Economic analysis of a medication assistance program for ambulatory care patients at an urban teaching clinic

Objective: Pharmacy-managed medication assistance programs (MAP) have the potential to recoup losses incurred by the pharmacy department, but whether this offsets the personnel cost of the program has not been well-established. The purpose of this study was to conduct a cost-benefit analysis of the pharmacy-managed MAP at an urban teaching clinic at Hartford Hospital in Hartford, Conn.

Methods: Patients were enrolled into a MAP at Hartford Hospital, a large urban teaching hospital, based on referrals from social workers. Patients were assisted in identifying pharmaceutical manufacturers providing free drugs and completing applications. The program's benefit was defined as the value of the drugs procured using average wholesale price (AWP) in 2005 US dollars. The costs to the pharmacy consisted of personnel cost (time administering the program times a wage rate of $18.02/h). The net benefit was calculated as the value of the drugs procured minus personnel costs. Threshold sensitivity analysis and Monte Carlo simulation were performed by varying drug value, personnel time, and wage rate.

Results: Over the 24-month period (May 2003 through April 2005), 143 patients were enrolled into the program, with 328 medication shipments received. The total benefit for all drugs was $83,504. An average of 4 h/wk were spent administering the program, resulting in personnel costs of $7,496. The program yielded a net benefit of $76,008. Upon sensitivity analysis, the break-even point was reached when the drug value was reduced to $7,496, personnel time increased to 4,634 hours (44.5 h/wk), or the wage rate increased to $200/h. When the value of the drug procured, personnel time, and wage rate were simultaneously varied within plausible ranges ($68,175–$171,584, 2 h/wk–40 h/wk, and $13.65/h–$43.60/h, respectively), Monte Carlo simulation demonstrated there was a 96.1% chance of the program being cost-beneficial.

For many patients, paying for prescription drugs is an ever-increasing financial burden. Patients' inability to afford their drugs may limit adherence to prescribed therapies and can negatively affect disease management. The most recent available data estimate that approximately 45 million Americans are uninsured and as a result incur nearly $41 billion in uncompensated care.1

Pharmaceutical manufacturer assistance programs have evolved to provide uninsured patients with supplies of medication at no cost. As an illustration of the resources available, in 2004, industry-sponsored assistance programs provided 22 million prescriptions free-of-charge to indigent patients at a value of more than $4 billion.² Eligibility criteria and enrollment requirements to obtain free drugs vary, but generally include documentation of limited income, lack of prescription drug coverage, and ineligibility for other forms of both public and private assistance.

Due to the large number of uninsured patients at Hartford Hospital's ambulatory care clinic, the pharmacy department initiated a MAP to provide free medications to qualified patients. As resources are scarce, this powerful yet simple pharmacoeconomic methodology is required to justify an outpatient MAP. A cost-benefit analysis was conducted to determine the financial benefit to the institution, as well as determine the resources required to administer the program.

PROGRAM DESCRIPTION

Hartford Hospital's pharmacy-managed MAP was established in cooperation with University of Connecticut's School of Pharmacy in August 2002. The program is managed through the joint efforts of a full-time university faculty member and the Department of Pharmacy Services at Hartford Hospital, which is a large, urban teaching hospital. The MAP program is administered by pharmacy research fellows on a revolving basis. MAP appointments are available to Hartford Hospital outpatients seen at the adult primary care clinics once a week for 3 hours (Wednesdays from 1 pm–4 pm).

Patients are referred to the program through the outpatient social services department. Upon referral, patients are interviewed to obtain demographic, medication, financial, and insurance information and are assisted by pharmacy research fellows in completing and submitting applications (usually via US mail) to pharmaceutical manufacturers. Only medications for chronic disease states such as hypertension, diabetes, and asthma are sought. Physicians are consulted to discuss the patients' current and future drug needs and to complete their sections of applications.

Manufacturer approval of applications and the shipping of drugs generally takes 4 to 6 weeks. To address this lag time in receiving drugs, those patients who are thought to be eligible are immediately authorized to receive the medication(s) they have applied for free-of-charge from the institution's outpatient pharmacy. Before the initiation of this pharmacy-managed MAP, indigent patients were billed directly for drugs utilized as outpatients. This did not usually result in the Department of Pharmacy Services being compensated for drugs dispensed. Therefore, patients are assisted in applying for drugs to replace those that otherwise would have been dispensed to the patients by the outpatient pharmacy.

Patients are generally approved for 3-month intervals by manufacturer programs. After initial patient enrollment into Hartford Hospital's MAP, patients are scheduled to return to the program office quarterly to be reassessed for eligibility, to identify newly prescribed and eligible medications, and to complete renewal applications and update financial documentation. Patients are contacted via telephone to set up renewal appointments or when additional information is required for manufacturer approval.

METHODS

A cost-benefit analysis was conducted from the hospital perspective. The net benefit was calculated by summing the value of all benefits as well as costs associated with the administration of the MAP. The costs were then subtracted from the benefits to calculate the net benefit:

Net Benefit = Σ Benefits – Σ Costs

COSTS

The costs of this MAP consisted of direct personnel costs plus fringe benefits (ie, health and dental insurance). Costs for clerical supplies were assumed to be minimal and therefore not included in the analysis. No additional expenditures for overhead were required of the pharmacy department. Personnel costs were measured by multiplying the time spent administering the program times the wage rate. The salary rate of the pharmacy fellow(s), including employee benefits, was $18.02/h. To assess the amount of time spent administering the program, each fellow kept a time log for the first 6 months of the program.

BENEFITS

The benefits to the institution were measured based on the value of the medications obtained free-of-charge from pharmaceutical manufacturers. Drug values were based on 2005 AWPs, in US dollars.⁵

SENSITIVITY ANALYSIS

Researchers conducted 2 types of sensitivity analyses-a threshold sensitivity analysis and a Monte Carlo simulation (probabilistic sensitivity analysis). Sensitivity analyses are used to test the robustness of an economic evaluation's overall conclusions. When conducting sensitivity analyses, one's confidence in the study conclusions is strengthened if the results do not significantly differ when key variables are altered within a range of plausible values. The value of medication procured, time spent administering the MAP, and wage rate were varied as part of the sensitivity analyses.

In threshold sensitivity analysis, the value of each variable is varied independently until the program becomes cost-neutral (ie, the break-even point is reached).4 In Monte Carlo simulation (utilizing the Crystal Ball 2000 software program) plausible ranges for each variable and the distribution of data points are assigned.4 A computer simulation is run in which random values from the variable ranges are selected multiple times, while outcomes (ie, the program's net benefit) are estimated and plotted. During each iteration, different values for the free medication procured, personnel time, and wage rate were substituted.

For this simulation, a triangular distribution was chosen, which sets the range of values by defining the likeliest, minimum, and maximum values for each variable. Since the program was evaluated from the perspective of the hospital pharmacy department, the minimum value ($68,175) for the benefit variable was the lowest estimable value of the drug when calculated using the institution's actual acquisition costs. The maximum value ($171,504) for the benefit variable was derived from a previous study of an outpatient MAP identified through a search of the medical literature.6 As mentioned previously, the likeliest benefit was the actual cumulative benefit observed in this program, valued in 2005 US dollars based on the drugs' AWP.

The minimum ($13.65/h, for a hospital pharmacy technician) and maximum ($43.60/h, for a hospital pharmacist) wages were taken from the 2005 Pharmacy Compensation Reports;⁷ the likeliest wage was assigned as that currently paid to the institution's pharmacy research fellows ($18.02/h).

The literature did not detail the minimum and maximum time spent on a similar MAP; consequently, the minimum (2 h/wk) and maximum (40 h/wk) expected time spent administering the MAP was used to define this variable range. The likeliest value of time spent on the MAP was determined through evaluation of the fellows' time logs.

RESULTS

An average of 4 hours per week were spent administering the program. Of these 4 hours, approximately 2.5 hours were spent meeting with patients, 1 hour was spent filing paperwork and corresponding with manufacturer programs, and the remainder of the time involved obtaining provider signatures and discussing treatment plans. The personnel costs associated with the administration of the program were $7,496. The benefit-to-cost ratio for the program was approximately 11:1, yielding a net benefit of $76,008.

Upon threshold sensitivity analysis, the MAP was found to be cost-neutral (net benefit=$0) when the value of the medication procured was reduced to $7,496, personnel time was increased to 44.5 hours per week, and wage was increased to $200 per hour.

DISCUSSION

Moreover, the program's claim of being cost-beneficial was found to be robust to changes in key variables when various sensitivity analyses were conducted. In order for the program to become cost-neutral, threshold sensitivity analysis revealed that the drug value would have to be reduced to $7,496, personnel time increased to 4,634 hours (44.5 h/wk), or the wage rate increased to $200 per hour. Since none of these values fall within what most researchers would consider "plausible ranges" for these variables, the conclusions are further strengthened. Additionally, Monte Carlo simulation demonstrated there was a 96.1% probability of the program being cost-beneficial.

The ability of a MAP to be a cost-beneficial endeavor has been demonstrated in at least 2 prior articles published in the medical literature.8,9 The first program, also implemented at Hartford Hospital, helped procure drugs free-of-charge (mostly chemotherapeutic agents) for inpatients and yielded a 6-month net benefit of $52,057. Similarly, a comprehensive outpatient prescription procurement program established in a cardiovascular clinic utilized manufacturer programs, drug samples, and state aid programs and yielded a net benefit of $66,876 from free drugs over a 6-month period. The MAP discussed in this article differs from the previously mentioned programs in that it focused on a broad outpatient population and utilized only manufacturer assistance programs to obtain drugs free-of-charge.

It is noteworthy to mention that economic benefits such as those demonstrated in this evaluation are not the only benefits resulting from the implementation of a MAP. Clinical benefits have also been realized. For example, Schoen et al10 demonstrated that the provision of a MAP to indigent patients resulted in improved drug adherence and clinical outcomes (eg, lower blood pressure and cholesterol levels), as well as a decrease in the number of hospitalizations when cardiovascular drugs were obtained. Additionally, Strum et al11 demonstrated improved clinical outcomes in patients with type 2 diabetes mellitus enrolled into a clinic-based MAP. In this study, mean hemoglobin A_1c and LDL-C levels decreased significantly over the year the patients were enrolled in the program compared to the year prior to enrollment (P<.001 for both end points).¹¹ Although not evaluated in this analysis, future evaluations of the MAP program at Hartford Hospital should include both humanistic and clinical end points.

CONCLUSIONS

This study illustrated that the MAP for ambulatory care patients at Hartford Hospital was cost-beneficial. The conclusions of the study were found to be robust to changes in drug value, personnel time, and wage rate.

Dr Gillespie and Dr Patel are Pharmacoeconomic and Outcomes Research fellows in the Department of Pharmacy Practice, School of Pharmacy, University of Connecticut, Storrs, Conn, and the Department of Pharmacy Services, Hartford Hospital, Hartford, Conn. Dr Henyan and Dr Sander are Cardiovascular Outcomes Research fellows in the Department of Pharmacy Practice, School of Pharmacy, University of Connecticut, and the Department of Pharmacy Services, Hartford Hospital. Dr White is co-director, Cardiovascular Pharmacology and Arrhythmia Research, Hartford Hospital; and associate professor of Pharmacy Practice, Department of Pharmacy Practice, School of Pharmacy, University of Connecticut. Dr Coleman is assistant professor of Pharmacy Practice, Department of Pharmacy Practice, School of Pharmacy, University of Connecticut; and director, Pharmacoeconomics and Outcomes Studies Group, Hartford Hospital. He can be reached at ccolema@harthosp.org.

REFERENCES

1. Lack of healthcare and prescription drug insurance in the US: a serious public health problem. Available at: http://www.pparx.org/americas_uninsured.php. Accessed January 3, 2006.

2. Fact sheet. How the pharmaceutical industry helps America's uninsured. Available at: http://www.pparx.org/help_uninsured.php. Accessed January 3, 2006.

3. Pharmaceutical Research and Manufacturers Association's (PhRMA) prescription patient assistance directory: resources & next steps. Available at: http:// http://www.disabilitybenefits101.org/ca/programs/health_coverage/prescription/phrma/resources.htm. Accessed January 3, 2006.

4. Gold MR, Siegel JE, Russell LB, Weinstein MC. Cost-effectiveness in health and medicine. New York: Oxford University Press;1996.

5. Fleming T. Red Book: Pharmacy's Fundamental Reference. Thomson Healthcare;2005.

6. Hotchkiss BD, Pearson C, Lisitano R. Pharmacy coordination of an indigent care program in a psychiatric facility. Am J Health-Syst Pharm. 1998;55:1293–1296.

7. Pharmacy Week 2005 Pharmacy Compensation Survey-Spring Edition. Available at: http:// http://www.pharmacyweek.com/job%5fseeker/salary/?e=404/. Accessed January 3, 2006.

8. Coleman CI, Reddy P, Quercia RA, et al. Cost-benefit analysis of a pharmacy managed medication assistance program for hospitalized indigent patients. Am J Health-Syst Pharm. 2003;60:378–382.

9. Schoen MD, Dischler JE, Connor SE, et al. Economic outcomes and medication compliance in a prescription procurement clinic for indigent patients with cardiovascular disease. Circulation. 1997;96(8):I-617.

10. Schoen MD, DiDomenico RJ, Connor S, Dischler JE, Bauman JL. Impact of the cost of prescription drugs on clinical outcomes in indigent patients with heart disease. Pharmacotherapy. 2001;21(12):1455–1463.

11. Strum MW, Hopkins R, West D, Harris B. Effects of a medication assistance program on health outcomes in patients with type 2 diabetes mellitus. Am J Health-Syst Pharm. 2005;62:1048–1052.