Prescription of contraindicated and interacting drugs in elderly patients admitted to hospital

Central dimensions of clinical practice evaluation: efficiency, appropriateness and effectiveness--II

That a treatment selected for a given condition works, or that it works better than alternative treatments, or that it was selected because it works as well as but is cheaper than alternative treatments, should be of pivotal concern to clinicians and is of central concern to patients and to health care managers. Attempts to address these concerns have resulted in what is now widely termed the 'effectiveness movement'. The protagonists of the movement have been concerned to create a culture of evaluation and inquiry within which the formulation of evidence-based clinical guidelines and their introduction into routine practice have played a prominent part. The need to ensure cost effectiveness of clinical intervention has been at least as emphasized as the need to ensure the clinical effectiveness of health care interventions. Although cost-effectiveness analyses are now an indispensable feature of practice guideline formulation and treatment evaluation, few studies have examined any deterioration in patient outcome associated with successful cost containment. An adequate understanding of the concept of clinical effectiveness and the associated aims of the 'effectiveness movement' is central to an understanding of the future nature and extent of health service provision, not simply in the UK but also internationally. Having examined the concepts of efficiency and appropriateness previously (O'Neill, Miles & Polychronis 1996, Journal of Evaluation in Clinical Practice 2, 13-27) we move in this second of two articles to a detailed explanation of the concept of effectiveness, and to an examination of the derivation and use of clinical practice guideline, concluding with a consideration of the role of practice guidelines in ensuring the cost effectiveness of health care intervention. The reservation is expressed that a 'guidelines culture', when established, will be manipulated by health care commissioners for largely political purposes, creating a systematic bias in the purchasing process that will actively disadvantage a range of patient groups.

An intervention on discharge polypharmacy

OBJECTIVE

To determine if providing a way to cancel pre-admission prescriptions would reduce the number of active drug prescriptions (RXs) at discharge.

DESIGN

A randomized non-blinded clinical trial.

SETTING

Inpatient acute medical service of a university affiliated Veterans Administration medical center.

PARTICIPANTS

Twelve medicine ward teams were randomized to control and intervention groups. Patients controlled had been discharged from these teams during 12 weeks and were receiving outpatient medications from this facility at hospital admission; control = 180, intervention = 168.

INTERVENTION

At discharge, intervention teams used a computer-generated drug list to cancel or renew previous outpatient RXs or to prescribe new medications. Control teams could not cancel outpatient drugs and wrote all medications on individual prescriptions.

MEASUREMENTS

The difference between admission and discharge RXs.

RESULTS

There were no significant differences in patients' age, sex, race, Charlson Index (CI), or LOS between patient groups at discharge. The intervention group had fewer RXs on admission (5.4 vs 6.2, P < .05) and at discharge was not significantly different (2.9 vs 2.9, P = .87) from the control group.

CONCLUSIONS

Providing a method for canceling pre-admission medications did not reduce the number of RXs at discharge. Further research is needed to evaluate the appropriateness of the large increase in RXs from admission to discharge for patients in acute hospital settings.

Potential drug--drug interactions in 5,125 mostly elderly out-patients in Gothenburg, Sweden

Of 5,125 mostly elderly out-patients, average age 78.2 years, 1,594 (31%) had at least one interacting drug combination according to the Swedish National Formulary. On average, each patient with drug-drug interactions (DDIs) had 1.6 DDIs. There was no difference in the sex or age between those having DDIs and those without. However, patients with one or more DDIs used a significantly larger number of drugs than those without DDIs, on average 8.1 versus 5.2. This significant difference was independent of age and sex. When classified for clinical significance according to the handbook "Drug Interactions and Updates" by Hansten and Horn, 155 (3%) patients had interactions of 'major clinical significance'. The most common were interactions between beta-adrenergic blockers and antidiabetics, followed by potassium-sparing diuretics and potassium, and carbamazepine and dextropropoxyphene. The methodology allows us to screen routinely for DDIs and to plan further studies with emphasis on clinical outcome.

Prescription drug dependence in the elderly population: demographic and clinical features of 100 inpatients

OBJECTIVE

To determine the prevalence of prescription drug dependence among elderly persons in an inpatient treatment setting, to identify apparent risk factors for drug dependence, and to ascertain what factors led to admission of these patients.

DESIGN

We reviewed the medical records of 100 elderly patients dependent on prescription drugs who were admitted to the Mayo Inpatient Addiction Program between 1974 and 1993.

MATERIAL AND METHODS

Demographic features, chronic medical disorders, categories of substance dependence, diagnoses of mental disorders, and Minnesota Multiphasic Personality Inventory data were compiled and analyzed.

RESULTS

The mean annual admissions rates for three substance use disorder groups among all elderly persons treated during the 20-year period of study were as follows: alcohol only, 72%; prescription drugs, 16%; and both alcohol and drugs, 12%. The group as a whole was socially intact. Female gender seemed to be a risk factor for drug dependence. By several measures, these elderly patients were characterized as a psychiatric population. The most frequent drug dependence involved sedatives or hypnotics. General medical data did not suggest that these elderly persons were more physically impaired than the general population.

CONCLUSION

In elderly patients, awareness of coexistent diagnoses is essential in avoiding the inappropriate administration of multiple pharmaceutical agents and the possible risk of associated drug abuse and dependence.

Distinguishing between the fit and frail elderly, and optimising pharmacotherapy

Frail older patients are at risk for adverse consequences from medications or other external stresses. No single marker, such as age or physical disability, or laboratory test can identify this group of patients. As a result, screening questionnaires have been developed and successfully used by nurses to help identify frail older patients upon admission to a hospital. A very short, 7-item screen with questions concerning cognitive ability, physical mobility, nutrition, number of medications used and hospitalisation within the previous month, was able to identify those patients who were more likely to be discharged to a nursing home, die, or incur a large hospitalisation cost for the institution. While the number of medications used was not an independent predictor of the outcome measures studied (e.g. discharge to a nursing home), data from the literature show that the number of medications prescribed is related to iatrogenic complications in older patients, and specific impairments in mobility and cognition. The proper choice and prescribed dose of a medication is extremely important in frail older patients who, for instance, are at increased risk from hip fracture with some benzodiazepines, and who have markedly diminished clearance of some drugs. A systematic approach is suggested for the prescription of medications in frail older persons which will help achieve optimal pharmacotherapy by using a limited number of medications, thoughtfully selecting medications which will not impair function, and prescribing an appropriate dose based on pharmacodynamic and pharmacokinetic changes that occur with age.

Medication use by the elderly

Medication use is correlated with the age-associated onset of chronic diseases for which drug therapy offers symptomatic relief, and assists in preventing the onset of disabling and life-threatening complications. However, high rates of medication use by older people raise a number of issues, ranging from concerns with rising expenditures for individuals and third-party insurers; increased risk of adverse drug reactions; toxic or interaction effects from concomitant use of multiple pharmaceutical agents, both physician and self-prescribed; and poor compliance with complex medication regimens by the more physically and mentally impaired. Although existing data do not support the theory of age as an independent predictor of drug-specific adverse reactions, older people have typically been excluded from clinical trials in the dynamic and changing field of pharmacotherapy. Furthermore, ingestion of a greater number of different agents clearly exposes individuals to higher risk of adverse reactions and interactions. Recent advances in information technology have facilitated multicenter clinical trials and post-marketing epidemiological surveillance studies of specific and concomitant medication use by individuals of all ages. The employment of such technology by insurers to determine appropriate prescribing or to control costs in the present limited state of knowledge is, however, premature. Furthermore, such techniques will not replace the need for careful clinical review of symptoms and total drug therapy by prescribing physicians, with modification of regimens and provision of appropriate information and instructions to older individuals and their caregivers.

Drug interactions in primary health care. A retrospective database study and its implications for the design of a computerized decision support system

OBJECTIVE

To investigate the occurrence of potential drug interactions in primary health care from the perspective of the prescribing general practitioner.

DESIGN

Retrospective database study of computer-based patient records with a query language. All drug prescriptions during a four year period were compared with concurrent or overlapping prescriptions for the same patient and these drug pairs were compared with a database of drug interactions from the Swedish drug catalogue.

SETTING

One health centre in Sweden with six general practitioners and two doctors on vocational training.

PARTICIPANTS

All patients who had visited a doctor at the health centre between 1 November 1986 and 31 October 1990.

MAIN OUTCOME MEASURES

The rate of potential interactions in relation to all drug prescriptions and the incidence rate of potential interactions for patients at risk (those receiving two or more drugs).

RESULTS

Approximately 55,000 drug prescriptions were analysed for potential drug interactions. A total of 1,074 cases of potential drug interactions were found, which corresponds to a rate of 1.9% of all drug prescriptions. The incidence rate of potential interactions was 12% for all patients at risk (those receiving two or more drugs) and 22% for elderly (> = 65 years of age) patients at risk. Major interactions were investigated concerning the extent to which the prescribing doctors were aware of the potential interactions.

CONCLUSION

Potential drug interactions occur at a high rate in general practice, in particular for elderly patients. Properly designed computer-based decision-support might increase the prescribing doctor's awareness of clinically significant interactions and improve the quality of drug treatment.

Decision support for drug prescription integrated with computer-based patient records in primary care

A conceptual model of an information system that integrates a controlled vocabulary, a patient database, and a knowledge base is described. Methods, design and components for the implementation of the system are discussed. It is argued that the key issue for the successful introduction of computer-based decision support in primary care today is integration with a computer-based patient record. Also important is that the knowledge acquisition process is based on the general practitioner's real needs. This has been achieved by, first, providing general practitioners with real patient data from a series of retrospective database studies; and second, letting a panel of general practitioners select, discuss and decide which computer reminders to implement. A hybrid representation scheme was chosen for the knowledge base. The combination of a standard procedural representation (the so-called Arden syntax) for the reminder knowledge with a semantic net representation for the medical factual knowledge facilitates knowledge sharing with other systems and knowledge reuse within the system.

Prospective evaluation of adverse drug interactions in the emergency department

STUDY OBJECTIVE

To determine the incidence and risk factors of clinically relevant adverse drug interactions occurring in emergency department patients.

DESIGN

This report describes the drug interactions identified in an emergency population. Patients' drug regimens were evaluated prospectively at the time of the emergency evaluation.

SETTING

University hospital ED.

TYPE OF PARTICIPANTS

Convenience sample of 341 patients.

INTERVENTIONS

Patients' medications on arrival at the ED (current medications) and medications initiated in the ED were entered into Hansten's computer-based drug interaction program to identify potential drug interactions. All potential drug interactions were brought to the attention of the attending emergency physician, whose subsequent actions were noted. Clinically relevant interactions were determined by a physician panel based on the ED attendings' actions, set criteria, and a review of hospital charts and four-week telephone follow-up of patients with potential drug interactions.

MEASUREMENTS AND MAIN RESULTS

Three hundred forty patients were enrolled. One hundred thirty-five potential drug interactions were identified in 61 patients. Among these 135 potential drug interactions and 61 patients, we identified 20 clinically relevant interactions in 15 patients. The incidence of clinically relevant interactions was significantly higher (chi 2 = 3.95, P = .047) among current medication (9.7%) than medication added in the ED (3.1%). Clinically relevant interaction from both current and ED-initiated medication was associated with taking three or more medications on ED arrival (P = .016 and .045, respectively). Multiple regression showed age of 60 years or older to be the sole predictor of clinically relevant interaction among current medication (P = .05).

CONCLUSION

Clinically relevant adverse drug interaction was significantly less from medication administered or prescribed in the ED than from current medication.

Adverse reactions as a cause of hospital admission in the aged

Truly elderly people comprise an increasingly large fraction of the population and consume a disproportionate amount of drugs. Over the last 25 years a number of different studies have illustrated that advancing age is associated with adverse drug reactions (ADRs). Advancing age is also associated with polypharmacy and multiple pathology, and this complex inter-relationship makes it difficult to conclude that age itself is a causative factor for ADRs. ADRs resulting in hospital admission have been the subject of study for many years, but it has not been consistently demonstrated that advancing age is a predisposing factor. Early studies used the method of intensive inpatient monitoring and identified digoxin, diuretics, aspirin, psychotropics and cytotoxics as drugs of concern. Smaller more recent studies have used more sophisticated statistical techniques to identify predisposing factors. Nonsteroidal anti-inflammatory drugs have been added to the list of drugs that may cause ADR-related hospital admission. Polypharmacy, and altered pharmacokinetics and pharmacodynamics are possible causative factors; however, variable compliance and multiple pathology may cause difficulties with attributing causality. Some basic guidelines for sensible prescribing would almost certainly result in fewer ADRs in the elderly, including those ADRs severe enough to result in hospital admission.

Iatrogenesis in the elderly. Drug-related problems

Drug reactions are generally related to the influence of age, toxicity, side effects, immunologic reactions, idiosyncratic reactions, drug-drug interactions, and drug-disease interactions. In addition to age-related changes, the elderly are susceptible to the incidence of adverse drug reactions because of polypharmacy, incorrect self-administration of drugs, omission of drugs, taking another's prescriptions, use of over-the-counter drugs, and medication errors by health-care personnel. To prevent or predict adverse drug reactions, the gerontologic nurse can obtain thorough drug histories, educate clients and health-care providers, use nursing measures to alleviate symptoms, and be astute for the potential for problems through drug review.

Drug-prescribing patterns in old age. A study of the impact of hospitalization on drug prescriptions and follow-up survey in patients 75 years and older

A prospective drug surveillance study was undertaken in 300 elderly patients admitted to a geriatric clinic. Prescribing patterns were determined on admission, at discharge and 3.6 and 18 months after discharge. Patients referred from long-term care institutions were on significantly more drugs than non-institutionalized subjects. A 34% reduction in the number of medicines prescribed at discharge was accompanied by a significant decrease in the mean number of prescriptions per patient, from 4.3 to 2.8, irrespective of whether the patient was institutionalized. Polypharmacy, defined by 5 or more concomitant drugs, declined from 43 to 17%. Dosage schedules were simplified in the majority of patients, as expressed by a significant decrease in the mean number of daily doses to be taken from 6.7 on admission to 4.4 at discharge. Cardiovascular drugs, diuretics and psychotropic drugs accounted for 64% of all drug prescriptions. At discharge, prescription frequencies were reduced for most medication categories, except diuretics and gastrointestinal drugs, which were being taken more often. The prescribing frequency of cardiac glycosides, the single most frequently prescribed drug class, decreased from 60 to 33% of the patients. Three months after discharge, prescribing patterns and frequencies were found to be very similar to the pre-admission situation. Eighteen months after discharge, overall drug use had increased by 15% compared to admission, and polypharmacy was recorded in 54% of patients. It is concluded that a substantial reduction in drug prescriptions was possible in the majority of elderly patients, particularly if they are institutionalized, on admission to a geriatric clinic.(ABSTRACT TRUNCATED AT 250 WORDS)

Detecting drug interactions: a review of the literature

This article provides a historic overview of drug interaction screening and reviews 19 studies that have sought to measure the frequency of drug interactions. Differences in study designs, methodologies, and definitions contribute to considerable variation in the reported incidence rates, which ranged from 2.2 to 70.3 percent for all potential drug interactions. The percentage of patients actually experiencing symptoms that could be attributed to a drug interaction, however, ranged from 0 to 11.1 percent. The relative importance of drug interactions as a clinical problem remains unclear. Screening programs that do more than simply identify large numbers of patients who receive potentially interacting drug combinations without indicating which subpopulations of these individuals are likely to be harmed by the drugs have not yet been developed.

Appropriateness of acute medical care for the elderly: an analysis of the literature

Over the past 30 years, an explosion in health care expenditures has occurred. Prior to 1960, health care accounted for 4.4% of the U.S. Gross National Product; today it is 11%. Before rational solutions to controlling this rise can be proposed, we must determine whether the care that we are currently paying for is appropriate to the needs of the elderly. This paper analyzes the literature regarding appropriateness of acute care provided to the elderly. We identified 17 articles that explicitly cited appropriate or inappropriate care (including under-, over- and misuse) provided in hospital and ambulatory settings and for procedures, and 19 articles that presented data on the appropriateness of medication use in the elderly. Virtually every study included in this review found at least double-digit levels of inappropriate care. Perhaps as much as one-fifth to one-quarter of acute hospital services or procedures were felt to be used for equivocal or inappropriate reasons, and two-fifths to one-half of the medications studied were overused in outpatients. The few studies that examined underuse or misuse of services also documented the existence of these phenomena. This was especially true for the ambulatory care of chronic physical and mental conditions and concerned the use of low-cost technologies (visits, preventive services, some medications). Thus, we conclude that there appears to be a substantial problem in the matching of acute services to the needs of elderly patients. This mismatch occurs both in terms of overuse and underuse, at least for areas where research has been conducted.

[Incidence and severity of drug interactions in the elderly: a prospective study of 639 patients]

The medicinal treatments of 639 patients aged over 65 were recorded on admission to hospital. The mean number of drugs consumed was 4.4 +/- 2.8 per patient. Drug interaction was found in 37 p. 100 of the patients on the basis of data published in the Vidal dictionary. The prevalence of interactions increased with the number of drugs prescribed. The medicinal families most frequently involved were digitalis derivatives, antiarrythmic agents, diuretics, anticoagulants and psychotropic drugs. Thirty patients (4.7 p. 100) presented with a side-effect that was directly ascribable to an interaction. Among the iatrogenic adverse reactions 11 were life-threatening, including 8 cases of severe dysrhythmia and 3 cases of gastrointestinal haemorrhages. Altogether, one-third of all iatrogenic disorders were consecutive to a drug interaction. Simple precautions would have considerably reduced the frequency of such side-effects.

Psychiatric admissions due to adverse drug reactions

This is a report of the findings of a 6-year study of hospitalizations caused by adverse psychiatric reactions to prescribed medications. Of 15,800 consecutive psychiatric admissions to two university hospitals, 112 (0.7%) were caused by adverse reactions to medications. In 67% of cases these admissions were due to extrapyramidal symptoms such as parkinsonism and/or akathisia, and coexisting neuroleptic-related depression. In 25% the admitting diagnosis was drug-induced delirium or psychosis; one third of these patients suffered from Parkinson's disease and had been treated with a combination of two or more antiparkinsonian agents. Older age, polydrug therapy, and the parenteral administration of neuroleptics at high dosages were important risk factors for severe adverse drug reactions leading to hospitalization.

Clinical pharmacokinetic considerations in the elderly. An update

There are numerous studies of drug handling in the elderly, but it is difficult to assess the significance of changes seen in vitro, or after single-dose administration, because they are often compensated by other mechanisms at steady-state. However, a knowledge of these studies is important as the results alert the investigator to possible treatment problems. The high incidence of adverse drug reaction in the elderly population leaves no doubt that improvements in therapy are needed. Research has been directed at seeking patterns of abnormality in the elderly on which to base recommendations for alterations in dosage regimens. The major shortcoming of this approach has been the failure to distinguish between the effect of chronological age on drug pharmacokinetics, and drug kinetics in elderly people with multiple pathology. The latter concern appreciates the variety of factors involved and the importance of treating each patient as an individual: presentation of mean data is confusing and misleading. The objective of drug treatment in any age group, but particularly in the elderly, is to administer the smallest possible dose which gives adequate therapeutic benefit throughout the entire dosage interval with the minimum of side effects. For most drugs the safe starting dose in the elderly is one-third to half that recommended in the young. Vigilance for potential side effects with plasma concentration monitoring, if available, should help keep toxicity to a minimum. When other medications are added or changed, the possibility of interaction should be anticipated. Methods for individualisation of dosage regimens and the use of sustained-release formulations in the elderly are discussed. Dosage alteration in the elderly in terms of reduced dose frequency, rather than dose size, may help improve compliance. A knowledge of the pharmacokinetics of a drug helps determine which approach will be most beneficial.

Influence of hospitalization on drug therapy in the elderly

In order to determine whether acute hospitalization leads to changes in the medications used by the elderly after discharge, we studied the medications used at admission and discharge for 197 elderly subjects admitted to one hospital. We found that 40% of all admission medications were discontinued by discharge and 45% of all discharge medications were newly started during the hospitalization. Although the number of drugs used did not increase significantly from admission to discharge (4.50 to 4.80, P = .128), the number of narcotics, laxatives, and antibiotics increased significantly. For those elderly subjects admitted to the hospital taking less than the average number of medications, the overall use of medication increased from admission to discharge (2.89 to 3.75, P less than .0001). These "low users" were discharged on more benzodiazepines, narcotics, laxatives, antibiotics, and cardiac medications. Our data suggest that during hospitalization admission medications are discontinued and new medications are started in large numbers and that these changes are accompanied by a tendency towards the increased use of certain categories of medications. These changes may place the elderly patient at increased risk of polypharmacy and its complications.

Effect of DRGs on three-month readmission rate of geriatric patients with congestive heart failure

We reviewed the three-month hospital readmission rates of 410 patients ages 70 years or older discharged alive with a first time diagnosis of congestive heart failure during the period January 1983-June 1986. The mean age was 79.8 years, 59.5 per cent were women. Mean length of initial hospital stay decreased from 10.8 days in 1983 to 7.8 days in 1986. One hundred-nineteen patients (29 per cent) were rehospitalized at least once within three months of initial discharge. The readmission rates by year were: 1983, 40.0 per cent; 1984, 27.5 per cent; 1985, 21.4 per cent; 1986, 23.2 per cent. During this same interval, the percentage of patients referred for home health care services increased from 3.3 per cent in 1983 to 13.0 per cent in 1984, 5.8 per cent in 1985, and 12.5 per cent in 1986. Thus, decreased length of hospital stay was associated with a parallel decline in early readmission rate and increased utilization of home health care services. Although this study has important methodologic limitations, the data suggest that shorter hospital stays under the DRG system are not necessarily associated with an increased rate of early rehospitalization.

Assessing the preventability of emergency hospital admissions. A method for evaluating the quality of medical care in a primary care facility

The quality of primary medical care was assessed by studying the events leading to 686 emergency admissions of patients from our hospital-based primary care practice. Independent physician reviewers determined that 59 (9 percent) of the admissions were potentially preventable; 40 were due to iatrogenic factors including inadequate follow-up and adverse drug reactions, 12 were due to lack of patient compliance, and seven were due to both iatrogenesis and noncompliance. Adverse drug reactions were the most common cause of iatrogenesis, and warfarin was the drug that most commonly caused an adverse reaction. Inadequate follow-up of abnormal physical findings, symptoms, and laboratory test results was also important. Patients with preventable admissions had more medical diagnoses (4.9 versus 4.1, p less than 0.01), were prescribed more medications (4.5 versus 3.7, p less than 0.01), and were older (66.5 years versus 60.2 years, p less than 0.01) than patients whose admissions were not preventable. It is concluded that a small percentage of emergency hospitalizations may be preventable and that systematic review of emergency hospitalizations may provide a means of measuring the quality of primary medical care.

The quality of medical care received by older persons

This paper reviews studies published since 1970 on the quality of medical services received by older persons. Although many of the studies were flawed in design or limited in scope, they suggest that there are problems in the quality of care received by older persons in ambulatory, hospital, and nursing home settings. Changes in health care delivery and financing should attempt to improve, rather than merely maintain, the quality of medical and health services received by this vulnerable population.

Determining drug-drug interactions and related effects on microcomputers

Algorithms for determining drug-drug interactions are described. They handle all the generic drugs in the British National Formulary (BNF) (Number 5, British Medical Association and Pharmaceutical Society of Great Britain, 1984) and allow these drugs to be identified by generic or trade name. They also accept and identify interactions for compound drugs prescribed by a single trade name. The algorithms can be implemented on any microcomputer and in a number of languages. The choice of algorithm and number of drugs handled depend on size and disk storage of the microcomputer. It was tested in collaboration with the Department of Geriatric Medicine, Liverpool University, and is used in a prescribing monitoring system operational in some wards of the Royal Liverpool Hospital. The drug data are stored in the form given in the BNF and can be easily updated from this publication. The algorithms can also handle drug allergies and cross-sensitivities.

Appropriate use of drugs by elderly people

The effectiveness and safety of prescribed medication theoretically depends on the patient correctly following the prescriber's instructions. However, patient compliance is often difficult to achieve. This paper discusses reasons for this in relation to elderly people, and attempts to remedy the situation are reviewed.