Correlation of employees' perception of workload with workload indicators.

In this study, the measurement of perception of pharmacists' workload did not correlate well with any workload indicators studied. Generally, census on the study date correlated well for pharmacy technicians working in various areas of the pharmacy. If it is agreed that perception of workload is the real measure of activity--that is, how busy the employees believe it is--then the indicators used in this study are not useful. Similar research using other indicators is warranted.

Evolving to provide pharmaceutical care without additional resources in a university hospital.

Our department is committed to a process of continuous quality improvement focusing on delivering the best possible pharmaceutical care services. Three committees, each with representation from pharmacists and pharmacy technicians from all areas of the department, were convened in 1991 to 1992 to further identify areas for service enhancement and to plan for the future. Based on the recommendations of these committees, further expansion in ambulatory services is a priority. Other changes will include further automation of the drug-distribution system, examination of the role of the pharmacy technician, development of an automated patient-care system with direct physician order entry, computerized documentation of clinical interventions, and reaffirmation of the role of the pharmacist as the professional responsible for proper use of medications at UMHC.

Implementation, justification, and expansion of ambulatory clinical pharmacy services.

In response to the current trends in health care utilization and provision, The UMHC Department of Pharmaceutical Services has as part of its mission the provision of comprehensive pharmaceutical services to its ambulatory patients. To achieve this, the Department began the process of identification of need and the provision of these services. Simultaneously, the Department is conducting research to document the impact that pharmacists can have on the cost and quality of care when they are actively involved in the therapeutic management of patients treated in the ambulatory setting.

Checking of unit dose cassettes by pharmacy technicians at three Minnesota hospitals.

A pilot project in which pharmacy technicians were trained to check unit dose cassettes filled by other technicians is described. With the approval of the state board of pharmacy, the Minnesota Society of Hospital Pharmacists (MSHP) conducted the nine-month project in three hospitals with different types of unit dose drug distribution systems. Twenty-seven technicians underwent didactic and practical training and were then validated as checkers if they scored 99.8% accuracy in checking carts into which errors had been deliberately introduced by the pharmacist auditor. The performance of validated technicians was audited monthly, and failed audits had to be repeated. Participating technicians did not check the preparation of first doses or extemporaneously prepared doses. In 100,000 doses audited, 60 errors by the validated checkers were identified. Of six technicians who failed a monthly audit, five passed a repeat audit. Pharmacists at the participating hospitals documented time they spent on clinical activities that would have been spent checking cassettes. In December 1990 a one-year extension of the project, expanded to 10 hospitals, began. With strict quality control measures, specially selected and trained pharmacy technicians performed unit dose cassette checking with an accuracy of at least 99.94%.

Effect of computer software on time required to prepare parenteral nutrient solutions.

Pharmacist and technician times required for the preparation of parenteral nutrient (PN) solutions were studied before and after the introduction of a software package that is used with an automated compounder. At a 580-bed teaching hospital, work sampling was used to collect data on how time was spent by personnel in the i.v. admixture pharmacy during six-week periods before and after the introduction of software that calculates the quantities of PN solution ingredients on a computer-generated work sheet and prints labels. The second data-collection period began when the software had been in use for four months. In the second study period, there was a significant decrease (28%) in total pharmacist time spent per PN solution; the mean +/- S.D. pharmacist times per PN solution per day for the two study periods were 14.03 +/- 3.24 minutes and 10.12 +/- 1.61 minutes, respectively. There were also significant decreases in pharmacist time spent performing calculations, checking calculations, and typing labels. After introduction of the software, technicians spent significantly less time typing labels and pumping base solutions. Overall, technicians spent significantly less time per PN solution in the second study period (20.15 +/- 3.50 versus 17.82 +/- 1.94 minutes). Use of the software allowed pharmacist staffing in the i.v. admixture pharmacy to be reduced, and the pharmacy resources were reallocated toward the provision of clinical services. The generation of PN labels and calculation worksheets by computer software reduced pharmacist and technician time requirements for PN solution preparation.

Comparison of automated and manual methods of syringe filling.

A study to measure the time and cost associated with an automated and a manual method of syringe filling is reported. A stopwatch was used to measure the time needed by an experienced pharmacy technician to prepare batches of 200 syringes of each of seven drugs by a manual method and an automated method, the Multi-Ad Fluid Dispensing System. For each drug and method, time-and-motion data were collected during the preparation of four batches. The accuracy of each method was determined by dividing the actual by the expected number of syringes filled per batch. Material costs were calculated by summing the contract costs of the necessary equipment. The total cost of each method was determined by adding the labor and material costs. For all the drugs, the mean total time required to prepare one batch of syringes by the automated method was significantly less than that for the manual method. There was no significant difference in accuracy between methods for any of the drugs. The annual labor costs of the automated and manual methods were $4056 and $5761, respectively, and the annual material costs were $3364 and $2260, respectively. The total annual cost of the automated method was $7419, compared with $8021 for the manual method. The Multi-Ad system was significantly faster and somewhat less costly overall than a manual method for batch preparation of syringes of seven drugs.

Quantifying unused orders for as-needed medications.

Nonuse of as-needed (p.r.n.) medication orders in a university hospital was quantified. A total of 882 patient charts for the last six months of 1984 were reviewed (seven randomly selected patients per month from each of 21 medical-surgical services). Orders for p.r.n. medications were quantified by therapeutic category and by whether the order was written on admission. Use or nonuse of orders was determined from pharmacy records of doses administered. Of 7735 p.r.n. orders, 4793 (62%) were unused. By therapeutic category, antacids were the least prescribed p.r.n. medications but also had the highest rate of nonuse. In every therapeutic category, the rate of unused orders was higher for orders written on the day of admission than for subsequent orders. By patient's medical-surgical service, the percentages of p.r.n. orders unused ranged from 50% for renal transplant patients to 81% for ophthalmology patients. Nonuse of p.r.n. medications in all therapeutic categories decreased as length of stay increased; the overall rate of nonuse was 80% for patients hospitalized for two days or less. Reduction of the number of p.r.n. orders written but not used should begin in two areas: orders written on day of admission, and orders for patients with short hospital stays.

Recording schedule II drug use in a decentralized drug distribution system.

A decentralized system for Schedule II controlled substance distribution and accounting that does not rely on proof-of-use sheets is described. Controlled substances are kept in a vault in the central pharmacy; technicians assigned to work in the controlled substances vault are responsible for monitoring, ordering, and storing these medications. The narcotic vault technicians also prepare narcotic boxes that are used by technicians in patient-care areas to transport and issue Schedule Ii substances to nursing units. Twice during each morning and evening shift, technicians in patient-care areas visit each nursing station, replenish the unit's stock of Schedule II medications, and document on a narcotic use form all doses that have been administered since the last technician visit. Nurses leave small cards preprinted with patients' names and room information in the narcotic drawers to alert technicians to administered doses of Schedule II substances; the technicians are responsible for reconciling medications missing from the drawers with doses recorded in patients' medication administration records as being administered. Periodic audits are performed to ensure that actual inventories and the recorded information are correct. The decentralized system for distribution and accounting of Schedule II substances has been successful in increasing the flexibility of inventories on individual nursing units and ensuring maximal pharmacy department control over the dispensation of these medications.

Chemical and statistical considerations in the determination of partition coefficients of weakly ionizable drugs and poisons.

Equations have been developed that relate the concentration (or a parameter directly proportional to concentration, such as optical absorbance) of a weakly ionizable solute in a water-immiscible phase, in equilibrium with an aqueous phase, to the pH of the aqueous phase, the partition coefficient of the unionized solute and the phase volume ratio. These relationships have been used in the design of experimental methods for determining partition coefficients, which require measurement of solute concentration in only one phase. Data obtained in this way permit ready recognition of deviations from assumptions made in the development of the model; these assumptions include insolubility of the ionized solute in the water-immiscible phase and lack of interaction between buffer components and solute. Conditions for optimal liquid-liquid extraction of weakly ionizable solutes are more easily recognized. With these techniques, the negative logarithm of the acid dissociation constant (pK'a) and the logarithm of the octanol-water partition coefficient (log P) have been measured for warfarin (pK'a = 5.15 +/- 0.04; log P = 2.82 +/- 0.06), strychnine (pK'a = 8.29 +/- 0.02; log P = 2.23 +/- 0.04), phenol (pK'a = 9.88 +/- 0.02; log P = 1.75 +/- 0.05), procaine (pK'a = 8.11 +/- 0.04; log P = 1.10 +/- 0.08), and ephedrine (pK'a = 9.92 +/- 0.01; log P = 1.65 +/- 0.04) at 21 degrees C.