PHARMACOECONOMICS Edited by William F. McGhan, J. Lyle Bootman, and Raymond J. Townsend
ENALAPRIL TO LISINOPRIL: ECONOMIC IMPACT OF A VOLUNTARY ANGIOTENSIN-CONVERTING ENZYME-INHIBITOR SUBSTITUTION PROGRAM IN A STAFF-MODEL HEALTH MAINTENANCE ORGANIZATION Kimberly P. McDonough, Ross H. Weaver, and Gary D. Viall
The cost-effectiveness of a voluntary program that switched enalapril to lisinopriltherapy in patients with benign essential hypertensionin a staff-modelhealth maintenance organization(HMO) was evaluated. DESIGN: The one-year nonrandomized,controlled trial was performed from November 1989 through October 1990. PARTIOPANTS: One hundred twenty-sevenpatients were entered into the study: 75 who converted from enalapril to lisinopriland 52 who remained on enalapril throughout the study period. Patients were excluded from analysis because of diagnosis (not benign essential hypertension)or insufficientdata collection.
INTERVENTIONS: Patients taking enalapril were asked by staff pharmacists if they were willing to consider switching from enalapril to lisinopril.To encourage patients, the HMO agreed to waive the drug rider copayrnent for three months. If patients were willing,their physicians were contacted and they established the lisinoprildosage. MAIN OUTCOME MEASURES: Total direct cost and savings resulting from converting patients from enalapril to lisinopril were measured and compared with costs of therapy for patients who remained on enalapril. RESULTS: The control and study groups were evenly matched according to demographics and concomitant drug therapy. Drug acquisitioncosts, costs associated with waiving drug rider copayrnent,pharmacy administrativecosts, costs of managing adverse events, costs of visits to physicians,and laboratorytest costs were assessed. Depending on the cost of capital assumed, net
KIMBERL Y P. MCDONOUGH, Phann.D., is the Clinical Coordinator, Harvard Community Health Plan of New England, One Hoppin St., Providence, RI 02903; ROSS H, WEAVER, Phann.D., M.B.A., is the Manager, New Products Marketing Research, ICI Pharmaceuticals Group, ICI Americas, Inc., Wilmington, DE; and GARY D. VIALL, B.S., is the Pharmacy Supervisor, Harvard Community Health Plan of New England. Reprinls: Kimberly P. McDonough, Phann.D. WILLIAM F. MCGHAN, Pharm.D., Ph.D., is a Professor and the Executive Director, Institute for Pharmaceutical Economics, and the Chairman, Department of Pharmacy Practice and Pharmacy Administration, Philadelphia College of Pharrnacy and Science, Philadelphia, PA; J. LYLE BooTMAN, Ph.D., is a Professor and the Dean, College of Pharmacy, University of Arizona, Tucson, AZ 85721; RAYMOND J. TOWNSEND, Phann.D., is the Vice President, Applied Healthcare Research, Glaxo Inc., Research Triangle Park, NC 27709.
savings ranged from $85 to $110 per patient converted from enalapril to lisinopril. Monthly net savings that ranged from $2.04 to $2.61 per patient were required to result in overall net savings within the first two years. In a regular practice setting, a net savings is realized in less than 12 months when patients are converted from enalapril to lisinoprilfor treatment of benign essential hypertension. The voluntary therapeuticinterchangeprogram provided a good means for achieving cost controls for pharmacy expenses.
Ann Pharmacother 1992;26:399-404, HOSPITALS HAVE USEDTHE CONCEPT of therapeutic substitu-
tion as a method for controlling pharmacy expenses for many years.'? Likewise, managed-care organizations are pursuing this avenue as a potential source of savings.' This article describes the cost-effectiveness of such a program in a staff-model health maintenance organization (HMO). The Harvard Community Health Plan of New England (HCHP/NE) is a 76 ODD-member staff-model HMO that serves members in Rhode Island and southeastern Massachusetts. The organization comprises five healthcare centers, each with a full-service pharmacy department. Each pharmacy is staffed with registered pharmacists and technicians who are employees of the HMO. Pharmacy service is available for 12 hours on weekdays and 4 hours on Saturdays and holidays. Members must use HCHP/NE pharmacies for all prescriptions that are filled during usual operational hours. Angiotensin-converting enzyme (ACE) inhibitors were selected for therapeutic substitution for several reasons. This class is recognized as a suitable first-line agent for the treatment of hypertension in many patients.v Use of ACE inhibitors has steadily increased at HCHP/NE over the past two years, with enalapril being the most widely prescribed agent. Both enalapril and lisinopril are very similar in therapeutic effectiveness when treating hypertension.v" however, contractual prices for lisinopril suggested that substantial savings could be achieved by preferential use of
The Annals ofPharmacotherapy
1992 March, Volume 26 •
this product. With these factors in mind, the pharmacy and therapeutics committee approved a program to convert patients from enalapril to lisinopril therapy. The effort to encourage conversion between these products was directed toward HMO members receiving enalapril. Previous efforts directed at the HCHP/NE medical staff resulted in only minimal changes in long-term prescribing. It was thought that involving patients in the conversion process would increase the number of participants. Patients were solicited by staff pharmacists when obtaining refills of their enalapril prescription. To encourage members in this process, the HMO agreed to waive the drug rider copayment ($2, $3, or $5 depending on the policy) for a period of three months for participants. Patients without a drug rider were given reduced prices, reflecting contractual savings, for their lisinopril prescriptions. The actual change in drug therapy required approval of the patient's medical provider. The medication dosing and adequacy of blood pressure (BP) control was left to the discretion of the physician as would be the case in a typical outpatient setting. To determine the success of this program, several goals were established for evaluation. The first goal was to quantify the total economic savings or costs resulting from the program. Although a savings of pharmacy acquisition costs could easily be achieved, these savings may have been offset by increased use of services elsewhere in the organization. Therefore, determining the total economic impact of this program was essential. The second goal was to assess the effectiveness of involving patients as a motivating factor for changing prescribing habits. The final goal was to determine comparable dose effectiveness between the products. Unfortunately, we were unable to determine comparable dose effectiveness in this study.
rates and pay-back timeframes. The one-tailed r-test was used for each discount rate to determine whether the money saved was statistically different from zero, and chi-square analysis evaluated whether the conversation resulted in savings rather than losses for the majority of patients. Analysis of the effectiveness of the patient-oriented approach to conversion from enalapril to lisinopril was made by prospectively tracking product utilization throughout the organization following implementation of the conversion program. Dose requirements and relative BP control of patients who chose to convert to lisinopril were recorded for the purpose of evaluating comparable doses between products. Patients were excluded from this comparison if they had not been controlled on enalapril for three months prior to conversion or if they did not achieve adequate BP control for three months after the conversion. BP control was defined as having at least three of the last four diastolic BP readings ~95 mm Hg. If fewer than four BP readings were available, the last two consecutive diastolic pressures were required to be ~95 mm Hg.
The control and study groups were evenly matched according to age, sex, smoking history, and concurrent drug therapy (Table 2). Of the 75 patients in the converting group, 7 were excluded because of the following reasons: congestive heart failure (2 patients), Grave's disease 0), active cocaine abuse 0), and insufficient charting data (3). Seven patients experienced adverse drug events and were excluded from the clinical use evaluation. Two other patients experienced adverse effects, but elected to continue therapy. The cost of managing all adverse reactions was included in the financial analysis. The remaining 61 charts were evaluated for clinical information. A total of 52 paTable 1. Patient Costs and Savings Associated with Enalapril-to-Lisinopril Conversion Program Costs clinic visits laboratory pharmacy administrative fixed variable adverse reactions waived copayment Savings drug expenses (monthly)
Methods A total of 125 patients were chosen for evaluation, 75 of whom participated in the conversion program; 50 chose to remain on enalapril. These patients were randomly selected from computerized records of pharmacy dispensing activities. The conversion program was initiated in May 1990, and data were collected from November 1989 through October 1990. Evaluations were made based on retrospective review of each patient's medical record by the investigators. Demographic information such as age, sex, and smoking history was obtained for each patient. Patient records were evaluated for BP control, dose of ACE inhibitor, and the presence of combination therapies. The frequency of office visits and laboratory testing was determined from six months prior to conversion through six months after conversion. This information was compared with records of patients who did not change therapy during a comparable 12-month period. Patients were excluded from the study if they were treated with an ACE inhibitor for reasons other than hypertension or if a concurrent illness would directly affect BP control. Patients who were treated with other antihypertensive medications were permitted in the study, provided that there were no additions, deletions, or dosage changes to these therapies during the study period. Economic evaluations were based on a comparison of the costs associated with implementing this program with the savings achieved through reduced drug acquisition prices. Costs associated with the conversion included additional laboratory monitoring, office visits, management of adverse drug reactions, lost revenue from the copayment waiver, and pharmacy administrative costs (Table I). These values reflect the actual costs to our institution and differ for other organizations. In order to adequately consider the time value of money, savings and costs were discounted at various rates (5, 10. and IS percent annually) and net present values were determined for 12,24, and 60 months." Various discount rates and timeframes were employed to allow other managed-care settings to interpret the findings of this study, using their settings' discount
The Annals ofPharmacotherapy
$28.00 0.00 5.61 1.40 4.43 8.67 4.32"
"Drug savings depend on a variety of factors, including dosages required to adequately control blood pressure, the number of patients receiving enalapril and lisinopril more than once daily, and product acquisition prices.
Table 2. Patient Demographics STUDY GROUP Number evaluated Exclusions Number for analysis Men Women Age (men) Age (women) Smokers Nonsmokers Unknown smoking history BP controlled on enalapril BP not controlled on enalapril Unknown BP control on enalapril BP =blood pressure.
1992 March, Volume 26
39 (64%) 22 (36%) 28-81 Y 41-71 Y 24 (39%) 29 (48%) 8 (13%) 27 (44%) 32 (53%) 2( 3%)
CONTROL GROUP 52 I 51 31 (61%) 20 (39%) 34-76 Y 36-75 y 17 (33%) 24 (47%)
10 (20%) 35 (69%) 9 (17%) 7 (14%)
tients were evaluated in the control (nonconverting) group, with one exclusion because of limited charting data. The frequency of office visits and laboratory tests was measured to assess any changes as a result of this program. With few exceptions, the therapy was converted from enalapril to lisinopril during an office visit. This visit was counted as a preconversion office encounter. The frequency of office visits averaged 2.85 per patient in the six months prior to conversion compared with 1.69 after changing therapy. From this information, it was assumed that an office visit was scheduled as a result of the conversion program. The total number of office visits averaged 4.54 per patient in the converting group during the 12-month study period compared with 2.6 office visits per patient in the control group. This difference was attributed to the fact that patients in the converting group were less likely to have controlled BP while on enalapril than the patients in the control group. Induction of a single office visit per conversion was substantiated through chart reviews. The frequency of laboratory tests was also greater during the preconversion period (1.95 per patient) compared with the six-month period after medication change (1.09 per patient). However, this incidence was comparable with findings in the control group. We noticed no causal relationship between the data of conversion and the ordering of laboratory monitors. From these results it was concluded that additional laboratory tests were not conducted as a result of the conversion program. A total of seven patients (nine percent) experienced adverse events resulting in cessation of therapy. Lisinopril was discontinued in this group and all patients returned to their previous treatment with enalapril. All adverse events were minor and resolved following the cessation of therapy (Table 3). Two additional patients experienced possible adverse effects, but both elected to continue lisinopril therapy. To determine the overall financial benefit of this program we compared savings from reduced drug acquisition expenses with the costs incurred by the program. Costs associated with this process included the additional office visit with a provider, pharmacy administrative costs, adverse reactions management, and waived copayment (Table 1). Savings were achieved through reduced medication acquisition costs. Pharmacy administrative costs were separated into two functions: fixed and variable. Fixed costs were associated with the time and supplies for the clinical pharmacist to set up the program prior to initiation. This figure was calculated to be $1020. Because these costs are fixed, the cost per patient would be less as more individuals participate in the program. Variable costs reflect the time that the dispensing pharmacist spent explaining the program to patients. This was calculated to be $1.40 per 3.5-minute encounter, based on the average pharmacist's salary and benefit costs to our organization. Monthly copayments ranged from $2 to $5 per patient, depending on the insurance policy, and were waived for three months. In addition, there were some patients who did not have drug riders and paid in full for their prescriptions. The average amount of copayment waived per person was $2.89 per month or $8.67 per conversion. Costs for managing adverse events were associated with an additional office visit, discarded medication, and copay-
ment loss. These costs were minimally offset by medication savings for the limited duration of lisinopril therapy. The total cost for managing these seven patients was $270.30, or $4.43 per converted patient (Table 4). Savings associated with the conversion process were determined on a patient-by-patient basis. Because the conversion process was left to the discretion of the provider, there was a wide variability in dose selection through the change of medication. In some cases, actual doses were reduced, some converted on a milligram-for-milligram basis, and still others resulted in increased doses. The costs of enalapril therapy were compared with the cost of therapy with lisinopril and a net saving or loss was calculated for each patient. These amounts are summarized for the entire group to show an average monthly savings of $4.32 per patient. In the first five months of the program, while 68 patients were evaluated in this trial, a total of 182 patients at HCHP/ NE converted from enalapril to lisinopril therapy. By applying the average costs and savings found in the sample group to the entire population, the cost benefit of the program was determined. All costs and expenses were analyzed on a spreadsheet program on a personal computer. Most of the costs associated with this program were incurred in the first several months, and the savings are generated over a longer period of time. For this reason, we analyzed the cost benefit of the program over a 6O-month period and discounted the savings at 5, 10, and 15 percent to account for the effects of inflation on the net present value of these dollars. Regardless of the discount rate used, we found that this program generated an overall savings to our organization within the first year. Net savings varied depending on the discounts applied and ranged from $85 to $110 per patient over a 24-month period (Figure 1). These savings are particularly significant when one considers the number of patients involved in the program. By knowing the total costs incurred by this program, we were able to calculate the savings in drug acquisition prices Table 3. Incidence of Adverse Reactions Following Conversion of Enalapril to Lisinopril" Headache Lightheadedness Constipation Depression Elevated cholesterol" Impotence Nausea
2 2 I I I 2 I
'Some patients experienced more than one adverse effect. bPatient was also receiving a thiazide diuretic.
Table 4. Cost of Adverse Drug Reactions Cost of lisinopril that was discarded when the patients discontinued therapy secondary to an adverse event Drug savings for the duration of lisinopril use instead of enalapril (the average patient took lisinopril for 14 days prior to resuming enalapril) Office visits for evaluation of adverse reactions Lost copayment waived TOTAL COST
The Annals ofPharmacotherapy •
196.00 21.00 $270.30
1992 March, Volume 26 • 401
that were needed to achieve a positive cost benefit for this program over a 12-month period. This was calculated by using a spreadsheet program on a personal computer and applying various discounted drug savings to the expenses of the program. This information is useful when deciding the benefit of initiating this program and in contractual negotiations. Based on our fmdings, a net monthly savings in drug acquisition prices ranging from $3.63 to $3.98 per patient, depending on discount rate applied, was necessary to result in a net savings within the first year (Table 5). Any savings in drug acquisition beyond these values provides additional fmancial benefits to the organization. As shown in Table 6, statistically significant savings were noted at 24 months and beyond, even when annual discount rates of up to 15 percent were considered. Although HCHP/NE actually saved money at 12 months because of the conversion, the savings did not reach statistical significance at any discount rate. Similar fmdings were seen with chi-square analysis, which evaluated whether the conversion resulted in savings rather than losses in more patients. For the timeframe of 24 months and beyond, the conversion resulted in statistically significantly more savings for HCHP/NE, rather than losses. The second objective of the study was to evaluate the effectiveness of involving patients in the decision-making process as a means to increase conversion from enalapril to lisinopril. During the first five months of the program, pharmacy records show that a total of 182 of the patients receiving enalapril (30 percent) voluntarily participated in the conversion process. This process had the added benefit of influencing new prescription generation. Despite the fact that the total number of ACE-inhibitor prescriptions did not change appreciably during the study period, a 50 percent change in the drug-prescribing pattern occurred (Figure 2). Unfortunately, we were unable to assess comparative dosing information between enalapril and lisinopril. Of the 75 patients studied, controlled BP was demonstrable prior
to conversion in only 27 individuals (36 percent). Of these patients, only 14 established controlled hypertension on lisinopril in the six months following conversion. This fmding was not related to enalapril or lisinopril effectiveness, but rather to an insufficient number of BP readings to establish control of hypertension as defined by the study protocol. This is based primarily on the fact that in an uncontrolled ambulatory setting, patients often visit providers on a sporadic basis, resulting in limited opportunities for BP monitoring.
Discussion This study demonstrates that, in a regular practice setting' a net savings may be seen in less than 12 months when patients are converted from enalapril to lisinopril. These findings include all direct costs, including physician/nurse office visits, laboratory tests, adverse-effect management, drug waste, pharmacy fixed and variable costs, and the time value of money. If extended to 24 months, HCHP/NE received a net savings of $85-110 per patient, depending on the discount rate used. These estimates are conservative, and HCHP/NE ultimately will receive greater savings per patient as a result of this program. Only a sample of patients who ultimately converted from enalapril to lisinopril are profiled in this study. Because fixed pharmacy administrative costs are a Table 5. Monthly Drug Savings Needed to Achieve Cost-Effectiveness over Two Years DIFFERENTIAL COST
DISCOUNT RATE (%)
0 5 10 IS
$3.63 3.79 3.88 3.98
$2.04 2.26 2.44 2.61
Table 6. Statistical Analysis of Savings Resulting from Conversion from Enalapril to Lisinopril over TIme, Using Various Discount Rates
240 DISCOUNT RATES
12-Month mean savings" t-value b
.!! 120 "0
P 24-Month mean savings t-value
P 36-Month mean savings r-value
Figure I. The discounted average savings resulting from the conversion from enalapril to lisinopril are shown. Using discount rates ranging from 0 to 15 percent annually, the conversion had a positive net present value in less than 12 months, and ranged from $150 to $220 per patient converted in five years.
P 60-Month mean savings t-value
8.25 0.80 NS
6.40 0.70 NS
5.21 0.58 NS
4.07 0.47 NS
$ 54.76 2.92