| Adverse Medication Event History Clarification: A Multidisciplinary Approach | Jerry R. McKee |
Setting: 450 bed intermediate care facility for the mentally retarded.
Practice Description: Four pharmacists, three technicians, and an office assistant serve this facility, which is home to a variety of multiply disabled adults. Pharmacists are involved clinically in quarterly treatment team meetings for all residents.
Practice Innovation: A pharmacist-led multidisciplinary program to review and categorize resident adverse medication event histories is described. Using facility-defined criteria, adverse medication event histories were reviewed and categorized into allergy/hypersensitivity, medication side effect, misinformation, and no known allergy categories by the multidisciplinary treatment team. The revised data are entered into the pharmacy computer system so that all new medication orders can be screened for potential allergies. This information is also printed on client documents maintained in the medical record.
Main Outcome Measurements: Pre- and post-analysis assessments were made and compared for each resident regarding documented adverse medication experience information (no data, no known allergies, medication alert).
Results: A total of 25 previously undocumented allergies were detected by this process, most for medications commonly prescribed at this facility. Some 29.9% of residents with no documented allergy information were updated through this project.
Conclusion: Clearly defined adverse medication event histories must be delineated for anyone receiving medications to avoid preventable medication misadventures
Key Words: Allergy, Adverse drug reaction, Clinical pharmacy services, Medication misadventures, Medication use.
Abbreviations Used: ADR = Adverse drug reaction; NKA = No known allergies.
Consult Pharm 1996; 11:942-46.
Patients with multiple disabilities often require many different medications, suffer from coexisting illnesses, and have othercharacteristics that increase their risk for an adverse drug reaction, also known as an adverse medication event or experience. Among hospitalized patients, the reported incidence of ADRs ranges from 1% to 30%.1-5 ADRs are a major cause of morbidity and in some cases mortality.6 Clearly, such medication misadventures also have a profoundly negative impact on the quality of life of those affected. Billions of health care dollars are spent annually as a result of patient noncompliance, preventable adverse medication effects, and harmful drug interactions, which often lead to hospitalization. Estimates are that one third of ADRs could be prevented.7 Therefore, preventing drug misadventuring is a clearly defined quality improvement issue in the arena of medication use.7,8
One of the problems with reporting and documenting adverse medication events is that of terminology. The term ADR means different things to different groups. The World Health Organization defines ADR as a "noxious and unintended [reaction that] occurs at doses normally used in man for the prophylaxis, diagnosis, or therapy of disease, or for the modification of physiological function." Attempts to distinguish true hypersensitivity reactions (allergic reactions) from other ADRs has been complicated by the lack of standardized terminology among health care professionals. Only 6%-10% of all ADRs result from immunologic causes, making them true allergic reactions.9
Western Carolina Center is a 450-bed residential intermediate-care facility for the mentally retarded. The population comprises multiply disabled individuals in the severe to profound range of mental retardation. With an average length of stay of 17 years, a great volume of medical information accumulates in the resident's records. This is compounded by families often being poor historians (upon admission) with regard to past medical and medication history.
The lack of clarity regarding appropriate categorization of ADR history information at this facility was identified when the pharmacy department switched to a computer system that allowed codification of medication allergy information and automatic screening of all new medication orders for possible allergies. A review of the database from the previous computer system indicated that, of 417 patients, 125 had no data listed (missing classification), 124 had "NKA-no known allergies" listed, and 168 had "med alert" information listed. "Med alert" information, as defined by the previous computer system, was adverse medication event data. Close inspection found these data to be a mix of allergy or hypersensitivity data, side effect information, and misinformation. A lack of information (29.9% with no data) could cause pharmacists to overlook possible medication-related allergic reactions, leading to preventable ADRs and medication misadventures. Likewise, operating on misinformation could have the equivalent result. A serious quality improvement issue was thus identified at the facility, one that pharmacists could assist in addressing. The scope of the problem having been identified, a potential solution was developed using a multidisciplinary team approach.
The purpose of this report is to outline the process of analyzing adverse medication event histories in a long-term care facility for the developmentally disabled. The objective of the project was to categorize these data as of allergies, predictable medication side effects, or misinformation as a means of improving risk management in the area of pharmacotherapy.
Methods
A multidisciplinary task force, including the directors of pharmacy, nursing, and medicine at the facility focused on the importance of clarifying existing patient "med alert" data into "NKA," allergies/hypersensitivity reactions, medication side effects, and misinformation categories. The implications of not addressing this issue (continued risk of preventable adverse drug experiences) were discussed. The protocol for conducting this process was developed along with identification of the mechanism for carrying out the project.
A multidisciplinary treatment team meeting is held at least quarterly for all residents of the facility. At that meeting, a pharmacist, nurse, physician, psychologist, team leader, and social worker discuss each resident's drug therapy, medical status, and related quality of life issues in an effort to address holistically each resident's care. Tying the screening into an existing function increased its chance of success and saved time overall.
In the treatment-team setting, almost all documentation is readily available to team members regarding adverse medication event history (with other historical data in central medical records, if needed). We used these operating definitions:
Clearly defined hypersensitivity reactions were documented on a red and white allergy sticker and placed on the outside of the resident's medical record. Those with no history of allergies were noted as NKA on the same type sticker. Medication side effects and other relevant medication-related adverse event information (including food or environmental allergies) were recorded on a separate "Med Alert" sticker and placed on the inside cover of the medical record. If known, a specific date was documented for each allergy or side effect noted. All pre-existing stickers were removed when the process was completed for a specific resident.
The pharmacy database, which screens automatically for medication allergies, was subsequently updated to correspond to the outcome of each resident's adverse medication event history screening discussion. This database is, in turn, used to print current allergy and Med Alert information on residents' medication administration records, quarterly medication order updates, and quarterly drug-regimen review worksheets, all of which are kept in the resident's permanent medical record.
Results and Discussion
Using the above outlined definitions, an initial three-month screening and review process was conducted. Some resistance in changing existing information was encoun-tered from medical and nursing personnel. The concerns were primarily related to questionable increased liability risks if existing medical record information was altered. In these few instances, attempts were made to address the concerns. When differences of opinion could not be resolved, appropriate steps were taken to assure that all databases were consistent with the existing information.
A second three-month period was then used for teams to review the accuracy of resulting documentation on the resident charts and pharmacy documents as a final quality assurance measure (through quarterly meetings). In completing the multidisciplinary review, we found that the pharmacy database and the resident record data varied in a large percentage of cases.
A post hoc analysis of the data is displayed in Table 1. Of the 125 individuals with no data initially, 2.6% had medication allergies. Of 124 patients originally classified as NKA, 1.4% had allergies. Of the remaining 168 cases who had Med Alert information initially, 23.1% had allergies and 2.6% had NKA.
TABLE 1. Comparison of Pre- and Post-Screening of Adverse Medication Event Information
| % Residents (n = 417) | |||
|---|---|---|---|
| Postscreening Classification | No Known Allergy (n=124) | Medication Alert (n=168) | Missing Classification (n=125) |
| No known allergy | 16.9 | 2.6 | 16.7 |
| Allergy | 1.4 | 8.1 | 2.6 |
| Medication alert | 10.2 | 14.2 | 9.5 |
| Allergy and alert | 0.9 | 15.0 | 0.9 |
Two primary categories of error were identified during the adverse event history screening process (Table 2). Errors of omission were determined for the 29.9% of cases that originally were missing classification status. Errors of commission comprised 9.1% of cases, consisting of both false negatives (no information) or false positive (misinformation). Of residents originally categorized as NKA, 6% had allergies (false negative because of incorrect information). Of the 3.1% who were false-positive cases, 12 cases were found to have NKA, while two cases were found by the team to have documented medication allergies.
Table 2. Types of Error Detected During Adverse Medication Event History Screening Process
| Error Type | No.Cases | % Total Cases |
| Errors of omission | ||
|
| 125 | 29.9 |
| Errors of commission | ||
|
|
25 | 6.0 |
|
| 13 | 3.1 |
| Total errors | 163 | 39.0 |
Post-evaluation, 29% of residents had documented medication allergy or hypersensitivity (Table 3). Some 37% had NKA, while 34% had some type of Med Alert information. The 25 residents with newly defined allergies were hypersensitive to 28 different medications. Most allergies identified were to medications commonly prescribed at this facility (Table 4). This finding again points out the importance of identification and documentation of medication allergy data to minimize preventable ADRs.
Table 3. Prescreening and Postscreening Classifications for Misclassified Cases
| % Residents (n = 417) | |||
|---|---|---|---|
| Original Classification | Post-Screening Classification | Misclassified | |
| No known allergy | 29.7 | 36.4 | 12.7 |
| Allergy | 0 | 12.2 | 12.2 |
| Medication alert | 40.3 | 34.4 | 25.8 |
| Allergy and alert | 0 | 17.0 | 17.0 |
Table 4. Medication Categories Identified by Screening for Drug Allergies
| Medication Category | No. Residents Identified |
| Antipsychotics/antiepileptics/antidepressants | 7 |
| Antibiotics (penicillins, cephalosporins, macrolides, sulfa drugs, aminoglycosides) | 13 |
| Analgesics (nonsteroidal and narcotic agents) | 5 |
| Intravenous imaging dyes | 3 |
Costs can be avoided and quality of life maintained by minimizing medication misadventures. Clinicians and administrators are jointly responsible for applying current technology and resources in fulfilling this responsibility. One method for improving baseline adverse medication event data for newly admitted patients is to have a pharmacist perform a thorough and systematic medication history with the families or guardians. This is standard procedure at our facility. The integrity of this baseline information can be maintained and updated by clear communication between pharmacy, nursing, and medical personnel.
The treatment team model is an ideal format for such an exchange to occur. In the absence of treatment teams, telephone or other oral communication may occur, or a physician's order may be used to communicate the change in information status. An effective ADR reporting system, with ongoing educational efforts regarding the importance of such reporting, must supplement any method of information sharing.
The adverse medication event clarification process described in this report involves pharmacists in an area in which they have clear expertise-management of drug experience information and its application in day-to-day practice-as a means of promoting safe, effective, and rational drug therapy. Many other types of long-term care facilities share in the problem of having adverse medication event misinformation accumulate over time. The experience of Western Carolina Center suggests that, periodically, health care professionals should critically analyze this information as an ongoing quality improvement project, to assure that the data are current, valid, and consistent across the medical record and computer information systems used by the various disciplines.
Conclusion
This adverse medication event history screening process identified 40% of residents of an ICF-MR as having data-classification errors, with the majority of errors (77%) being ones of omission. Classification errors were found in all categories reviewed. Many cases previously documented as either NKA or missing classification were found to have allergies, Med Alerts, or both.
Clearly defined adverse medication event histories must be delineated for anyone receiving medications to avoid preventable medication misadventures. Equally important is assuring the ongoing validity of this information.
References
1. Melmon KL. Preventable drug reactions-causes and cures. N Engl J Med 1971; 284: 1361-8.
2. Linkewich JA. Adverse effects of drugs. In: Osol A, ed. Remington's pharmaceutical sciences. 17th ed. Easton, PA: Mack Publishing Co., 1985; 132-5.
3. Visconti JA, Smith MC. The role of hospital personnel in reporting adverse drug reactions. Am J Hosp Pharm 1967; 24: 273-5.
4. Lawson LA, Pruemer JM. Assessment and management of adverse drug reactions. Am Drug 1986; 193: 129-34.
5. Berry LL, Segal R, Sherrin TP et al. Sensitivity and specificity of three methods of detecting adverse drug reactions. Am J Hosp Pharm 1988; 45: 1534-9.
6. Allen WO, Calder WF. Adverse drug reactions. Hosp Pharm Cost Contain 1987; 5(Nov): 1-4.
7. Bates DW, Cullen DJ, Laird N et al. Incidence of adverse drug events and potential adverse drug events. JAMA 1995; 274: 29-34.
8. Manasse HR Jr. Medication use in an imperfect world: drug misadventuring as an issue of public policy, Part 1. Am J Hosp Pharm 1989; 46: 929-44.
9. Guill MF. Allergic drug reactions: identification and management. Hosp Formul 1991; 26: 582-9.
Jerry R. McKee, RPh, MS, is Director of Pharmacy Services, Western Carolina Center,Morganton, North Carolina.
Address for reprints: Jerry R. McKee, Director of Pharmacy Services, Western Carolina Center, 300 Enola Road, Morganton, NC 28655.
Acknowledgments: To James
Bodfish, PhD, for statistical analysis of data, and Melvin Cline,
RPh, Robert Allen, RPh, MS, and Robert Lowe, PharmD, for manuscript
preparation.