| Nurses' Satisfaction with Pharmaceutical Services for Home Care Hospices |
Candace L. Haugtvedt Linda D. MacKeigan Stephen W. Birdwell Donald L. Bennett |
Since the establishment of Connecticut Hospice as a National Cancer Institute demonstration project in 1974, hospice in the United States has grown from a grass-roots volunteer movement to a regulated industry reimbursed by Medicare (beginning in 1983), Medicaid (beginning in 1990) and private insurance programs.1 The aim of hospice is to allow patients to die comfortably, without pain, and at home with their family to the extent possible.2 Medicare, Medicaid, and private insurance programs in the United States emphasize home hospice care.3 Data from 1992 provided by the Ohio Hospice Organization indicated that about one third of 70 Ohio hospice programs were home-care agency based, one third were hospital-based, and one third were free-standing organizations.
A hospice team-made up of nurses, physicians, therapists, social workers, pharmacists, clergy, and dietitians-tries to meet the patient's physical, emotional, spiritual, and financial needs so the patient is comfortable and better able to deal with approaching death. Many hospice programs have a pharmacist as an integral part of their interdisciplinary team.4,5 The pharmacist's primary role on hospice teams has been to assure effective pain management through the use of different types of analgesics, administration techniques, and dosage schedules.6,7 Many other important consultant services provided by pharmacists are useful to hospice teams, including recommendations for symptom control of expected side effects and drug counseling information.8-12 Because hospice is designed to improve the quality of life of the terminally ill, pharmacists may be able to improve a patient's quality of life substantially by effectively controlling their pain. Pain has been found to affect all dimensions of quality of life-physical, psychological, social, and spiritual.13-15
The purpose of this study was to assess the quality of the pharmaceutical services provided to hospice programs in Columbus, Ohio. In the study, nursing satisfaction with pharmaceutical services provided by three hospice programs served as a measure of pharmaceutical service quality. Hospice pharmacists were found to interact primarily with hospice nurses, who in turn had direct contact with their patients. Nurses were thus viewed as the hospice pharmacy's primary customer, and nursing satisfaction was chosen as the best indicator of the quality of the pharmaceutical services provided to the hospice programs.
Background
The study area was served by three hospice programs: Hospice of Riverside, Hospice of Columbus, and The Mount Carmel Hospice. Each of these programs is unique with respect to the provision of pharmaceutical services. A brief description of each program and its pharmaceutical services at the time of the study follows.
Affiliated with Riverside Methodist Hospitals, a not-for-profit, teaching hospital in the U.S. Healthcare System, Hospice of Riverside was begun in 1986. A total of 11 primary care nurses cared for patients in their homes, and seven nurses served the nine-bed inpatient unit. The Hospice of Riverside had its own pharmacist who devoted all of her time (30 hours a week) to the hospice program; she was employed by the Department of Pharmacy at Riverside Methodist Hospitals. She acted as a consultant pharmacist for the hospice staff on pain and symptom control, made house calls, provided drug information, educated the nursing staff, monitored patients' drug therapy, served on a quality improvement committee for the hospice program, and attended all weekly hospice team meetings. This pharmacist had a B.S. in pharmacy degree with 16 years' clinical experience and an office on-site. Drug dispensing for inpatients was handled by the Riverside Methodist Hospitals' Pharmacy Department, and for home patients by a local community pharmacy. The staff pharmacists of the local community pharmacy also had B.S. in pharmacy degrees with no advanced training.
The Hospice of Columbus, affiliated with the City of Columbus Health Department and the first free-standing hospice in Columbus, began to serve the community in 1978. The program employed seven full-time and two part-time nurses. The nurses were on-call from 4:00 p.m. to 8:00 a.m. each day as well as all day on weekends. The Hospice of Columbus contracted with a local community pharmacy (the same one as Riverside) to dispense and deliver medications to patients. Nurses did not receive any formal orientation to pharmaceutical services or educational programs from the staff of the community pharmacy, but could call them with drug information questions. A pharmacist was not attending hospice team meetings at the time of this study.
Affiliated with Mount Carmel Medical Center, a nonprofit teaching hospital in the Sisters of the Holy Cross Healthcare System, Mount Carmel Hospice employed 15 primary care nurses directly involved in providing care in patients' homes, staffing two shifts per day (days and evenings), with nights and weekends covered on an on-call basis. Pharmaceutical services were provided by the outpatient pharmacy at Mount Carmel Medical Center. Pharmaceutical services provided to the hospice included dispensing (mainly narcotic analgesics, antinausea agents, and laxatives); providing drug information to nurses and physicians on dosing, drug costs, therapeutic alternatives, generic equivalents, and side effects; calling physicians for new prescriptions; delivering drugs directly to patients' homes; and participating in hospice team meetings. Outpatient pharmacy personnel included a full-time pharmacist, full-time technician, and part-time pharmacist. Both pharmacists had a B.S. in pharmacy with 16 and 12 years' experience, respectively. Formal educational programs were not provided to the hospice nurses by the pharmacy staff at the time of this study.
Methods
All full- and part-time primary care hospice nurses employed by the three hospice programs were included in this study. The quality of hospice pharmaceutical services provided to each hospice program was assessed by asking these nurses to complete a self-administered questionnaire regarding their satisfaction with the services.
The questionnaire was developed by interviewing nursing administrators at the Mount Carmel Hospice and Hospice of Columbus, the hospice pharmacist at Hospice of Riverside, and the director of pharmacy at Mount Carmel Medical Center about aspects of service they would like to see assessed. Areas identified included the following:
Using established guidelines for questionnaire development, 13 questions were developed to assess these areas of interest.16 The instrument was pilot-tested on four nursing administrators of the Mount Carmel Hospice who were not involved in the primary care of patients (but had previous experience in this area) to assess content clarity, ease of reading, and face validity. Based on their comments, one question was added about the pharmacist as a resource for pain and symptom control, and some questions were reworded. A question on nurses' perception of the profitability of providing pharmaceutical services to the hospice program was retained despite a comment that it was irrelevant.
The final questionnaire included 14 questions, with five unfavorably worded items, four favorably worded items, two neutral items, two demographic items, and one open-ended question. Ten questions used a seven-point rating scale anchored at 1 = strongly disagree and 7 = strongly agree. For the question assessing overall level of satisfaction the response scale was anchored at 1 = extremely dissatisfied and 7 = extremely satisfied. The open-ended question was used to reveal additional problem areas and details not identified by close-ended questions. Questionnaires were self-administered in April 1993. A copy of the questionnaire is available from the authors upon request.
Subsequent to establishing adequate homogeneity of items by Cronbach's alpha coefficient, item scores were aggregated to produce a Likert scale score (overall satisfaction). Unfavorably worded items on the questionnaire were reverse coded so that a higher score on an item indicated a more favorable rating. For example, "The pharmacists could do a better job solving problems with drug therapy in my hospice patients," was reverse coded. Cronbach's alpha for the summated scale of the 11 items was 0.66. The item on profitability correlated very poorly with the other items on the scale. Since its inclusion had been challeged during pilot testing, it was deleted from the scale, increasing Cronbach's alpha to 0.71. The overall satisfaction score was then
determined by summing the scores on these 10 items (range 7-70). The a priori level of significance was set at 0.05 (two-tailed) for all analyses. The data were analyzed using the SAS Statistical Package (Cary, NC).
Descriptive statistics for each item were determined for each hospice program. Analysis of variance was used to determine if there were significant differences in the mean values for overall satisfaction across groups. To determine the specific nature of program differences, individual items were tested separately using the Kruskal-Wallis test. A nonparametric test was chosen because the response to each item was measured on an ordinal scale and the sample size was restricted. Statistically significant differences on the Kruskal-Wallis test were explored by conducting post-hoc pairwise comparisons using the Wilcoxon Rank Sum test. The level of significance was p < 0.05.
Alpha slippage because of multiple testing was handled by using the Bonferroni correction to adjust the alpha level to 0.003 (0.05/14 where 14 was the number of tests conducted). Because this adjustment is extremely conservative and because the sample size was constrained by the nature of the target population, tests were also conducted with no adjustment to the a priori alpha level.
Results
All questionnaires were collected from Mount Carmel Hospice nurses (100% response rate). Six out of eight questionnaires were returned from Hospice of Columbus nurses (75% response rate). Of the two nonrespondents, one nurse was on vacation and the other was sick. All questionnaires were returned by Hospice of Riverside nurses (100% response rate).Thus, total of 25 out of 27 questionnaires were returned, for an overall response rate of 92.6%.
Tables 1 and 2 summarize the work experience of the nurses in the three programs. Hospice nurses working at the Mount Carmel Hospice had considerably less work experience in hospice than nurses in the other two programs. The average years with their current employer was roughly equivalent for two of the hospice programs, with Mount Carmel nurses having about one year less experience.
Table 3 summarizes the descriptive statistics for each item included in the self-administered questionnaire. Analysis of variance for overall satisfaction among hospice groups resulted in an F value of 1.37 (p = 0.27), indicating no statistical difference. As Table 3 shows, the mean values were significantly different between the three hospice programs on the item "pharmacist as a resource" (p = 0.006) and for the item "timeliness of delivery" (p = 0.05).
Results of pairwise comparisons using the Wilcoxon Rank Sum test on the item "pharmacist as a resource" were: Riverside and Mount Carmel (X2 = 5.9, p = 0.015); Mount Carmel and Columbus (X2 = 1.1, p = 0.29); Riverside and Columbus (X2 = 9.16, p = 0.0025). With the Bonferroni correction for multiple testing, the difference between Hospice of Riverside and Hospice of Columbus nurses remained significant.
"Timeliness of delivery" had a mean of 5.75 for Mount Carmel, compared with 3.91 for Riverside, and 3.83 for Columbus nurses (midpoint 4), indicating that Mount Carmel nurses were satisfied with delivery time, whereas the others were slightly dissatisfied. Results of pairwise comparisons using the Wilcoxon Rank Sum test on the item "timeliness of delivery" were: Riverside and Mount Carmel (X2 = 4.94, p = 0.0263); Mount Carmel and Columbus (X2 = 3.90, p = 0.0483); and Riverside and Columbus
(X2 = 0.023, p = 0.8785).
Discussion
The quality assessment of hospice pharmaceutical services indicated no difference in overall satisfaction with the pharmaceutical services provided to nurses in the three hospice programs in Columbus, Ohio. Although not statistically significant, the mean was the highest for nurses employed by Mount Carmel Hospice (63.5 out of 70). This may be because timeliness of medication delivery to their hospice patients is a very important component of overall nurses' satisfaction with pharmaceutical services. Pharmacy managers may want to monitor how much time is required to deliver a stat medication as well as routine medications.
Hospice of Riverside nurses were more satisfied than Mount Carmel or Hospice of Columbus nurses with the pharmacist as a resource for information on pain and symptom control. This may be because a pharmacist was dedicated exclusively to serving Hospice of Riverside patients and their nurses, acting also as a consultant rather than primarily as a dispenser of medications. The pharmacist at Riverside provided educational programs for the nurses, was an integral part of their orientation process, and attended all team meetings. Riverside's pharmacist also had an office on-site with the hospice nurses for more one-on-one interaction, whereas interaction was primarily by telephone in the other two hospice programs. The implications for this finding for pharmacy managers may include dedicating a position exclusively to the hospice program and providing that pharmacist with an office on-site with the hospice nurses, and having minimal dispensing responsibilities.
The results of this study on the quality of pharmaceutical services are generalizable only to the target population of primary care home hospice nurses in Columbus, Ohio. However, the approach to quality assessment is one that other pharmacy managers may use to conduct similar studies on their own hospice pharmacies. Also, the hospice programs in Columbus closely match the national averages reported by the National Hospice Organization in 1990: two thirds of hospice patients were over age 65 (and hence qualified for Medicare), 84% suffered from cancer, 4% AIDS, 3% cardiovascular disease, and 9% other terminal illnesses. Therefore, these hospice programs and their nurses are fairly representative of hospice programs nationwide.
Another limitation of this study is the limited evidence of reliability and validity of the questionnaire, since it was developed for the purpose of this study with no prior testing.
Conclusion
Hospice nurses in Columbus, Ohio, were satisfied overall with the pharmaceutical services provided to them. A specific area in which the pharmacy at Mount Carmel Medical Center excelled over the other two programs was in timeliness of medication delivery. All hospice programs provided services to patients in the greater Columbus metropolitan area, with Mount Carmel less centrally located than the local community pharmacy. Mount Carmel worked closely with its distribution center to make hospice prescription deliveries a priority, however. Pharmacists at Mount Carmel and the local community pharmacy could both work to improve nurses' satisfaction in the specific area of the pharmacist as a resource for pain and symptom control, an area in which Hospice of Riverside excelled. The pharmacists' level of education and clinical experience were roughly equivalent across all of the hospices. The critical differences between the hospice programs were that Riverside's pharmacist had an office on-site with the hospice nurses, was devoted exclusively to the program, and had initiated many educational programs for the nurses, including an orientation to pharmaceutical services.
Future studies with greater generalizability could be designed
by randomly sampling from the entire population of hospice nurses
in the United States to determine the relationship between satisfaction
with pharmaceutical services and the structure and process of
pharmaceutical services provided.
Table 1. Years of Work as Hospice Nurse
| Program | Mean | ± | S.D. | (Range) |
| ||||
| Riverside | 5.09 | ± | 3.32 | (1-13) |
| Mt. Carmel | 2.00 | ± | 1.31 | (1-4) |
| Columbus | 3.33 | ± | 2.73 | (1-8) |
Table 2. Nurses' Years of Work with Current Employer
| Program | Mean | ± | S.D. | (Range) |
| ||||
| Riverside | 5.64 | ± | 3.88 | (2-9) |
| Mt. Carmel | 4.63 | ± | 5.26 | (1-17) |
| Columbus | 5.67 | ± | 4.93 | (1-12) |
Table 3. Descriptive Statistics for Nurses' Satisfaction with Pharmacy Services by Hospice Program
| Item |
Mean Riverside (n=11) |
(± S.D.) Mt. Carmel (n=8) |
Value Columbus (n=6) |
| |||
| General satisfaction (a) | 5.82 ± 1.47 | 5.75 ± 1.98 | 5.83 ± 0.41 |
| Pharmacist problem solving | 5.18 ± 2.18 | 6.00 ± 1.60 | 4.83 ± 1.83 |
| Pharmacist availability (b) | 5.72 ± 1.90 | 6.63 ± 0.74 | 6.33 ± 0.82 |
| Timeliness of delivery (c) | 3.91 ± 1.81 | 5.75 ± 1.28 | 3.83 ± 1.72 |
| Pharmacy profitability | 3.60 ± 1.58 | 3.13 ± 1.25 | 2.50 ± 1.38 |
| Pharmacy accessibility | 6.00 ± 1.18 | 6.88 ± 0.35 | 5.83 ± 1.60 |
| Pharmacist as team member | 6.45 ± 1.81 | 7.00 ± 0.00 | 7.00 ± 0.00 |
| Quality drug information | 5.81 ± 1.47 | 6.75 ± 0.71 | 6.17 ± 0.75 |
| Pharmacist as caring | 6.45 ± 0.93 | 6.25 ± 0.71 | 6.33 ± 0.82 |
| Pharmacist as resourcec | 6.91 ± 0.30 | 6.25 ± 0.71 | 5.80 ± 0.75 |
| Know pharmacy procedures | 6.36 ± 1.03 | 6.25 ± 1.75 | 6.33 ± 0.82 |
| Satisfaction (sum) (d) | 58.63 ± 8.89 | 63.50 ± 4.28 | 58.30 ± 5.75 |
| |||
1. Department of Health and Human Services, Health
Care Financing Administration, Medicare Program. Hospice care:
final rule. Fed Regist 1983: 56032.
2. Mor V. Hospice care systems: structure, process,
costs and outcome. New York: Springer Publishing Co., 1987.
3. Feather AD. Medicare hospice benefit: early program
experiences. Health Care Financ Rev 1988: 99-111.
4. Schoenike S, Brown S. The pharmacist's role in
hospice care. NARD J 1990 (May); 112(5): 77-81.
5. Whigham WD, Roberts KB. Considerations when proposing
consultant pharmacist services to hospice programs. Contemp Pharm
Pract 1988; 5: 239-45.
6. Lindeman L. Missions of mercy: independents explore
the boundaries of hospice pharmacy. NARD J 1992 (Dec); 113(12):
37-43.
7. Arter SG, Dube J, Mahoney JJ. Hospice care and
the pharmacist. Am Pharm 1987 (Sept); NS27(9): 32-6.
8. Cramer R. Opportunities for pharmacists in hospice
organizations. Am J Hosp Pharm 1988; 45: 76.
9. Toal DR. Pharmacists' roles in the information
age: hospices. Consult Pharm 1987; 2: 290-1.
10. Salzer LB. Changing information into dollars.
Consult Pharm 1986; 1: 17-22.
11. Walters JK. Hospice care and the pharmacist:
pharmacist's role in counseling hospice patients and their families.
NARD J 1984 (Apr); 106: 61-5.
12. Murphy DH. Delicate art of caring: treating
the whole person. Am Pharm 1984 (Jun); NS24: 68-70.
13. Ferrell B, Wisdom C, Wenzl C. Quality of life
as an outcome variable in the management of cancer pain. Cancer
1989; 63: 2321-7.
14. Ferrell B, Wisdom C, Wenzl C, Brown J. Effects
of controlled release morphine on QOL for cancer pain. Oncol
Nurs Forum 1989; 16: 521-6.
15. Padilla G, Ferrell B, Grant M, Rhiner M. Defining
the content domain of quality of life for cancer patients' pain.
Cancer Nurs 1990; 13: 108-15.
16. Alreck PL, Settle RB. The survey research handbook.
Homewood,IL: Richard D. Irwin, Inc., 1985.