Abstract
The results of a 2011 survey evaluating pharmacy services at small and rural Illinois hospitals are presented and compared with data from similar surveys in 2001 and 1991.
A questionnaire modeled on the previous survey instruments but updated to reflect contemporary pharmacy practice was mailed to pharmacy directors at 86 small hospitals (i.e., <150 staffed beds) and rural hospitals (i.e., located outside metropolitan areas).
The response rate was 46.5%. The survey data indicated that 57.5% of hospitals represented in the 2011 survey had a centralized drug distribution system, 35.0% had a hybrid system, and 7.5% had a decentralized system. The most commonly reported form of technology was automated dispensing cabinets, which were in use at 75.0% of hospitals in 2011, compared with 34.8% of hospitals represented in the 2001 survey. Barcode verification of medication doses before dispensing and at the time of administration was performed at 50% and 70% of hospitals, respectively. While the provision of clinical pharmacy services has risen sharply since 1991, substantial changes were not observed between 2001 and 2011 except in the provision of compliance and drug histories (67.6% of hospitals in 2011 versus 46.8% in 2001) and pharmacist participation in medical emergency responses (54.0% versus 34.0%).
A 2011 survey of pharmacy departments in small and rural Illinois hospitals provided information on the use of automation and health information technologies and showed changes in the provision of many clinical pharmacy services since 1991.
Hospitals often serve as the hub of the health care delivery system in small and rural communities in the United States and, as such, play a vital role in ensuring access to safe and effective care.1 However, the economic viability of these facilities is a constant concern because of shrinking reimbursement for services combined with the rising costs of technology and personnel. In addition, the size of the rural population is declining in some areas, and this may result in lower patient volumes for rural hospitals.2 Economic issues such as these may lead to difficult decisions about the services that can be provided, and in some cases existing services must be reduced or curtailed. Pharmacy services—and clinical pharmacy services in particular— may be subject to such cuts.
At the same time, recent changes in the U.S. health care system may create new opportunities for small and rural hospitals and pharmacists in those facilities. For example, the Patient Protection and Affordable Care Act includes measures to expand the health care work force, particularly in primary care, and provides new funding for entities designated as Rural Community Health Centers and Federally Qualified Health Centers.3 The law also allows rural hospitals to participate in the federal 340B prescription drug discount program and has provisions to equalize Medicare and Medicaid reimbursement of rural and nonrural hospitals.4
Furthermore, although technology is expensive, it offers the promise of increased efficiency and expanded patient access. Telemedicine and telepharmacy are of particular interest to rural hospitals for this reason.1,5 In addition, electronic medical record (EMR) and computerized prescriber-order-entry (CPOE) systems are strongly recommended for adoption by hospitals. In rural settings and critical access hospitals, CPOE can enable not only improved safety and reduced variation in care but also remote or centralized order processing that reduces personnel costs.6,–8
It is important to understand how the practice of pharmacy in small and rural hospitals has been influenced by these and other recent issues affecting health care delivery and financing. Large national surveys of hospital pharmacy services, such as those conducted by the American Society of Health-System Pharmacists (ASHP), have been informative but typically do not yield results that allow for the characterization of rural hospital practices.9,–11 Further, while national surveys may help in our understanding of pharmacy practice on a national level, the results may not accurately reflect such practice within a single state or region and thus may not be useful in making policy decisions on a state or regional basis.
In 1991 and 2001, we conducted surveys that assessed the provision of pharmacy services in rural hospitals in Illinois, including both distributive and clinical services.12,13 The purpose of the study described in this article was to update our earlier findings by examining the extent and types of pharmacy services, including the use of technology, in small and rural hospitals in Illinois in 2011. We also examined trends that have occurred over the past two decades by comparing the results with those of previous surveys. Data from the 2011 survey pertaining to human resources and staffing issues, as well as findings specific to critical access hospitals, will be presented in future issues of this journal.
Methods
A questionnaire was developed and mailed to pharmacy directors at small and rural hospitals in Illinois. The questionnaire was designed to identify and quantify the provision of pharmacy services in these hospitals, including drug distribution systems and services, clinical pharmacy services, and human resources and staffing; it used methods similar to those of the surveys conducted in 1991 and 2001.12,13 While modeled after the past survey instruments, the 2011 questionnaire was revised to make it more relevant to contemporary practice, including the use of health information technology. The draft questionnaire was reviewed or completed by five individuals with experience in rural hospital practice or survey research methods, and revisions were made based on their recommendations, resulting in the final 54-item questionnaire used in the survey.
A publically available list of 88 small and rural hospitals published by the Illinois Hospital Association (IHA) was used to identify hospitals for inclusion in the survey.14 Hospitals on this list either had a capacity of fewer than 150 acute care staffed beds or were located in a rural area (i.e., outside a metropolitan statistical area [MSA]).15 We excluded 1 hospital because it was a long-term acute care facility without medical or surgical beds. All the remaining hospitals were crosschecked to ensure that each had a pharmacy license registered with the Illinois Department of Financial and Professional Regulation; 1 hospital did not and was eliminated from the survey pool, resulting in a final survey population of 86 hospitals.
The questionnaire was initially mailed to the director of pharmacy at each of the 86 hospitals in November 2011. A cover letter and a self-addressed, stamped return envelope were included in each mailing. To maximize the response rate, duplicate mailings were sent to nonrespondents 4, 8, and 12 weeks after the initial mailing. Reminder telephone calls were made to nonrespondents 11 weeks after the initial mailing; a final mailing targeted those who indicated their willingness to complete the questionnaire. In addition, e-mails encouraging response were periodically sent to pharmacy directors who were members of the Illinois Council of Health-System Pharmacists.
Each hospital was assigned a unique identification number. Data obtained from usable responses were entered and saved using commercially available database management software (Office Access and Office Excel, Microsoft Corporation, Redmond, WA). In addition, data on the hospitals of both respondents and nonrespondents were obtained from the American Hospital Association’s (AHA) comprehensive guide to U.S. health care institutions in order to assess nonresponse bias and validate certain items in the self-reported information.16 Statistical software (SAS, version 9.2, release 03.08, SAS Institute Inc., Cary, NC) was used to generate descriptive statistics to characterize the respondents. Inferential statistics were used (separate variance t tests for independent groups) for comparisons between groups based on selected hospital characteristics using an a priori level of significance of 0.05.
All results are expressed in terms of the number of affirmative responses relative to the number of usable responses for each question or criterion; in some instances where the numbers of item responses within a given survey topic area varied, data are presented as a fraction and corresponding percentage. When appropriate, results are compared with those from the 1991 and 2001 surveys.12,13 This study was conducted in accordance with policies of the institutional review board at the University of Illinois at Chicago.
Results
Hospital and pharmacy characteristics
A total of 40 usable questionnaires were returned, yielding an overall response rate of 46.5%. Among the 40 pharmacy directors who responded to the survey, 29 (72.5%) were based at facilities defined as rural hospitals and 39 (97.5%) were based at small hospitals.
Table 1 presents data on several variables related to hospital size and level of activity, as reported by respondents in 2011 and in the earlier surveys. Relative to the hospitals represented in the 2001 and 1991 surveys, those represented in the 2011 survey averaged fewer staffed beds (mean ± S.D., 53.5 ± 40.6) and a lower daily census (mean ± S.D., 31.6 ± 32.8). The average number of annual admissions was lower in 2011 versus 2001 (mean ± S.D., 2871 ± 5083 versus 3720 ± 3669).
Hospital Capacity and Demographics, by Survey Year
| Characteristic | Mean ± S.D. Value (No. Usable Responses) | ||
|---|---|---|---|
| 2011 | 2001 | 1991 | |
| Staffed bedsa | 53.5 ± 40.6 (40) | 76.7 ± 63.3 (39) | 93.8 ± 63.9 (77) |
| Average daily censusb | 31.6 ± 32.8 (39) | 41.0 ± 41.3 (40) | 51.2 ± 44.3 (77) |
| Annual admissionsc | 2,871.1 ± 5,083.7 (29) | 3,720.2 ± 3,669.2 (30) | …d |
| Annual outpatient visitse | 54,047.0 ± 50,107.0 (27) | 48,615.6 ± 57,755.2 (20) | … |
| Average length of stay | 3.3 ± 1.7 (39) | 4.06 ± 4.2 (40) | 4.9 ± 1.1 (77) |
| Characteristic | Mean ± S.D. Value (No. Usable Responses) | ||
|---|---|---|---|
| 2011 | 2001 | 1991 | |
| Staffed bedsa | 53.5 ± 40.6 (40) | 76.7 ± 63.3 (39) | 93.8 ± 63.9 (77) |
| Average daily censusb | 31.6 ± 32.8 (39) | 41.0 ± 41.3 (40) | 51.2 ± 44.3 (77) |
| Annual admissionsc | 2,871.1 ± 5,083.7 (29) | 3,720.2 ± 3,669.2 (30) | …d |
| Annual outpatient visitse | 54,047.0 ± 50,107.0 (27) | 48,615.6 ± 57,755.2 (20) | … |
| Average length of stay | 3.3 ± 1.7 (39) | 4.06 ± 4.2 (40) | 4.9 ± 1.1 (77) |
Number of inpatient beds regularly maintained and staffed for use; in the 2001 and 1991 surveys, these were reported as “operational beds.”
Inpatients excluding newborns.
Patients admitted for inpatient services.
Not surveyed.
Visits by patients not lodged at the hospital while receiving care.
Hospital Capacity and Demographics, by Survey Year
| Characteristic | Mean ± S.D. Value (No. Usable Responses) | ||
|---|---|---|---|
| 2011 | 2001 | 1991 | |
| Staffed bedsa | 53.5 ± 40.6 (40) | 76.7 ± 63.3 (39) | 93.8 ± 63.9 (77) |
| Average daily censusb | 31.6 ± 32.8 (39) | 41.0 ± 41.3 (40) | 51.2 ± 44.3 (77) |
| Annual admissionsc | 2,871.1 ± 5,083.7 (29) | 3,720.2 ± 3,669.2 (30) | …d |
| Annual outpatient visitse | 54,047.0 ± 50,107.0 (27) | 48,615.6 ± 57,755.2 (20) | … |
| Average length of stay | 3.3 ± 1.7 (39) | 4.06 ± 4.2 (40) | 4.9 ± 1.1 (77) |
| Characteristic | Mean ± S.D. Value (No. Usable Responses) | ||
|---|---|---|---|
| 2011 | 2001 | 1991 | |
| Staffed bedsa | 53.5 ± 40.6 (40) | 76.7 ± 63.3 (39) | 93.8 ± 63.9 (77) |
| Average daily censusb | 31.6 ± 32.8 (39) | 41.0 ± 41.3 (40) | 51.2 ± 44.3 (77) |
| Annual admissionsc | 2,871.1 ± 5,083.7 (29) | 3,720.2 ± 3,669.2 (30) | …d |
| Annual outpatient visitse | 54,047.0 ± 50,107.0 (27) | 48,615.6 ± 57,755.2 (20) | … |
| Average length of stay | 3.3 ± 1.7 (39) | 4.06 ± 4.2 (40) | 4.9 ± 1.1 (77) |
Number of inpatient beds regularly maintained and staffed for use; in the 2001 and 1991 surveys, these were reported as “operational beds.”
Inpatients excluding newborns.
Patients admitted for inpatient services.
Not surveyed.
Visits by patients not lodged at the hospital while receiving care.
As shown in Table 2, the mean total pharmacy expenditures for 2011 were $3,307,978. As expected, pharmaceuticals accounted for the majority of pharmacy expenditures (61.9%). When adjusted for inpatient volume, the survey data indicated that small and rural hospitals spent approximately $1,152 per patient admission, or $320 per patient-day, in 2011.
Hospital Pharmacy Expenditures, by Survey Year
| Category | Mean ± S.D. ($) | ||
|---|---|---|---|
| 2011 | 2001 | 1991 | |
| Total expendituresa | 3,307,978 ± 4,676,593 | 1,626,233 ± 1,881,396 | 1,348,980 ± 1,726,627 |
| Pharmaceuticalsb | 2,048,732 ± 2,700,827 | 1,196,871 ± 1,482,138 | …c |
| Pharmacy personneld | 732,718 ± 859,089 | 315,671 ± 234,913 | … |
| Category | Mean ± S.D. ($) | ||
|---|---|---|---|
| 2011 | 2001 | 1991 | |
| Total expendituresa | 3,307,978 ± 4,676,593 | 1,626,233 ± 1,881,396 | 1,348,980 ± 1,726,627 |
| Pharmaceuticalsb | 2,048,732 ± 2,700,827 | 1,196,871 ± 1,482,138 | …c |
| Pharmacy personneld | 732,718 ± 859,089 | 315,671 ± 234,913 | … |
Inclusive of drugs, supplies, personnel, and all other expense categories and (if applicable) all cost centers within the department.
Inclusive of pharmaceuticals and intravenous solutions and (if applicable) all cost centers within the department.
Not surveyed.
Inclusive of all pharmacy personnel and (if applicable) all cost centers within the department.
Hospital Pharmacy Expenditures, by Survey Year
| Category | Mean ± S.D. ($) | ||
|---|---|---|---|
| 2011 | 2001 | 1991 | |
| Total expendituresa | 3,307,978 ± 4,676,593 | 1,626,233 ± 1,881,396 | 1,348,980 ± 1,726,627 |
| Pharmaceuticalsb | 2,048,732 ± 2,700,827 | 1,196,871 ± 1,482,138 | …c |
| Pharmacy personneld | 732,718 ± 859,089 | 315,671 ± 234,913 | … |
| Category | Mean ± S.D. ($) | ||
|---|---|---|---|
| 2011 | 2001 | 1991 | |
| Total expendituresa | 3,307,978 ± 4,676,593 | 1,626,233 ± 1,881,396 | 1,348,980 ± 1,726,627 |
| Pharmaceuticalsb | 2,048,732 ± 2,700,827 | 1,196,871 ± 1,482,138 | …c |
| Pharmacy personneld | 732,718 ± 859,089 | 315,671 ± 234,913 | … |
Inclusive of drugs, supplies, personnel, and all other expense categories and (if applicable) all cost centers within the department.
Inclusive of pharmaceuticals and intravenous solutions and (if applicable) all cost centers within the department.
Not surveyed.
Inclusive of all pharmacy personnel and (if applicable) all cost centers within the department.
Clinical pharmacy services and policy
Survey participants were asked about the pharmacist deployment model in place at their hospital. Most respondents (26/39, 66.7%) reported the use of a “patient-centered or integrated model,” defined on the questionnaire as one in which nearly all pharmacists have distributive and clinical responsibilities. Two of 39 respondents (5.1%) reported the use of a “clinical specialist-centered model,” with distinct roles for distributive pharmacists and clinical specialists; the remaining 11 (28.2%) reported a “drug distribution-centered” model, in which pharmacists perform mostly distributive functions and provide limited clinical services.
The survey examined the types of drug policies used by pharmacy departments to influence medication use in small and rural hospitals. Most of the 39 pharmacy directors submitting usable responses in this topic area reported the use of automatic generic substitution (n = 34, 87.2%) and automatic therapeutic interchange by pharmacists (n = 33, 84.6%). Most of the respondents also reported the use of pharmacist interventions and recommendations to monitor prescriber compliance with policies (n = 34, 87.2%) and the development of clinical practice guidelines including medications (n = 25, 64.1%). Approximately half of the respondents reported the use of a policy requiring pharmacist approval of nonformulary medications (n = 20, 51.3%) and pharmacist prescribing under protocol (n = 21, 53.8%). On the other hand, only a quarter of respondents reported that pharmacists at their hospital could give prior approval for restricted drugs. A majority of the respondents (n = 23, 59.0%) reported the use of retrospective medication-use evaluation, and one third reported the use of pharmacist-provided education programs about medication costs. Some respondents (n = 7, 17.9%) reported that pharmacists at their hospital were engaged in clinical practice in outpatient clinics.
Respondents were also asked about the provision and frequency of provision of certain clinical pharmacy services; Table 3 presents comparative data from the 2011, 2001, and 1991 surveys. The percentage of small and rural hospitals in Illinois that provided each of the listed patient-specific clinical pharmacy services increased from 1991 to 2001 but did not change substantially from 2001 to 2011, with the exception of compliance and drug histories (46.8% of hospitals in 2001 and 67.6% of hospitals in 2011) and medical emergency response services (34.0% of hospitals in 2001 and 54.0% of hospitals in 2011). The only service for which there was a substantial decrease was nutritional support, which was provided by 83.0% of hospitals in 2001 and 73.0% of hospitals in 2011.
Provision of Clinical Pharmacy Services, by Survey Year
| Service | Fraction (%) of Usable Responses | ||
|---|---|---|---|
| 2011 | 2001 | 1991 | |
| Drug therapy monitoring | 36/37 (97.3) | 46/47 (97.9) | 54/74 (73.0) |
| Patient education and counseling | 34/37 (91.9) | 44/47 (93.6) | 19/74 (25.7) |
| Pharmacokinetic consultation | 36/38 (94.7) | 42/47 (89.4) | 24/74 (32.4) |
| Nutritional support | 27/37 (73.0) | 39/47 (83.0) | 28/74 (37.8) |
| Compliance and drug histories | 25/37 (67.6) | 22/47 (46.8) | 0/74 (0) |
| Medical emergency response | 20/37 (54.0) | 16/47 (34.0) | 23/75 (30.7) |
| Service | Fraction (%) of Usable Responses | ||
|---|---|---|---|
| 2011 | 2001 | 1991 | |
| Drug therapy monitoring | 36/37 (97.3) | 46/47 (97.9) | 54/74 (73.0) |
| Patient education and counseling | 34/37 (91.9) | 44/47 (93.6) | 19/74 (25.7) |
| Pharmacokinetic consultation | 36/38 (94.7) | 42/47 (89.4) | 24/74 (32.4) |
| Nutritional support | 27/37 (73.0) | 39/47 (83.0) | 28/74 (37.8) |
| Compliance and drug histories | 25/37 (67.6) | 22/47 (46.8) | 0/74 (0) |
| Medical emergency response | 20/37 (54.0) | 16/47 (34.0) | 23/75 (30.7) |
Provision of Clinical Pharmacy Services, by Survey Year
| Service | Fraction (%) of Usable Responses | ||
|---|---|---|---|
| 2011 | 2001 | 1991 | |
| Drug therapy monitoring | 36/37 (97.3) | 46/47 (97.9) | 54/74 (73.0) |
| Patient education and counseling | 34/37 (91.9) | 44/47 (93.6) | 19/74 (25.7) |
| Pharmacokinetic consultation | 36/38 (94.7) | 42/47 (89.4) | 24/74 (32.4) |
| Nutritional support | 27/37 (73.0) | 39/47 (83.0) | 28/74 (37.8) |
| Compliance and drug histories | 25/37 (67.6) | 22/47 (46.8) | 0/74 (0) |
| Medical emergency response | 20/37 (54.0) | 16/47 (34.0) | 23/75 (30.7) |
| Service | Fraction (%) of Usable Responses | ||
|---|---|---|---|
| 2011 | 2001 | 1991 | |
| Drug therapy monitoring | 36/37 (97.3) | 46/47 (97.9) | 54/74 (73.0) |
| Patient education and counseling | 34/37 (91.9) | 44/47 (93.6) | 19/74 (25.7) |
| Pharmacokinetic consultation | 36/38 (94.7) | 42/47 (89.4) | 24/74 (32.4) |
| Nutritional support | 27/37 (73.0) | 39/47 (83.0) | 28/74 (37.8) |
| Compliance and drug histories | 25/37 (67.6) | 22/47 (46.8) | 0/74 (0) |
| Medical emergency response | 20/37 (54.0) | 16/47 (34.0) | 23/75 (30.7) |
Figure 1 provides information on the use of selected patient-oriented clinical pharmacy services provided in 2011 and the frequency of provision (response options: never, infrequently, sometimes, often, always). The reported frequencies of provision varied for different clinical pharmacy services. For example, most hospitals that provided pharmacokinetic consultation did so often or always. On the other hand, the majority of responding pharmacy directors at small and rural Illinois hospitals reported that pharmacists never or infrequently participated in medical rounds with physicians.
Provision and frequency of clinical pharmacy services. The numbers of respondents indicating the use of each listed service ranged from 36 to 38.
Pharmacy directors were also asked about the type of EMR system used in their hospital. About one third of the respondents (12/38, 31.6%) reported the use of a complete EMR, and just over half (21/38, 55.3%) indicated the use of a partial EMR. Five of the 38 respondents (13.2%) indicated that their hospital did not have an EMR. A large majority of respondents at hospitals with an EMR (31/33, 93.9%) reported that pharmacists were able to access the system. More than two thirds of respondents (26/38, 68.4%) indicated that a CPOE system was in use; clinical decision-support (CDS) functionality was reported by 11 of 20 pharmacy directors (55%) who provided information on the use of CDS. Among the 26 respondents at hospitals with a CPOE system, 60.0% indicated that prescribers used it infrequently (i.e., for less than 50% of orders).
Survey participants were also asked about the use of an e-prescribing system to write and communicate prescriptions for outpatients. Among pharmacy directors in hospitals that had outpatient clinics, about half (20/37, 54.1%) reported that an e-prescribing system was in use. Relatively few respondents (5/38, 13.2%) indicated that their hospital used imaging technology for outpatient or retail prescriptions.
The pharmacy department computer system was described as a pharmacy computer system integrated into the hospital system or EMR in nearly all of the hospitals represented in the survey data (36/38, 94.7%). The majority of the hospitals (23/38, 60.5%) did not use imaging technology for inpatient order transcription.
The pharmacy directors were also asked about the use of offsite or remote medication order review or entry. Most of the respondents (26/39, 66.7%) indicated that their hospital did not utilize remote review or order entry; some respondents (11/39, 28.2%) said they used these services only when the pharmacy was closed.
Dispensing process and distribution-related services
Survey participants were asked about the hours of operation of the hospital pharmacy, and the survey data indicated a mean ± S.D. of 81.8 ± 40.5 hours per week. They were also asked how drugs were made available during the times when the pharmacy was closed. Among the 35 usable responses, methods identified included the use of automated dispensing cabinets (ADCs) (n = 25, 71.4%), the use of floor stock (n = 4, 11.4%), the use of a separate night cabinet (n = 4, 11.4%), and “nurse access to the pharmacy” (n = 1, 2.9%).
A centralized drug distribution system (e.g., traditional manual unit dose, centralized robot system) was in use at 23 of the 40 hospitals represented in the survey (57.5%), a decentralized system (e.g., satellite pharmacies, ADCs) at 7.5% (n = 3), and a hybrid of both at 35.0% (n = 14). The majority of hospitals (n = 30, 75.0%) had ADCs, with 93.3% (n = 28) of hospitals using a patient-specific medication profile. Respondents were also questioned about the location of the ADCs within the hospital; general patient care, peri-operative, and procedure areas were listed as the most common locations.
Pharmacy directors were asked if pharmacist verification of medication orders was required before the medications could be obtained by another health care provider from an ADC. A large majority of survey participants (26/30, 86.7%) responded affirmatively to this question. When asked if a bar-coding system is used to verify the medication doses before dispensing, half (20/40, 50.0%) said such a system was in use.
When asked about the type of facility the pharmacy uses for the preparation of sterile products at the hospital, 17 of 40 survey participants (42.5%) reported the use of a glove box and 8 (20.0%) reported having one or more laminar flow hoods located in a physically separate i.v. room. Thirteen of the 40 respondents (32.5%) reported having one or more laminar flow hoods located in a cleanroom, which was defined on the questionnaire as a specially constructed facility with ≤10,000 particles per cubic foot of air. Notably, only 1 respondent (2.5%) reported that one or more laminar flow hoods were located in the main pharmacy work area. Three quarters of the pharmacy directors surveyed (n = 30, 75.0%) reported that their hospital was in compliance with United States Pharmacopeia (USP) chapter 797 standards on pharmaceutical compounding of sterile products. Three respondents (7.5%) indicated that their hospital did not meet those standards, and an additional 7 (17.5%) responded that they did not know or were not sure.
Respondents were asked about the types of automation used in sterile product preparation, such as a standalone robotic device for small-volume parenterals, syringe-filling devices or pumps, and automated sterile product compounding pumps. Most respondents (34/40, 85.0%) reported using none of the automation or technology listed. When asked about the percentage of sterile product doses prepared by nurses on the nursing units, a large majority of respondents (35/39, 89.7%) reported that 0–25% of doses were prepared in that manner, while 4 indicated nurse preparation of more than 50% of sterile product doses.
The survey also examined the extent of outsourcing of preparation of pharmaceutical products. The outsourcing of the compounding of total parenteral nutrition preparations and other sterile products was frequent in the survey respondents’ hospitals (n = 19, 47.5%). The most commonly outsourced products were those for use in patient-controlled analgesia.
Pharmacy directors were asked about the use of bar-coding and other technology for drug preparation and dispensing. Some reported the use of a repackaging machine for unit dose packing (19/38, 50.0%), the use of bar-coding technology for inventory control (5/38, 13.2%), and the use of medication carousels for cassette-replenishment of ADCs (2/38, 5.3%). More than one third of the respondents (13/38, 34.2%) reported using none of the automation or technology listed. Electronic inventory control using radiofrequency identification was not in use at any of the respondents’ hospitals.
The survey participants were also asked about the use or application of telemedicine or telepharmacy technology in the hospital pharmacy department. Only 3 of the 40 respondents (7.5%) reported the use of those technologies.
The questionnaire also asked pharmacy directors about the types of technologies used in hospital clinics or outpatient pharmacies, such as outpatient pharmacy pill-counting machines, an outpatient pharmacy prescription-filling system (e.g., robot), or physician office-or clinic-based outpatient prescription dispensing machines. Most of the respondents (35/36, 97.2%) reported using none of these technologies; 1 reported the use of outpatient pharmacy pill-counting machines.
Medication administration and monitoring
Respondents were asked about methods of storing patient-specific medications on nursing units. Of the 39 pharmacy directors submitting usable response data, just over half (n = 20, 51.3%) reported that all such medications were obtained from ADCs, and the remainder reported using one or more of the following: a patient-specific drawer located in the medication room (n = 16, 41.0%), a patient-specific drawer located in the patient’s room (n = 3, 7.7%), a nurse server (i.e., a supply storage unit) located in the medication room (n = 3, 7.7%), or a nurse server located in the patient’s room (n = 3, 7.7%).
When asked about the technologies used in the drug administration process, the majority of the 39 respondents indicated the use of barcode verification of medications at the time of administration (n = 29, 74.4%) and the use of smart infusion pumps (n = 28, 71.8%). On the other hand, only 8 respondents (20.5%) reported barcode verification of medication at the time of removal from ADCs.
Survey participants were also asked who is responsible for medication education and counseling for inpatients. The vast majority of respondents (38/39, 97.4%) reported that nurses were primarily responsible for those tasks; 30.8% (n = 12) indicated that pharmacists had primary responsibility, and 17.9% (n = 7) indicated that physicians had primary responsibility for inpatient medication education and counseling.
The pharmacy department heads reported providing various medication monitoring services. Nearly all of the 39 respondents who submitted usable data in this area reported providing adverse drug reaction monitoring and reporting (n = 38, 97.4%), drug information to physicians and other health care providers (n = 37, 94.9%), and medication error reporting (n = 37, 94.9%) as the main monitoring services. Poison information services were the least frequently provided of the listed service types (n = 9, 23.1%).
Discussion
The results of this survey help characterize hospital pharmacy practice in small and rural hospitals in Illinois in 2011. Hospitals represented in the survey were, on average, smaller than in the past two surveys (2001 and 1991), with a mean average daily census of approximately 32, which is 23% lower than the mean value in 2001 and 38% lower than the mean value in 1991. This finding is consistent with our previous study and other data.12,17
Despite the size of hospitals, small and rural hospitals represented in the survey reported high rates of provision of clinical pharmacy services. Nearly all respondent hospitals reported that pharmacists provided some clinical services such as drug therapy monitoring, pharmacokinetic consultation, and patient education and counseling—though in the case of patient education and counseling, those services were provided by pharmacists to only a small portion of patients, and nursing staff were considered primarily responsible for this task in most hospitals. A majority of respondents also reported pharmacists’ participation in medical emergency responses, the provision of nutritional support, and compliance and drug histories. The extent of clinical pharmacy services in the hospitals represented in the survey is reassuring because previous studies have demonstrated that these and other pharmacy services both improve clinical outcomes and reduce costs.18,–22
For the most part, the percentage of respondents indicating that their hospital was providing each specific type of clinical pharmacy service increased from 1991 to 2001 but less so from 2001 to 2011. Between 2001 and 2011 the most dramatic increase occurred in the areas of medical emergency response and provision of compliance and drug histories. Pharmacists’ participation on medical emergency response teams has been widely discussed in the literature and is associated with improved patient safety, provision of optimized pharmacotherapy regimens, prevention of medication errors, and decreased mortality.23,–25 As a result, it is not surprising that this was found to be a commonly provided pharmacy service. Overall, more than half (54.0%) of these hospitals had pharmacists serving on emergency response teams. The increase in provision of drug histories is also not unexpected given that medication reconciliation is now required as a part of accreditation by the Joint Commission.26 The ASHP national survey on monitoring and patient education has also shown that the majority of hospitals provide drug histories and medication reconciliation as a medication safety practice.11
In addition to the provision of specific clinical services, hospitals appear to have evolved the general model that is used for deployment of pharmacists to one that is patient centered, meaning that nearly all pharmacists have distributive and clinical responsibilities. This integration of pharmacy services in small and rural hospitals is consistent with practice trends at the national level; recent survey data indicate that 64.7% of all hospitals and 67.0% of hospitals with less than 100 beds use a patient-centered integrated model.9, 27, 28
We also found significant changes in the medication distribution systems used in small and rural hospitals in Illinois. The 2001 survey data indicated that 70% of hospitals relied on a manual, centralized, unit dose system for distribution of maintenance doses, and only 22% of hospitals had a decentralized automated system using ADCs to distribute medications for patients.9 In 2011, pharmacy directors at 75% of hospitals reported using ADCs as the primary method of supplying medications.
The survey data indicate substantial use of bar-coding technology in small and rural hospitals in Illinois. Half of the survey participants reported the use of machine-readable coding to verify doses before dispensing from the pharmacy, and nearly 75% reported the verification of medication barcodes at the time of administration. In contrast, 40% of hospitals in the ASHP survey reported the use of machine-readable coding before dispensing, and 50% reported barcode-assisted medication administration.9 Increased use of these and other technologies has been associated with improved medication safety and, because of improved efficiency, may also explain how hospitals have been able to increase provision of clinical services.29
There is still further improvement to be made in the implementation of technology in small and rural hospitals in Illinois. For example, while close to 70% of hospitals represented in the survey indicated the use of a CPOE system to order medications for inpatients, a majority reported infrequent use of CPOE by physicians, suggesting that other personnel may be entering orders for physicians. Many of the clinical advantages of CPOE are lost in such a process; for example, CDS systems built into the CPOE process can play a significant role in directing prescribers toward optimal, evidence-based drug therapy.30
We also found that relatively few hospitals (about one third) were using remote pharmacist medication order entry and review. Such practices can be used by small hospitals after hours to improve access to verified medications and, where such processes are centralized, to increase efficiency.7,31
In addition, while telemedicine and telepharmacy have been widely discussed in the literature and may be particularly applicable to small and rural hospitals, only 7.5% of respondents reported using these technologies. A national survey found that the use of telepharmacy technology in rural hospitals is not widespread and its application varies across the United States.5 This finding is consistent with the fact that many states do not have regulations for the use of telepharmacy in hospitals.5 There also appears to be a lack of consensus on the definition of telepharmacy. Our questionnaire did not provide a definition; as a result, respondents may have had different interpretations of what sorts of technologies and systems can be classified under the term “telepharmacy.”
The preparation of intravenous drugs has become increasingly complex, and this has the potential to be an area of difficulty for small and rural hospitals, particularly given newer practice standards such as those specified in USP chapter 797.32 Not surprisingly, about one third (32.5%) of pharmacy directors surveyed reported that their hospital had a cleanroom, although 17.5% of respondents expressed uncertainty about compliance with USP chapter 797. These results are not dissimilar to those of a national survey indicating that many hospitals were not fully compliant with ASHP’s Guidelines on Quality Assurance for Pharmacy-Prepared Sterile Products.33 Perhaps for this reason we found that the rate of outsourcing of sterile product preparation appears to be increasing among small and rural hospitals in Illinois. In 2001, just 21% of hospital pharmacy directors surveyed reported outsourcing of medication preparation, while 43% reported that practice in the 2011 survey.9
Limitations
This study had several notable limitations. First, the overall response rate (46.5%) was not as high as in our previous surveys (66% in 2001 and 81% in 1991).12,13 This decline is consistent with other literature.34 The generalizability of survey data can be assessed according to the degree to which the sampled population differs from the survey population as a whole (i.e., the degree of non-response bias).35 We investigated the potential for nonresponse bias by comparing selected characteristics of the survey respondents’ hospitals with abstracted data from the 2012 AHA Guide to the Health Care Field for all 86 hospitals in our survey population.16 Variables abstracted were the average daily census, annual number of admissions, outpatient visits, and average length of stay. The values for hospitals not represented in the survey did not differ drastically from the values for nonrepresented hospitals for any of these variables.
Another potential limitation was the use of the IHA list to define the population of small and rural hospitals. This approach differed from that used in 1991 and 2001, which focused strictly on rural hospitals. However, while the IHA list included some hospitals that were small but not rural, the majority of hospitals represented in our survey (72.5%) were rural. Further, the definition used by IHA to determine rurality (i.e., location outside an MSA) was the same as that used in our previous surveys, and we believe our results are therefore comparable across surveys.
Last, there is the possibility that some of the items in our questionnaire were unclear or misinterpreted by respondents, leading to response errors. We included definitions within the questionnaire wherever we felt terms might be interpreted in multiple ways. However, some terms in the questionnaire were not defined. We attempted to validate some of the responses by comparing hospital demographic data reported by pharmacy directors with data reported in the 2012 AHA Guide. While we found some differences for individual hospitals, the mean values were not significantly different. Further, many of the items in our questionnaire were the same or similar to those used in previously published surveys.
Conclusion
A 2011 survey of pharmacy departments in small and rural Illinois hospitals provided information on the use of automation and health information technologies and showed changes in the provision of many clinical pharmacy services since 1991.
Footnotes
This is the first in a new series of articles focusing on pharmacy practice in small and rural hospitals. Two of the authors of this article, Stephanie Y. Crawford, Ph.D., M.P.H., B.S.Pharm., and Glen T. Schumock, Pharm.D., M.B.A., Ph.D., will serve as field editors for the series, soliciting and reviewing future articles. Unsolicited submissions are also welcome.
The authors have declared no potential conflicts of interest.
The authors acknowledge Scott A. Meyers, M.S., and the Illinois Council of Health-System Pharmacists for their support of this project; Michael G. Weaver, M.B.A., for assistance in reviewing the questionnaire; the pharmacy directors who completed the questionnaire; the Illinois Hospital Association for providing the list of small and rural hospitals; and the Illinois Department of Financial and Professional Regulation for providing data on licensed pharmacies in Illinois.
References
Comments