| Analysis of Total Pharmacy Services Offered to Patients Admitted to Long-Term Care Facilities |
Lisa M. DeBoer Caroline A. Gaither |
In the mid-1970s, little, if any, motive existed for pharmacists to dramatically improve care provided to long-term patients. Cheung and Kayne3 described the health care system at that time as follows: "The pharmacist is compensated from profit based on the number of medications he/she dispenses, and there is no economic and professional incentive motivating the pharmacist to improve care or reduce cost of drug therapy." However, the growing elderly population and the increasing cost of health care of the 1990s is driving demand for increased quality of care at a reasonable cost for these individuals.
A number of studies have investigated the positive contributions of consultant pharmacists to increase quality of care provided to patients.4-7 In a comprehensive literature analysis of pharmacist-conducted drug reviews, Kidder8 combined results from 23 studies to estimate a national annual savings generated by drug reviews. Reductions in medications prescribed to patients, hospitalizations, and nursing time spent on drug administration during periods of pharmacist review were evaluated. Kidder estimated the national annual savings from drug reviews to be $306,566,000. Reduced medication use accounted for about one-fourth ($81,566,000) of this total savings generated by the consultant pharmacist, and averted hospitalizations accounted for $224 million. Additionally, Kidder estimated a savings of $154 million for reduced nursing time; this figure was not included in the annual savings since most institutions use this time to offer increased care to patients.
Although much evidence supports the benefits of consultant pharmacy services, cost savings generated by total pharmacy services (consultant and dispensing pharmacists) remain unstudied. Changes in therapy often are implemented by the dispensing pharmacist before review of a medication regimen by the consultant pharmacist. To account for all aspects that promote quality and cost-effective care, activities of the dispensing pharmacist also need to be evaluated.
In the current study we analyzed the impact of a long-term care pharmacy on drug regimens and medical care offered to patients newly admitted to long-term care facilities. Newly admitted patients were chosen because the pharmacy services could be accurately quantified by analyzing patients not previously under the care of the pharmacy. Few, if any, studies have analyzed pharmacy services offered to only newly admitted patients.
Objectives
The objectives of this study were as follows:
Methods
Study Site
Pharmacy The site of this study was a long-term care pharmacy in Grand Rapids, Michigan. The pharmacy serves a variety of off-site facilities via a courier service. During the study period, the pharmacy employed three full-time pharmacists, four pharmacy technicians, a billing clerk, an intern, and department manager. All employees assisted in the data-collection process.
Long-term care facilities Twelve long-term care facilities were included. Although six facilities offered a higher degree of care, we believed the 12 homes accurately represented the diverse mix of area nursing homes because of the overall degree of care that they provided to patients. Designation of the six homes offering a higher degree of care was based on subjective comments from the consulting pharmacists and the ability of the facilities to offer skilled services such as intravenous drug therapy and physical and occupational therapy. Approximately 1,000 patients resided at these facilities.
Based on previous admission rates at these facilities, the pharmacy expected to serve approximately 300 new patients during the three-month study period. The census at each long-term care facility was used to predict the expected share of the total study patients. Facilities having a higher than expected turnover rate (more than their expected share of the study patients) were analyzed to determine if the patients residing at these facilities had a higher number of interventions.
Inclusion Criteria
Patients were included in the study if they were admitted to one of the 12 long-term care facilities served by the pharmacy and remained at facility for at least one medical chart/history review by the consultant pharmacist. Patients who were away for a short time (e.g., hospital) were included in the study for a maximum of three chart reviews.
Patients admitted from March 1, 1992, through May 31, 1992, were included. All interventions made within the first three months of admission and the first three monthly pharmacist consultations were analyzed. Therefore, data collection lasted a total of six months (three months after May 31, 1992). We believed the three-month period to be an optimal time for the consultant pharmacist to review individual patient charts fully, interact with the physician, and recommend appropriate changes/modifications in drug therapy.
Sample Characteristics
Patient and nursing home characteristics were collected and evaluated. Patient information was categorized as follows:
These characteristics have been examined in other studies and were believed to be appropriate.3,4,6
Data Collection
The data collection method separated interventions according to those made by the consultant pharmacist (external intervention) while at the long-term care facility reviewing patient charts and those made by the dispensing pharmacist (internal intervention). External interventions were recorded on a form used previously to record interventions made by the consultant pharmacist. When an external intervention was necessary, the consultant pharmacist completed the form and awaited a response from the physician. External interventions were categorized by the researchers after collection of all interventions.
Interventions made at the pharmacy were recorded on an internal pharmacy intervention form. This form was developed by the researchers for this study. When a change in an order was necessary, the pharmacist completed the form; the pharmacist also would indicate the type of intervention. Orders were not changed until approved by the physician.
Interventions were categorized as drug interactions, drug duplications, inappropriate prescribing of a medication, dosage changes, order clarifications, and requests to document a diagnosis. Drug interactions included a medication that would interact with a medication the patient was currently taking, an allergy or adverse reaction that the patient had documented on his her chart, or an interaction with a physiologic condition. Drug duplications included two or more medications used to treat a condition that only required one of the medications. Inappropriate prescribing occurred when a medication was prescribed for an incorrect dose (300 mg rather than 30 mg), interval (three times daily rather than twice daily), or length of time (no discontinuation date for an antibiotic).
Data Analysis
Cost savings Cost savings for interventions approved by the physician were calculated as follows: (1) discontinuation of a routine medication (a medication to be administered at a given time on a specified day) was based on the price at the pharmacy for a 30-day supply of the medication; (2) medications taken on an as-needed basis (p.r.n.) were calculated based on consumption of that medication by the patient during the previous 30 days; (3) if a medication was changed to a less frequent dosing schedule, the value was calculated by subtracting the cost of a 30-day supply of the new regimen from the cost of a 30-day supply of the original dosing regimen; and (4) if a medication was changed to a less expensive brand or product, the savings were calculated by subtracting the cost of a 30-day supply of the new product from the cost of a 30-day supply of the original product. Savings for medications prescribed for a specific length of time or for a particular amount were calculated based on the prescribed amount of time or specified dose rather than a 30-day supply (if an antibiotic was prescribed for 10 days and an intervention discontinued the order, the savings would be a 10-day supply of the medication less the doses the patient actually received). Cost savings were not calculated for interventions that discontinued a medication not being used.
Nursing time In addition to evaluating savings from reduced medication usage, saved nursing time attributed to reduced medication administration was analyzed. Each dose that no longer had to be administered was multiplied by 1.07 minutes. This number was used based on Farner and Hicks' study,9 which calculated nursing time involvement in medication-related activities to be 1.07 minutes per dose based on the 72-hour unit dose supply system. A 15-day supply of medication was dispensed by the pharmacy to the long-term care facilities. Because so little information was available on the time of administration with the 15-day supply system, the value for the 72-hour system was used. Of the traditional (bulk), 24-hour unit dose, and the 72-hour unit dose systems, the efficiency of the 15-day supply system would be most like that of the 72-hour unit dose system. Additionally, 1.07 minutes was the quickest delivery of the three dose systems analyzed, thus minimizing the possibility of overestimation.
Statistics Descriptive statistics (means and standard deviations) were calculated to describe characteristics of patients and long-term care facilities. The distribution of interventions (as defined by the internal and external intervention forms) were analyzed to observe for trends based on type of intervention (internal or external) and facility at which the intervention occurred. The total savings were calculated and divided by the number of patients in the study; this figure was the average savings per patient. In addition, average savings per intervention and average savings at each facility were calculated. Reductions in nursing time were calculated in a similar manner.
Interventions were categorized and compared in several ways. The number of interventions at a given facility was weighted according to the number of patients a facility had in the study. Thus, an interventions/patient number was used for these comparisons. The total numbers of internal interventions and external interventions were compared using the Student's t test for the paired data. Interventions also were compared based on whether they were made at a facility offering a low or high degree of care; the Student's t test was used to test for differences in the number of internal and external interventions at these two types of long-term care facilities. Facilities having greater than 50% of their patients with a billing status of either Medicaid or Medicare also were analyzed using the Student's t test to determine if they had a significantly greater number of internal or external interventions than other facilities.
Residency before admission was based on nursing home estimates. Actual patient preadmission residency status was not available for the study as this information is not routinely supplied to the pharmacy by the nursing home and the resources available to conduct the study did not allow for retrieval of this information. Therefore, the director of nursing at each facility was asked to estimate the percentage of patients who came to the facility from the hospital, home, another long-term care facility, or other. The estimate (given as a percentage) was multiplied by the number of patients a given facility had in the study. For example, if a nursing home had 10 patients in the study, and they estimated that 50% (0.5) of their admissions came from the hospital, then five patients (10 ¥ 0.5) were considered to have come from the hospital. The number of medications upon admission and at the end of the study period also were tabulated. These values were compared using the paired t test. The level of significance was set at p < 0.05.
Results
Table 1 lists the facilities included in the study, average daily census at each facility, number of study patients each long-term care facility was expected to have in the study, and number of patients each facility actually had in the study. Three facilities had greater than 25 patients in the study and these three facilities accounted for more than 50% of the study population. Based on the percentage of the total study population residing at each facility, one can see the turnover rate a facility had during the study period. For example, Home D, which has a census of 98 patients, was expected to have approximately 10% of the study patients. This is based on the assumption that every facility has an equal turnover rate; 98 of the 1,000 available patients served by the facility is approximately 10%. However, Home D actually had 18.3% of the study patients, indicating that it has a higher than expected turnover rate. Home G had less than its expected share of patients (0.6% instead of the expected 3.5%).
Patient Characteristics
After analyzing reports of patients admitted from March 1 through May 31, 1992, and excluding those not meeting the entry criteria, 164 patients were included in the study. A total of 53 men (average age: 78.0 ± 11.5 years) and 111 women (average age: 81.7 (± 9.5 years) were included in the study. Diagnoses of the patients in the study are listed in Table 2. Hypertension and congestive heart failure were the most common. The average number of medications upon admission (7.0 ± 3.7) was significantly less (p <. 0.05) than the number at the end of the study period or upon discharge, whichever came first (7.3 ± 3.6).
Before admission, 120 (73.2%) patients were in the hospital, 19 (11.5%) patients were living at home, 13 (8.0%) patients resided at another long-term care facility, and 12 (7.3%) patients were at a supportive care home, adult foster care home, home for the aged, or other community-based long-term care facility. These values were calculated based on estimates provided by each long-term care facility and were weighted according to the number of patients a facility had in the study.
The percentage of patients who remained in the study for the maximum three months was 66.5%. Reasons for discharge from the long-term care facility were not available to the investigators.
The most common methods of payment for medications were Medicare (45%), private pay (cash) (30%), and Medicaid (18%). More internal and external interventions occurred at long-term care facilities with greater than 50% of their patients with a Medicare or Medicaid billing status (p < 0.005).
Interventions
Table 3 shows the total number of internal and external interventions by facility. Of the 46 internal interventions, 18 were categorized as incomplete or unclear orders (amount saved = $168.07), 10 were considered inappropriate prescribing (amount saved = $108.48), 6 were therapeutic duplications (amount saved = $41.68), 1 was for a drug interaction (amount saved = $0), and 11 were other types of interventions (early refill, not covered by patient's insurance; amount saved = $105.15). Examples of internal interventions included an order for ampicillin for a patient already taking amoxicillin, ampicillin ordered on a patient with a penicillin allergy, and orders for medications without documented strengths. Inappropriate prescribing occurred most often with the antibiotics, but also occurred with other medications. Although incomplete/unclear orders accounted for the largest cost savings ($168.07), the most expensive type of intervention was inappropriate prescribing. These 10 interventions, which questioned the appropriateness of a given prescription, averaged $10.85 per intervention. The average savings generated by an internal intervention was $9.20. Internal interventions that had savings attributed to them (18 of the 46 internal interventions) had an average savings of $23.52 per intervention.
The external interventions were as follows: 19 requests for physicians to document a diagnosis for a particular medication (amount saved = $0), 15 recommendations for a dosage change (amount saved = $170.68), 7 requests for discontinuation of a medication (amount saved = $186.97), 7 requests for physicians to document a diagnosis for a particular medication that subsequently resulted in discontinuation of that medication (amount saved = $336.13), 5 requests to obtain specific laboratory tests ($0) and 1 request to change a medication (amount saved = $35.00). Examples of external interventions included a request to change a famotidine dosage from twice daily to bedtime only and a request to document a diagnosis for terfenadine. By requesting the physician to document a diagnosis and justify the use of a given medication, the physician reconsidered the appropriateness of the medication and, in some cases, chose to discontinue it. The average cost savings for implemented interventions requesting documentation of a diagnosis for a particular medication was $48.00. The average cost savings for an external intervention was $13.50; for those external interventions with a savings attributed to them (24 of the 56 external interventions), the average savings was $30.37.
The number of internal and external interventions at each long-term care facility is listed in Table 3. Total savings for internal and external interventions based on a 30-day supply of medication were $423.38 and $728.78, respectively. There were no significant differences between the long-term care facilities in the total number of internal and external interventions per patient. However, significantly (p < 0.005) more internal interventions per patient were made at facilities offering a higher degree of care. More internal and external interventions per patient occurred at long-term care facilities with more than 50% of their patients with a Medicare or Medicaid billing status (p < 0.005). The number of external interventions made at long-term care facilities offering a higher degree of care was not significantly different from those offering a lower degree of care.
Table 3 also lists the number of doses that no longer had to be administered because of interventions. Not all interventions had a reduction in the number of doses. Internal and external interventions reduced the number of doses by 540 and 895 doses, respectively. Using 1.07 minutes as the time it takes to deliver a medication, saved nursing time was calculated to be 25.6 hours.
All internal interventions received a response. This was expected because the pharmacist would call the nursing home or the physician to verify a medication order and wait for a response before processing and delivering the medication. The average length of time spent on an internal intervention was 4.2 ± 3.3 minutes. External interventions received a positive response (physician agreed with recommendation) 87% of the time. Overall, seven external interventions were not approved. Three interventions were requests for the physician to document a diagnosis for a particular medication, and, in all three cases, the patient was discharged before the physician could respond. The other four recommendations were for the physician to review a therapy and change the dosage of a medication. The physician did not agree with these recommendations.
Discussion
Three hundred patients were expected to be included in the study; however, only 164 met the entry criteria. Many patients were admitted to a long-term care facility for a transitional period between hospital and home and thus were not at the facility long enough to be included. Actual reasons for discharge from the facility were not available. Overall, the distribution of men and women accurately represents the distribution commonly seen in long-term care facilities.5,6 The distribution of diagnoses indicates that the patients were not critically ill, but they did have diagnoses that necessitated treatment with prescription medications.
The increase in average number of medications per patient during the study was unexpected. A retrospective look at the orders made during the study (those orders occurring after admission but before the end of any given patient's three-month study period) revealed 39 orders for aspirin, acetaminophen, and antacids. All were to be used p.r.n. by the patient. Although these orders are important, they represent nonprescription items that any person is likely to need, regardless of whether they are in a long-term care facility, and they were written to allow the patient to have the medication, if needed. These 39 orders accounted for 78% of the increase in medications during the three-month study period. Thus, the 0.3 medication per patient increase would no longer be significant.
Most patients were in the hospital before admission. This makes the role of the dispensing pharmacist more critical, in that long-term care facilities with a large percentage of patients in the hospital before admission tended to have a greater number of internal interventions. Orders for patients who were discharged from a hospital and admitted to a long-term care facility often are incomplete and unclear, accounting for the greatest number of internal interventions. This is usually not the fault of the facility; rather the confusion may develop when the transfer form is sent with the patient upon discharge from the hospital. Because these transfer forms are commonly illegible and unclear, the pharmacist must call the facility or the physician to clarify orders.
Briefly touched upon earlier was the idea of turnover rate at the facilities. Coincidentally, the facility with the largest turnover rate also had the largest number of internal interventions even after accounting for its largest percentage of patients in the study. This long-term care facility (Home D) also received, on the average, 95% of its patients from hospitals. The most probable explanation would be that long-term care facilities receiving a large number of patients from hospitals tend to have a greater number of internal interventions. For reasons given above, the dispensing pharmacist may have a greater role than the consultant pharmacist when making interventions at long-term care facilities whose patients are in the hospital before admission.
Long-term care facilities offering a higher degree of care had a greater number of internal interventions per patient. Internal and external interventions were higher in facilities with more than 50% of their patients with Medicaid or Medicare as the sole billing type. Again, these findings could result from these facilities admitting more patients from the hospital.
Both internal and external interventions provided a reduction of almost 1,500 doses. This provided patients with an additional 25.6 hours per month of nursing time previously spent distributing medications. In general, interventions would continually be made on these patients, as well as other patients currently at the long-term care facility. Therefore, only those medications truly necessary would be distributed and the rest of the time would be devoted to increasing the quality of patient care.
Although the dispensing pharmacist demonstrated significant contributions to patients residing in long-term care facilities, the work of the consultant pharmacist has been continually analyzed as a means of increasing the quality of patient care. As this study demonstrated, dispensing and consulting pharmacists have a role in providing optimal patient care. Interventions made by both types of pharmacists on the 164 patients could continually save more than $1,000 every month. This translates to a monthly savings of $7.03 per patient. In addition, interventions would continually be made on these patients as their stay at the nursing home continued. Therefore, total savings will increase for these patients as well as other patients admitted to and residing at the long-term care facilities.
A retrospective look at the interventions indicated that, of the $423.38 saved by internal interventions, $174.55 would have continually been spent routine medication orders had the pharmacist not made the intervention. Additionally, of the $728.78 saved on external interventions, $715.37 could be considered to be a continuous savings. This number is larger because of the nature of consultant pharmacists' work. They intervene on a patient's continuous medications, whereas the dispensing pharmacist intervenes before the patient receives the medication. Combined, this total savings of $889.92 ($174.55 + $715.37) represents an amount continually saved every month secondary to interventions made on these 164 patients ($5.43/patient). Thus, the continuous annual savings would be $10,679.16. ($889.93 multiplied by 12, since original values were based on a 30-day supply).
The above savings calculation is based on a single three-month period of new admissions. Four periods in a year, $42,716.64 could be saved through the combined efforts of the dispensing and the consulting pharmacists for patients when they are first admitted. Additionally, these savings should include interventions on orders not considered routine. These values would be $248.73 ($428.38 - $174.55) for internal interventions and $13.41 ($728.78 - $715.37) for external interventions. In four groups of new admissions, the annual savings from noncontinuous interventions would total $1,048.56. The annual savings for continuous and noncontinuous interventions is $43,764.72.
This is quite a phenomenal number considering Kidder showed the reduction in medications from pharmacy interventions to be only 25% of the total savings produced by pharmacy services.8 The other 75%, which was attributed to averted hospitalizations, we did not analyze. Hypothetically, inclusion of an averted hospitalizations figure would greatly increase the total annual savings.
Our current study had some limitations: We assumed the majority of the interventions were made during the first three months of admission; future studies might analyze longer time periods. Because our study examined only interventions made shortly after admission, a study lasting six months or one year could examine initial interventions and the continued role of the dispensing and the consultant pharmacists after the first three months.
Conclusion
This study presents a way for pharmacists serving long-term care facilities to monitor their contributions. The quality of care for long-term care patients is improved by the combined efforts of dispensing and consulting pharmacists. Many of the medications would have continued had the interventions not been made. Interventions on 164 patients resulted in an average monthly savings of $7.03/patient, of which $5.43 will continually be saved from the monthly medication bill. In addition, the increase in free nursing time gives patients a higher quality of care. Future studies are needed to further define the role of overall pharmacy services in long-term care facilities. Both dispensing and consultant pharmacists contribute significantly in the nursing home setting.
| Expected No. (%) | Actual No. | ||
| Facility | No. Licensed Beds | Study Patientsa (% of total) | Study Patientsa (% of total) |
| Home A | 49 | 8.2 (5.0) | 5 (3.0) |
| Home B | 63 | 10.5 (6.4) | 7 (4.3) |
| Home C | 100 | 16.7 (10.2) | 11 (6.7) |
| Home D | 98 | 16.4 (10.0) | 30 (18.3) |
| Home E | 44 | 7.4 (4.5) | 5 (3.0) |
| Home F | 70/123b | 11.7 (7.1) | 12 (7.3) |
| Home G | 34 | 5.7 (3.5) | 1 (0.6) |
| Home H | 69 | 11.5 (7.0) | 10 (6.1) |
| Home I | 50/71b | 8.4 (5.1) | 11 (6.7) |
| Home J | 180/207b | 30.1 (18.3) | 28 (17.1) |
| Home K | 125 | 20.9 (12.7) | 27 (16.5) |
| Home L | 99 | 16.6 (10.1) | 17 (10.4) |
| Diagnosis | No. Patients (%) |
| Hypertension | 78 (47.6) |
| Congestive Heart Failure | 46 (28.0) |
| Arthritisb | 43 (26.2) |
| Diabetes | 39 (23.8) |
| Cerebrovascular accident | 33 (20.1) |
| Dementia | 29 (17.7) |
| Anemia | 28 (17.1) |
| Hypothyroidism | 24 (14.6) |
| Constipation | 24 (14.6) |
| Chronic Obstructive pulmonary diseasec | 23 (14.0) |
| Cancer | 20 (12.2) |
| Atrial fibrillation | 19 (11.6) |
| Status post-leg/hip fracture | 19 (11.6) |
| Gastrointestinal ulcer disease | 15 (9.1) |
| Depression | 14 (8.5) |
| Alzheimer Disease | 13 (7.9) |
| Coronary artery release | 13 (7.9) |
| Renal failure | 12 (7.3) |
| Peripheral vascular disease | 12 (7.3) |
| Chronic urinary tract infection | 11 (6.7) |
Table 3. Internal and External Interventions by Facility
| Facility | No. Residents | Internala | Externalb |
| Home A | 5 | 0 (0) | 4 (17.68) |
| Home B | 7 | 0 (0) | 0 (0) |
| Home Cc | 11 | 1 (1.60) | 3 (112.59) |
| Home Dc | 30 | 20 (242.54) | 4 (171.97) |
| Home E | 5 | 0 (0) | 3 (53.80) |
| Home F | 12 | 0 (0) | 1 (1.12) |
| Home G | 1 | 0 (0) | 0 (0) |
| Home H | 10 | 1 (19.40) | 2 (6.59 |
| Home Ic | 11 | 3 (0) | 3 (5.29 |
| Home Jc | 28 | 7 (88.88) | 13 (140.26) |
| Home Kc | 27 | 7 (9.48) | 16 (72.67) |
| Home Lc | 17 | 7 (61.48) | 5 (156.81) |
| Total | 164 | 46 (423.38) | 54 (728.78) |
| Doses eliminated | 540 | 895 | |
1. Thompson JF, McGhan WF, Ruffalo RL et al. Clinical pharmacists
prescribing drug therapy in a geriatric setting. J Am Geriatrics
Soc 1984; 32:154-9.
2. Morton MR. Clinical reasoning behind the federal indicators.
J Pharm Pract 1978; 1:178-88.
3. Cheung A, Kayne R. An application of clinical pharmacy services
in extended care facilities. Calif Pharm 1975; 23:22-25, 28, 43.
4. Cooper JW, Bagwell CG. Contribution of the consultant pharmacist
to rational drug usage in the long-term care facility. J Am Geriatrics
Soc 1978; 26:513-20.
5. Wade WE, Morton WR. Consulting activities in a private-pay
long-term care facility. Consult Pharm 1987; 2:399-403.
6. Strandberg LR, Dawson GW, Mathieson D et al. Effect of comprehensive
pharmaceutical services on drug use in long-term care facilities.
Am J Hosp Pharm 1980; 37:92-4.
7. Chrymko MM, Conrad WF. Effect of removing clinical pharmacy
input. Am J Hosp Pharm 1982; 39:641.
8. Kidder SW. Cost-benefit of pharmacist-conducted drug-regimen
reviews. Consult Pharm 1987; 2:394-8.
9. Farner J, Hicks CI. The impact of drug distribution systems
in nurses' time involvement in medication related activities in
long-term care facilities. Drug Intell Clin Pharm 1976; 10:458-62.
Address for reprints: Caroline A. Gaither, College of Pharmacy, The University of Michigan, 428 Church St., Ann Arbor, MI 48109-1065.
Acknowledgments: To the staff of the long-term care pharmacy for their contributions in the data-collection process.
Copyright © 1996, American Society of Consultant Pharmacists, Inc. All rights reserved.