Prescribing psychotropic medication for elderly patients

The inappropriate use of medications by elderly patients has become a public health concern because of its prevalence and its potential impact on patient autonomy. The use of psychotropic medication in particular has been associated with falls, confusion and morbidity in the elderly population. [1–3]

Inappropriate prescribing and use of medication is a complex phenomenon involving the patient, the physician and the pharmacist, among others. Prescriptions for psychotropic medication might be considered inappropriate if use is prolonged or pharmacological effects are duplicated, if drugs are chosen that are inappropriate for geriatric use because of potential accumulation or if drug combinations increase the risk of side effects. [4] Some of the many health promotion interventions aimed at patients have helped educate and change the attitudes and behaviours of elderly people and have resulted in the safer use of medications. [5] The few interventions aimed at physicians and pharmacists have been based on the premise that insufficient knowledge is the basis of inappropriate prescribing. [6]

The important role family physicians play in promoting the proper use of medications is highlighted by the report that they wrote 80% of benzodiazepine prescriptions in Quebec in 1990. [7] Although some studies have outlined the demographic characteristics of physicians who are high-volume prescribers, [8–10] few have examined the many variables that influence physicians when they decide whether to prescribe psychotropic medications. We interviewed physicians in an effort to understand better their perspectives and the decision-making process associated with prescribing psychotropic medications for elderly patients. Ultimately, we wanted to know whether physicians perceive a need to change some of their practises in this area.

Methods

We used a qualitative approach based on the grounded theory because it is suited for studies that are exploratory and hypothesis generating in nature and for studies that examine complex social realities. [11] Grounded theory advocates a highly inductive approach to gathering data; the conceptual framework emerges empirically from the field during the course of the study. Among the qualitative tools available, the semi-structured interview provided the most flexibility for an exploration of the perceptions, motivations and experiences of physicians.

All 12 physicians who offered medical consultation services in 22 private apartment buildings (3570 apartments) for elderly people in a suburban region of Montreal were invited to participate in the study; 9 physicians consented to be interviewed. Because the study was exploratory we had no prior knowledge of the physicians' volume or appropriateness of prescriptions.

Between February and April 1996 a semi-structured interview of 60-90 minutes' duration was conducted with each physician; each interview was tape recorded and transcribed verbatim. Physicians were asked to refer to typical patients in their practice as a basis for discussion. The interview was structured to determine physician characteristics (e.g., years of experience, practice setting, affiliations with other health care institutions), the physician's perceptions of the patient population receiving care in this setting (e.g., attitudes toward patients, pathologies seen, approaches taken with those who have anxiety or depression) and the use of psychotropic medications (e.g., most frequently prescribed drugs, perceived side effects, nature of follow-up offered). For the purposes of our study, psychotropic medications included sedatives, hypnotics, anxiolytics and antidepressants. We continued to analyse the results until saturation was obtained for most of the categories; association is the point at which additional data repeatedly confirm the interpretation already formed.

The study received ethical approval from the research committee of the local acute care hospital. [The identity of the hospital was revealed to the editor but is not disclosed here to protect the confidentiality of the physicians.]

Results

The physicians visited the facilities from once every 2 weeks to twice a week. All but one of the physicians were also practising in other institutions such as acute care hospitals, long-term care hospitals and community health centres, and all physicians had a relatively large elderly population in their practices.

The physicians were unanimous in their view of the aging process as a very negative experience. They see new patients as having difficulty dealing with change in many aspects of their lives (e.g., decreased health status and physical autonomy, death of spouse, loss of friends and relocation) and a high level of psychological suffering. Moreover, many physicians described in-house conditions that isolated elderly residents and exacerbated their anxiety and suffering. Often these retirement buildings, described as "ghetto-like," are located far from natural centres of activity.

Many of the patients had been using psychotropic medication for a long time, some for as long as 20 years. Most physicians described a very strong attachment of many of their patients, women in particular, to anxiolytic drugs. Patients previously described as fragile and vulnerable were said to become demanding and difficult when their use of psychotropic medication was addressed, and to resist any attempt to change their prescription. Many of the physicians thought patients would seek out another physician if they were not satisfied with their prescription, and they took this into consideration before prescribing.

When deciding whether to prescribe, the risk associated with the particular drug was one of the factors considered by the physicians. Most felt that it is generally more detrimental for the patients' health not to prescribe. In their view psychotropic medication helps the elderly patient remain functional and is the least problematic solution. The physicians interviewed said they are clearly concerned about preventing short-term deterioration but do not seem to attend to possible long-term consequences. Moreover, the physicians stated that they often do not see side effects and that patients often do not report them; their perception of a positive benefit-risk ratio for prescribing is thus reinforced.

Most of the physicians reported difficulty in establishing clear-cut diagnoses for elderly patients and in differentiating between anxiety and other medical conditions (e.g., angina). They felt that this uncertainty often encouraged them to consider patients' "lay expertise" regarding their own medication and to submit to their requests for a prescription.

Scepticism regarding nonpharmacological approaches for the treatment of conditions such as anxiety and insomnia was expressed by the physicians interviewed. They identified common mild alternatives (e.g., warm milk, not watching violent movies before bed) and considered them to be ineffective for elderly people with chronic problems and thought that psychotherapeutic approaches were "doomed to failure." Thus, the decision to prescribe medication was often seen as the most effective way to help the patient.

Physicians clearly differentiated between prescribing de novo and re-prescribing a psychotropic medication for a patient for whom they did not initiate the prescription. When they initiate a prescription, they said they are very careful about their diagnosis and indication; short-acting medications are often prescribed in small doses and quantities, with a limited number of renewals. If they decide to re-prescribe, most physicians perceived their role as one of a "gatekeeper" whose responsibility it is to control the type and quantity of medication used. They said they do so by using various strategies (e.g., switching to less well-known drugs or to drugs with different half-lives). Most said they base these negotiations on a strong physician-patient relationship. None of those interviewed saw a need to review their approach to prescribing for elderly patients or for additional medical education.

Although the physicians said they try to educate individual patients, they were conscious of their limitations. They recognized that the inappropriate use of psychotropic medication for elderly patients was a public health problem, but they felt that it was beyond the scope of the individual physician. One physician recognized the difficulty of intervening with psychosocial problems.

Interpretation

The physicians in our study did not question whether they should prescribe psychotropic medications; they reported having difficulty dealing with the suffering they witness without prescribing medication. These findings support the results of others. [12, 13] The physician's role as gatekeeper is legitimate. However, when there is no clear medical indication and physicians fail to convince patients to change, the physician's emphasis shifts to maintaining control over the dosage and type of drug prescribed.

There appears to be a need to improve physicians' communication skills so that they can respond more appropriately to patient's requests for medication. Because none of the physicians interviewed saw a need for continuing medical education, perhaps other strategies should be considered.

The inappropriate prescribing and use of medications is a complex matter that involves more than one professional body. Promoting links between physicians and community pharmacists is one option that might alleviate the relative isolation of physicians in these settings. Thus, a multifaceted intervention strategy should result in long-term modification of inappropriate prescribing of psychotropic medications for elderly people.

Although findings in this study cannot be generalized because of the small number of physicians interviewed, its exploratory nature provides interesting grounds from which to generate further research hypotheses and guide the development of intervention strategies.

We thank Ms. Christine Dore for conducting the interviews.

This study was supported by a research grant from the Programme regional de subvention pour des projets d'intervention, d'etudes et d'analyses en sante publique, Ministere de la Sante et des Services sociaux du Quebec. Dr. Collin has a fellowship from the Conseil Quebecois de la Recherche sociale.

Competing interests: None declared.

This article has been peer reviewed.

Correspondence to: Dr. Johanne Collin, Groupe de recherche sur les aspects sociaux de la sante et de la prevention, Universite de Montreal, CP 6128, succursale Centre-ville, Montreal QC H3C 3J7; fax 514 343-2334; johanne.collin@umontreal.ca