pharm. 2000;57:753–5. 3. Lee AJ, Boro MS, Knapp KK, et al. Clinical and economic outcomes of pharmacist recommendations in a Veterans Affairs medical center. Am J Health Syst Pharm. 2002;59:2070–7. 4. Schumock GT, Butler MG, Meek PD, et al. Evidence of the economic benefit of clinical pharmacy services: 1996–2000. Pharmacotherapy. 2003;23:113–32. 5. Blakey SA, Hixson-Wallace JA. Clinical and economic effects of pharmacy services in a geriatric ambulatory clinic. Pharmacotherapy. 2000;20:1198–203. 6. Hanlon JT, Weinberger M, Samsa G, et al. A randomized, controlled trial of a clinical pharmacist intervention to improve inappropriate prescribing in elderly outpatients with polypharmacy. Am J Med. 1996;100:428–37. 7. Campbell RK, Saulie BA. Providing pharmaceutical care in a physician office. J Am Pharm Assoc. 1998;38:495–9.
Journal of the American Pharmacists Association
93
Enhancing Pharmacists’ Recommendation Process in an Internal Medicine Resident Clinic Kimberly K. Daugherty and Karin Kangas
Received April 3, 2003, and in revised form June 3, 2003. Accepted for publication August 4, 2003.
ABSTRACT Objective: To determine the number and type of recommendations made by pharmacists in an internal medicine residency clinic, the percentage of recommendations that were addressed and acted upon by the residents, whether changes needed to be made to the recommendation process, and areas in which pharmacy services could be improved. Setting: Outpatient, urban, internal medicine clinic. Practice Description: The clinic operates Monday through Thursday afternoons. Each of these afternoons, six to eight medical residents evaluate patients. Pharmacists provide services on Monday and Thursdays. The pharmacists spend the mornings before patients’ visits reviewing charts and making recommendations for more appropriate medication use. Originally, pharmacists’ recommendations were listed on a form that was attached to the front of each patient’s chart but did not become a part of the medical record. At times, pharmacists made recommendations while a medical resident presented the patient’s case to the attending physician.
Kimberly K. Daugherty, PharmD, BCPS, is Assistant Professor of Pharmacy Practice, Ferris State University, Grand Rapids, Mich. Karin Kangas, PharmD, is Ambulatory Care Clinical Pharmacy Specialist, Spectrum Health, Grand Rapids, Mich. Correspondence: Kimberly K. Daugherty, PharmD, BCPS, 21 Michigan, NE, Suite 425, Grand Rapids, MI 49503. Fax: 616-391-3783. E-mail: [email protected] Disclosure: Kangas is employed by Spectrum Health HealthCare System, which is financially responsible for the internal medicine residency clinic used in this study. Daugherty works in the internal medicine clinic but receives no financial support from the health care system. The authors declare no other conflicts of interest or financial interests in any product or service mentioned in this article, including grants, employment, gifts, stock holdings, or honoraria.
Practice Innovation: Data for calendar year 2002 were analyzed retrospectively to identify a recommendation acceptance rate, areas of potential improvement, and actions that might enhance pharmacists’ effectiveness. Main Outcome Measures: Number of recommendations made, number of each type of recommendation made, percentage of overall recommendations addressed and acted upon by the residents, and percentage of each type of recommendation addressed and acted upon by the resident physician. Results: For 61 patients, 135 recommendations were made by pharmacists during 2002. Of these, 72 (53.3%) were therapeutic interventions and 63 (46.7%) were recommendations for laboratory monitoring. Overall, 66 (49.0%) recommendations were addressed and acted upon by the residents, while 69 (51.1%) were not addressed by the residents. Because of deficiencies identified in the communication process, a new form was developed that includes spaces for pharmacists’ recommendations and residents’ responses. Other potential future enhancements include providing more education for residents and implementing a pharmacist-run disease clinic. Conclusion: This retrospective analysis showed that pharmacists were able to identify many therapeutic and laboratory interventions, thus serving a useful role in patient care and physician education in an internal medicine residency clinic. Keywords: Pharmacy services, medical residents, internal medicine clinic, ambulatory care, pharmaceutical care. J Am Pharm Assoc. 2004;44:89–94. Vol. 44, No. 1
January/February 2004
www.japha.org
Journal of the American Pharmacists Association
89
EXPERIENCE
Acceptance of Pharmacists’ Recommendations
S
pending on prescription drugs has increased faster over the past 5 years than has spending on hospital and physician services.1 The number of new prescription medications approved each year, the increasing age of the U.S. population, use of direct-to-consumer advertising, and the aggressive marketing of medications to physicians by pharmaceutical manufacturers have all combined to help increase the spending on prescription medications.2,3 However, the prescribing of these medications is not always appropriate.1,3 Because of the need to use medications appropriately, many pharmacists have begun providing cognitive services. Cognitive services are defined as clinical activities intended to improve medication prescribing and use.1 Over the past 25 years, pharmacist-provided cognitive services have become a valuable component of many health care systems.4 Many authors have demonstrated the economic and clinical benefits of pharmacists’ services in a variety of clinic settings and for a range of disease states.1,3–5 The ambulatory care setting is optimal for the provision of pharmaceutical care and cognitive services because the pharmacist has access to the patient’s records and is able to review all of the patient’s medication requirements.7 The number of patient visits in the ambulatory care setting has increased over the past few years, but the number of primary care physicians has not.5 Because of the increased demand on their time, many primary care physicians find it difficult to stay abreast of current therapeutic options and to review complex drug regimens.2 One way of addressing this problem is to have pharmacists provide physician education and identify drug-related problems in the ambulatory care setting.2,6 As a result of the growing body of evidence indicating that pharmacists are effective in promoting appropriate medication use, pharmacists’ services were begun at one of the outpatient, urban, internal medicine residency clinics within a large nonuniversity
AT A GLANCE Synopsis: During 2002, pharmacists working two afternoons per week in an internal medicine clinic staffed by medical residents made 135 recommendations for changes in drug therapy or additional laboratory monitoring. About one half of these recommendations were acted upon by residents, somewhat lower than in some past published studies. After analyzing their experiences, authors of this article concluded that improvements were needed in the way they communicated with residents. A new form was developed that became a part of patients’ medical records and provided a place for residents to respond to pharmacists’ comments. Analysis: Integrating new pharmacy services into established practice models is challenging. By taking time to assess successes and analyze areas of improvement, pharmacists can increase the utility and value they bring to health care teams.
90
Journal of the American Pharmacists Association
affiliated medical center (Spectrum Health HealthCare System, Grand Rapids, Mich.).
Objectives The objectives of this project were to determine the number and types of recommendations made by the pharmacists in an internal medicine residency clinic, the percentage of recommendations that were addressed and acted upon by the residents, whether changes needed to be made to the recommendation process, and areas in which clinic pharmacists’ services could be improved in the future.
Practice Description and Pharmacists’ Role The internal medicine residency clinic that was the setting for this project schedules patients in the afternoons Monday through Thursday. An average of 40 adult patients receive medical care in the clinic each afternoon. These cases are usually complex, with patients presenting with multiple disease states (e.g., a patient with concomitant diabetes, hypertension, hyperlipidemia, and heart failure). Six to eight medical residents evaluate patients each afternoon. All residents must discuss the patient’s case with an attending physician before completing the patient’s visit. During January through December 2002, two pharmacists provided new clinical services, one on Mondays and other on Thursdays. Both provided drug information, patient counseling, and resident education. Pharmacists spent time each morning before patients’ visits reviewing charts and making recommendations for more appropriate medication use. The pharmacists’ recommendations were based on current therapeutic guidelines and medically accepted standards of care, but difficulties were encountered in establishing a system that resulted in improvements in patients’ drug regimens. At the beginning of this project, pharmacists were placing their recommendations on a yellow form that was attached to the front of each patient’s chart (see Figure 1). Because this document was used for communication only and was not approved by the health system, it did not become a permanent part of the chart. Residents typically did not save the form, and they therefore would frequently not remember to address the recommendations later. The pharmacists tried discussing the recommendations with the medical residents during patient presentations to the attending physician, but this was not always possible because of the number of residents in the clinic and the other pharmacy services that were being provided.
Practice Analysis The clinic pharmacists kept records of their recommendations during 2002, their first year in the clinic. Each record contained the patient’s name, medical record number, date of birth, and date of serwww.japha.org
January/February 2004
Vol. 44, No. 1
Acceptance of Pharmacists’ Recommendations
EXPERIENCE
Figure 1. Pharmacist Recommendation Form in Use During Time Period of Analysis Response:
Patient Name: Medical Record No.: Pharmacist Name:
vice; medical resident’s name; and the type of recommendation made. These records were used to create the database for this project. Inclusion criteria for this analysis included cases for which pharmacists made recommendations in 2002 and patients who were present for the appointment on which the pharmacy recommendation was to be made. Records for patients who did not keep or who canceled the appointment for which a pharmacist’s recommendation was made and records for patients whose charts were unavailable for review were excluded from the analysis. After receiving approval from the Spectrum Health Institutional Review Board and a waiver of informed consent, the pharmacists currently working in the clinic conducted a retrospective review in March 2003 of the charts for patients for whom pharmacy recommendations were made in 2002. The charts were used to determine whether a resident addressed and acted upon the pharmacist’s recommendation and at which patient visit the recommendation was acted upon, if at all. Only information contained in patients’ charts was used for this project. We analyzed the data collected for this retrospective analysis to determine the total number of recommendations made, the number of each type of recommendation made, the percentage of overall recommendations addressed and acted upon by the residents, and the percentage of each type of recommendation addressed and acted upon by the resident. A recommendation was considered addressed if the resident explained on the chart what the recommendation was. A recommendation was considered acted upon if the resident explained why it was or was not accepted (e.g., “labs will be done next week” or “aspirin cannot be used because of a history of bleeding”). The outcomes for this project included information on the different types of recommendations made and accepted by the residents, changes that needed to be made to the recommendation process, and areas in which clinic pharmacists’ services could be improved in the future.
Results Seventy-four patient charts were initially evaluated. Thirteen charts were excluded (11 patients did not keep or canceled their physician’s appointment, and charts for 2 patients were unavailable for review). Therefore, 61 charts were available for analysis. Vol. 44, No. 1
January/February 2004
www.japha.org
The records contained 135 pharmacists’ recommendations for patients—72 (53.3%) therapeutic interventions and 63 (46.7%) recommendations for laboratory testing. Overall, 66 (49.0%) recommendations were addressed by the residents, while 69 (51.1%) were never addressed. Of the 66 pharmacists’ recommendations addressed, 59 (43.7%) were acted upon by the resident at the patient visit immediately following the making of the recommendation, 4 (3.0%) were in the resident’s dictation to be discussed at the patient’s next appointment and were acted upon, and 3 (2.2%) were in the resident’s dictation to be discussed at the patient’s next appointment and were not acted upon. The 59 recommendations acted upon at the patient’s current visit included 28 (47.5%) therapeutic interventions and 31 (52.5%) orders for labortory monitoring. Most recommendations dictated for future discussion involved laboratory tests. Tables 1 and 2 show the results of the therapeutic and laboratory interventions, including details on the 69 recommendations that were not addressed.
Discussion and Practice Innovation The overall rate of residents addressing and acting upon the pharmacist recommendation in our experience, 49.0%, is lower than rates identified in other published reports. A study by Lee et al.3 revealed 92.8% acceptance of the pharmacists’ recommendations. In more than 30.0% of these cases, the patient showed improvement in the medical problem after a pharmacist recommendation was received. Several reasons may explain why the acceptance rate seen in our analysis was lower than that in Lee et al.’s study. The latter study took place in a Veterans Affairs Medical Center where pharmacists’ recommendations were made in written form but also included in the patient’s electronic chart. This allowed the providers to review the recommendation at their convenience. In our clinic, the recommendation form was not part of the patient’s permanent record, so the resident could not review pharmacists’ recommendations later. The decision to address and act upon a recommendation had to be made at that patient visit. In some instances, a patient was seen by a different resident who may not have been comfortable addressing the particular pharmacist recommendation because he or she Journal of the American Pharmacists Association
91
EXPERIENCE
Acceptance of Pharmacists’ Recommendations
Table 1. Results of Pharmacists’ Recommendations for Therapeutic Interventions
Type of Recommendation
Acted Upon at Current Visit No. (%)
Change dose
Dictated to be Discussed at Next Visit and Acted Upon No. (%)
Dictated to be Discussed at Next Visit and Not Acted Upon No. (%)
Never Addressed No. (%)
Total No. (%)
11
(39.3)
0 (0)
0 (0)
11
(25.6)
22
Discontinue drug
3
(10.7)
0 (0)
0 (0)
5
(11.6)
8
(30.6) (11.1)
Add medication
2
(7.1)
0 (0)
0 (0)
12
(27.9)
14
(19.4)
Inhaler training
4
(14.3)
1 (100)
0 (0)
5
(11.6)
10
(13.9)
Other
8
(28.9)
0 (0)
0 (0)
10
(23.3)
18
(25)
(39)
1 (1)
0 (0)
43
(60)
72
(100)
Total
28
Table 2. Results of Pharmacists’ Recommendations for Laboratory Monitoring
Recommended Test
Acted Upon at Current Visit No. (%)
Dictated to be Discussed at Next Visit and Acted Upon No. (%)
Dictated to be Discussed at Next Visit and Not Acted Upon No. (%)
Never Addressed No. (%)
Total No. (%)
Glycosylated hemoglobin (A1c)
7
(22.6)
1 (33.3)
1
(33.3)
4
(15.4)
13
(20.6)
Urine microalbumin
3
(9.7)
0 (0)
0
(0)
6
(23.1)
9
(14.3)
Lipid profiles
11
(35.5)
1 (33.3)
0
(0)
5
(19.2)
17
Liver function enzymes
6
(19.4)
0 (0)
1
(33.3)
4
(15.4)
11
(17.5)
Basic metabolic panel
2
(6.5)
0 (0)
1
(33.3)
6
(23.1)
9
(14.3)
Other Total
2 31
(6.5)
1 (33.3)
0
(0)
1
(49)
3 (5)
3
(5)
26
did not know the patient. Other times, the appointment was too short to allow addressing a particular recommendation. Also, Lee et al. evaluated acceptance by providers and did not state whether those providers were attending physicians, residents, or other health care providers.3 Our study evaluated acceptance among only resident physicians. In our experience, many attending physicians and other providers are more familiar with pharmacists’ services and more receptive to recommendations from pharmacists than are medical residents. In a study by Blakey and Hixson-Wallace5 in a geriatric ambulatory care clinic, physicians accepted 98.6% of pharmacists’ recommendations. This study showed a decrease in the number of drugs used per patient, positive or neutral clinical outcomes based on the recommendations, and a cost savings to the clinic. Again, several reasons may explain the much lower acceptance rate in our analysis. In the previous study, the pharmacists’ recommendations became a permanent part of the physician’s chart and the pharmacists also discussed the recommendations with the physicians before they decided whether to accept the recommendation. In our study, pharmacists’ recommendations were not included in patients’ charts and were not discussed on a regular basis with residents before they evaluated the recommendations. The second difference was that the pharmacists in the prior study documented the results of their discussions with physicians, noting whether recommendations were accepted.5 In our experience, documentation of whether recommendations were addressed was left up to the resident, so many of the recommendations may have been considered but not documented. Hanlon et al.6 compared pharmacist-written recommendations provided to medical residents with usual care. Recommendations 92
Journal of the American Pharmacists Association
(3.8) (41)
4 63
(27)
(6.3) (100)
were written for both the intervention and usual care groups, but only the intervention group actually received the written recommendation. The recommendations for the usual care group were only used for chart review to determine whether the recommended changes would have been made by the residents without the pharmacists’ recommendation. The intervention group’s acceptance rate was similar to that in our experience. Resident physicians enacted pharmacists’ recommendations about one half of the time in the intervention group, compared with about one fourth of the time in the control group, a significant difference. The physicians in this prior study were highly satisfied with the pharmacists’ input and thought their advice was helpful. All three of these previous studies show that regardless of the number of recommendations accepted, a pharmacist can provide useful education and improvements in patient care. After reviewing the results of the above studies, we identified three flaws in our recommendation process: (1) no permanent documentation of recommendations was included in patients’ charts for future review by residents; (2) the pharmacists were not able to discuss all their recommendations with residents before they made treatment decisions; (3) patients’ charts did not include documentation as to whether recommendations were addressed and acted upon by the residents or the reasons why they were or were not acted upon by the resident physician. To address the first and third flaws, we developed a new recommendation form (see Figure 2). This new form includes a block for pharmacists’ recommendations and also spaces for residents to document whether they accepted the recommendations and to provide the reasons the recommendations were or were not implemented. The second flaw was harder to address because of the nature of www.japha.org
January/February 2004
Vol. 44, No. 1
Acceptance of Pharmacists’ Recommendations
EXPERIENCE
Figure 2. Pharmacist Recommendation Form Developed Based on Analysis of Results and Other Studies
Pharmacy Recommendation Patient Name:
Date of Recommendation:
Medical Record No.:
Physician Seeing Patient:
Recommendations:
Response:
1.
1. Accepted: ❑ Yes ❑ No If no, why not:
2.
2. Accepted: ❑ Yes ❑ No If no, why not:
3.
3. Accepted: ❑ Yes ❑ No If no, why not:
the clinic. During clinic hours, there may be as many as eight residents on duty, and a pharmacist cannot personally speak with every resident about every recommendation made. When discussing the new recommendation forms with the residents, we encouraged them to speak with the pharmacist about the recommendations. Evaluation of the recommendations not addressed by medical residents revealed several areas in which pharmacists’ services could be improved in the future. One area was resident education on treatment guidelines for several common diseases, such as diabetes, asthma, and heart failure, which may help increase the acceptance of recommendations made for laboratory monitoring and dose changes. Another area was the development of pharmacist-run disease clinics in which patients’ drug therapy could be monitored between appointments with medical residents.
Limitations Our analysis did not involve testable hypotheses or comparison groups, precluding statistical analysis. Because our analysis was retrospective, the percentage of addressed and acted upon recommendations may have actually been higher than is reported because of the lack of documentation in patients’ charts. The retrospective nature of this analysis also did not allow for collection of data on the number of times pharmacists talked with residents and whether recommendations were implemented based on such discussions. Many recommendations may have been discussed when the resident presented the patient’s case, but the resident did not document this discussion in the chart. Finally, patient outcomes related to the recommendations could not be determined.
Vol. 44, No. 1
January/February 2004
www.japha.org
Conclusion Our retrospective analysis showed that in 2002, the addition of pharmacists within an internal medicine residency clinic proved useful in patient care and physician education. Pharmacists made recommendations for 72 therapeutic and 63 laboratory interventions. Medical residents addressed and acted upon 66 (49.0%) of pharmacists’ recommendations. Improvements in the recommendation forms, resident education, and implementation of a pharmacist-managed disease state clinic should further enhance patient care at this practice site.
References 1. Farris KB, Kumbera P, Halterman T, Fang G. Outcomes-based pharmacist reimbursement: reimbursing pharmacists for cognitive services. J Managed Care Pharm. 2002;8:383–93. 2. Zunker RJ, Carlson DL. Economics of using pharmacists as advisors to physicians in risk-sharing contracts. Am J Health Syst Pharm. 2000;57:753–5. 3. Lee AJ, Boro MS, Knapp KK, et al. Clinical and economic outcomes of pharmacist recommendations in a Veterans Affairs medical center. Am J Health Syst Pharm. 2002;59:2070–7. 4. Schumock GT, Butler MG, Meek PD, et al. Evidence of the economic benefit of clinical pharmacy services: 1996–2000. Pharmacotherapy. 2003;23:113–32. 5. Blakey SA, Hixson-Wallace JA. Clinical and economic effects of pharmacy services in a geriatric ambulatory clinic. Pharmacotherapy. 2000;20:1198–203. 6. Hanlon JT, Weinberger M, Samsa G, et al. A randomized, controlled trial of a clinical pharmacist intervention to improve inappropriate prescribing in elderly outpatients with polypharmacy. Am J Med. 1996;100:428–37. 7. Campbell RK, Saulie BA. Providing pharmaceutical care in a physician office. J Am Pharm Assoc. 1998;38:495–9.
Journal of the American Pharmacists Association
93