Stephen G. Bryant

Pharmacists' Dispensing Accuracy in a High-Volume Outpatient Pharmacy Service: Focus on Risk Management:

A 12-day peer-review audit was performed in the outpatient pharmacy of a large teaching hospital. The audit process was not masked, that is, the pharmacists were aware of the peer-review evaluation. During the 12-day period, 9394 prescription forms and their corresponding pharmaceutical products were examined manually before being delivered to the patient. A total of 1165 (12.4 percent) dispensing errors were detected, with 141 (1.5 percent) of these considered potentially serious. Seventy-six prescriptions contained two errors and four prescriptions contained three. A linear relationship (r2 = 0.78; P < 0.001) existed between the number of potentially serious errors and the total number of prescriptions filled. There were no statistically significant differences in the dispensing-error rate for the eight pharmacists audited. There was a trend for the number of pharmacist-hours containing at least one potentially serious dispensing error to increase as the prescription-filling rate accelerated. Outpatient pharmacies with high volumes should set a limit to the number of prescriptions filled by their pharmacists and should experiment with quality assurance systems to reduce dispensing errors and subsequent legal liabilities.

Postmarketing surveillance: accuracy of patient drug attribution judgments.

Data from two samples of ambulatory patients participating in a postmarketing surveillance study, one receiving antibiotics and another receiving tricyclic antidepressant agents, are presented, indicating that patients appear to be capable of correctly discriminating probable adverse drug reactions from other adverse clinical events. However, attribution accuracy depended both on the surveillance method and on how reports of the adverse clinical events were obtained. Discrimination was better when patients were reporting adverse clinical events spontaneously than when the interviewer probed for recall in a systematic inquiry. Discrimination was also better when the adverse clinical events were obtained from a staff‐initiated surveillance method than from a self‐monitoring, patient‐initiated telephone‐reporting method–probably because the latter method generates an excessively strong tendency to report mainly those adverse clinical events suspected of being drug related.

Increased frequency of doxycycline side effects.

As part of a study investigating new methods of detecting adverse drug reactions, adult outpatients receiving new prescriptions for 1 of 4 antibiotics (doxycycline, penicillin VK, ampicillin, or tetracycline) were assigned to 1 of 2 methods of monitoring adverse drug reactions. Data were collected from a total of 457 staff‐initiated interviews and from 1467 patients who were asked to telephone to report possible adverse effects. Doxycycline is considered by authoritative sources to be generally comparable in side effects to other tetracyclines and penicillins, with the exception that it increases photosensitivity. Results from both monitoring methods consistently indicated at least a 3‐fold higher frequency of nausea or vomiting with doxycycline relative to the other antibiotics. Complaints of skin rash also were at least 4 times more frequent with doxycycline, depending on the particular sample.

Postmarketing surveillance by patient self-monitoring : trazodone versus fluoxetine

This article presents incidence estimates and relative risks for a number of adverse clinical events reported by outpatients being treated with either trazodone or fluoxetine. Data were collected via an innovative method of patient self-monitoring. Many of the suggested differences between the two drugs are quite consistent with expected adverse drug reactions documented in both the package insert data for outpatients and with reports in the literature. Findings not so readily anticipated for trazodone, however, include higher relative frequencies for muscle weakness or soreness, skin swelling, and urinary complaints compared with fluoxetine; for fluoxetine, data are presented indicating a higher incidence of various psychologic/psychiatric adverse clinical events, including delusions and hallucinations, aggression, and suicidal ideation. Other possible interpretations of the results are discussed.

Patient Drug Attributions and Postmarketing Surveillance

Although studies have shown that patients can distinguish probable adverse drug reactions (ADRs) from adverse clinical events (ACEs) caused by other factors, it is not known whether these attribution judgments add any independent validity to other accepted methods of identifying ADRs, such as physician assessments or epidemiologic data. Data from 2487 patients receiving fluoxetine and 815 receiving trazodone were used to see whether such information was redundant when added to standard statistical analysis directed toward detecting ADRs. Relative risk values for 14 trazodone or fluoxetine ADRs were selected because each was significantly identified by an innovative postmarketing surveillance system. In one analysis, all patient reports were used to compute relative risk; in the other, only reports attributed by patients to the study drug were included. Results indicate that taking into account patient attribution judgments results in a consistent, albeit modest, increase in the discriminatory power of this monitoring method.

Monoamine Oxidase Inhibitors: Safety and Efficacy Issues:

Controversy has historically centered on the use of monoamine oxidase inhibitors (MAOI) in the treatment of depression because of safety and efficacy issues. Hypertensive crisis following ingestion of foods containing tyramine is the most feared problem associated with MAOI therapy. The authors conclude that only four tyramine-containing foods clearly warrant absolute prohibition, but indicate situations where moderation should apply. Although MAOI remain second-line agents in the treatment of endogenous depression, it is unclear whether MAOI or tricyclic antidepressants should be the drugs of choice in atypical depression and this question necessitates future research.

Detecting Adverse Drug Reactions in Postmarketing Surveillance: Interview Validity

As part of a large-scale study developing new methods of postmarketing surveillance, computer-assisted interview information was collected from 231 outpatients receiving a target drug chosen from two pharmacological classes (antibiotics and tricyclic antidepressants) for which adverse drug reactions (ADRs) are well-documented. The standardized telephone interview used to describe and quantify all adverse clinical events (ACEs) has some questions that focus only on spontaneously reported ACEs and other detailed questions covering various specific organ systems. Analyses show that spontaneously reported ACEs generally provide the most reliable source of true ADRs. The results have important implications bearing upon the ease with which a set of ACEs may be inappropriately interpreted as possible adverse drug reactions.

Patient-Initiated Postmarketing Surveillance: A Validation Study

A new patient‐initiated, pharmacy‐based postmarketing surveillance system is described. At the time a new prescription for a targeted drug was filled, 2705 outpatients (experimentals) randomly assigned to the new system had a printed notice attached to their medication bags: the information requested them to report any “new or unusual symptoms” during the next 2 weeks by a toll‐free telephone number to a trained nonprofessional who conducted a standardized adverse drug reaction (ADR) interview. To help validate the new system, another sample of 1109 patients (controls) did not receive a request for self‐monitoring but were interviewed by telephone 2 weeks later. Target drugs were chosen from two classes for which side effect profiles are well identified: oral antibiotics and tricyclic antidepressants. Results show that within both drug classes, all patient‐initiated reports closely matched those obtained from controls; the experimental and control groups also reported predictably high relative frequencies for the most commonly expected ADRs. Additional analyses suggest that a patient‐initiated monitoring system could prove to be a promising complement to existing physician‐based surveillance systems.

Long-term versus short-term amitriptyline side effects as measured by a postmarketing surveillance system.

As part of a large-scale postmarketing surveillance study, the adverse clinical events (ACEs) reported by 85 outpatients taking amitriptyline were investigated. Two discrete groups of patients were identified based on their duration of amitriptyline treatment: 45 had started the drug within 2 weeks of their interview (mean = 10.1 days, SD = 1.6 days), while 40 were much longer term tricyclic antidepressant patients (mean = 227.2 days,SD = 135 days). Our analysis of amitriptyline side effects reported by each of these two discrete groups challenges the common clinical impression that tricyclic side effects, in general, abate with continued treatment. Shorter term patients were much better able to correctly attribute their adverse clinical events to their drug therapy. Anticholinergic side effects were reported as new symptoms by the long-term patients just as frequently with similar ratings of subjective severity. These reports of adverse drug reactions of recent onset by long-term amitriptyline users may reflect the fact that such symptoms fluctuate in their occurrence and may not be recognized as potentially drug-induced until some threshold for patient tolerance is exceeded.

Postmarketing Surveillance Of Adverse Drug Reactions: Patient Self-Monitoring

Background: This report summarizes our experience with a new approach to postmarketing drug surveillance using a pharmacy-based patient self-monitoring strategy, developed in collaboration with Eckerd Drug Company, the American Association of Retired Persons Pharmacy Service, and other pharmacies nationwide. Methods: Patients presenting prescriptions to collaborating pharmacies for a targeted drug or a standard drug used as a control received an entry form asking them to register and then call a toll-free telephone number to report possible drug reactions. When contacted by patients, study staff conducted a standardized, computer-driven interview. Two brief mail questionnaires were also employed. Results: Original validation data gathered from 1984 through 1986 indicated that the most commonly expected adverse drug reactions caused by antibiotic and tricyclic antidepressants reported by 162 self-monitoring patients closely matched those elicited from a comparable control sample of 1109 patients who were independently interviewed by our staff. Results from subsequent studies are also described. Conclusions: We believe this method has great promise for providing not only a cost-effective, complementary, early alerting mechanism for detecting adverse drug reactions, but also the additional possibility for discovering unsuspected therapeutic benefits of newly marketed drugs.