Benzodiazepines: Challenge and Opportunity

BENZODIAZEPINES: THE CHALLENGE AND THE OPPORTUNITY
Mark Rapoport, MD, FRCPC Professor of Psychiatry at the University of Toronto, Staff Psychiatrist at Sunnybrook Health Sciences Centre, Canada

Key highlights:

• Nationally-representative US data indicates that sedative-hypnotic prescriptions for older adults doubled between 2003 and 2015.
• Sedative-hypnotics were self-reported to be used by 8.3% of the older adult population, and were prescribed in 25% of ambulatory care visits for those with insomnia.
• An educational brochure directed at patients using sedative-hypnotics and at their physicians in Quebec, Canada led almost half of the sample to stop their medications, and three-quarters of those who initiated a taper were successful at six months.

Although benzodiazepines continue to be commonly prescribed among older adults, a practical easily-scalable intervention can lead to successful discontinuation of these often dangerous medications at a community-wide level.

Maust et al., (American Journal of Geriatric Psychiatry 2019) conducted the first large-scale American estimate of prescription medication and over-the-counter agents specifically used for sleep in older adults since 1979.  The authors found that 35.4% of just over 1000 community-dwelling older adults between the ages of 65 and 80 endorsed using a sleep product.  The most common type was an over-the-counter sleep aid such as "Tylenol PM" or "Benadryl," followed by herbal or natural sleep aids.  Prescription sleep medication was endorsed by 8.3% of the sample (benzodiazepines and z-drugs), and prescription pain medication by 5.0%, specifically for sleep.  Of the full sample more than two-thirds endorsed any sleep difficulty and less than half had discussed ways to improve sleep with their physician.  Poor perceived physical health was a significant predictor of prescription pain or sleep medication used specifically for sleep, but other demographic factors and perceived mental health were not significant predictors.

Agarwal and Landon (JAMA Open 2019) published another study of nationally representative American data.  This study used pharmacy data from ambulatory visits and did not rely on self-report as the study by Maust et al. did, but no data was provided about sleep complaints or non-prescription sleep products.  Agarwal and Landon used a serial cross sectional approach to compare benzodiazepine prescription prevalence in 27.6 million ambulatory care visits in 2003 and 62.6 million visits in 2015.  The authors found that benzodiazepine visit rates doubled from approximately 3.8% to 7.4% of visits.  The doubling of rates was also true among adults age 65 and over, with the unadjusted estimated benzodiazepine visit rate of approximately 3.1% in 2003 and approximately 6.4% in 2015.  Among ambulatory visits for insomnia, the benzodiazepine visit rate was about one-quarter in 2015, with no significant change from 2003, although no such data were provided for the older adults sub-sample.

These are important studies for those of us working with older adults because in both cases, the authors used nationally representative samples, and examined the use of these agents, specifically for sleep rather than for other indications such as anxiety or low mood. In the Maust et al. study, it is difficult to rely on the estimated prevalence of sleep related symptoms, as respondents were asked about how many nights they had sleeping problems, but 67.7% of those who reported sleep difficulty included those who had such a problem only one night per week.  The authors did not explore the daytime impact of sleep difficulties, which may have further led to an over-estimation of sleep difficulties. Similarly, the rates of use of sleep products include those who reported only occasional use, and frequency and duration of use were not reported in this manuscript, which relied on self-report of use of such products.

In the Agarwal and Landon study, objective prescription data and a larger sample led to more precise estimates, but their delegation of indication for the prescription relied on electronic medical record data of variable quality.  Nonetheless, we can conclude from these studies that among nationally representative American samples of older adults, the majority have sleep complaints at least once a week of varying severity, 8.3% self-report receiving a prescription for sleeping medication specifically for sleep, benzodiazepines prescriptions are doubling for older adults, and about one quarter of ambulatory visits for insomnia (all ages) lead to a benzodiazepine prescription. Other research has indicated that benzodiazepine use is often chronic, despite known risks, minimal benefits in the short-term for insomnia, and stopping these medications can be very challenging.

Martin et al. (2018), randomized community pharmacies in Quebec, Canada to a pharmacist-led educational intervention targeted at patients and their physicians, or to treatment as usual.  The primary outcome was discontinuation of prescriptions for inappropriate medications at six months following the intervention.  The medications targeted were sedative-hypnotics, first generation antihistamines, glyburide, and nonsteroidal anti-inflammatory drugs, for patients age 65 years and older who had filled a prescription for one of these medications for at least three consecutive months.  Four hundred eighty-nine patients were enrolled from 69 pharmacies, of whom 89% completed the trial.  Patients on antipsychotics, cholinesterase inhibitors, or with a Mini-mental Status Examination (MMSE) less than 24 were ineligible for participation.  The intervention included educational material for patients comprising a drug-specific brochure with information about why the medication may be inappropriate, potential alternatives for treatment, and for benzodiazepines, a visual tapering protocol.  There was also an educational material package for physicians in the form of an evidence-based pharmaceutical opinion including the rationale for why de-prescribing was recommended, evidence about drug harms, credible sources of recommendations, potential safer therapeutic alternatives, and personalized participant data.  A robust effect was found, with 43% of those in the intervention group no longer filling a prescription for one of these potentially inappropriate medications compared with 12% in the control group (a number needed to treat of 3.2).  For sedative hypnotics, 43.2% of patients had stopped by six months compared with 9.0% in the control group (number needed to treat of 2.9).  Although 38% of sedative-hypnotic users reported withdrawal symptoms during the taper, 75% of users who initiated tapering successfully completed the protocol.

The magnitude of this effect was greater than in an earlier study by the same group using only the educational brochure, and not material sent to the prescribing physician, with a number needed to treat of five, and a 54% success rate in tapering sedative-hypnotics.  Although the main limitation of this study is the relatively short six month follow-up, the principal finding is that a low tech educational brochure to patients and their physicians has a large effect in reducing these prescriptions, off-setting at least to some degree the rising prescription prevalence of this potentially harmful medication. 

For further reading:

• Maust, D.T., Solway, E.S., Clark, S.J., Kirch, M., Singer, D.C., Malani, P. Prescription and nonprescription sleep product use among older adults in the United States. American Journal of Geriatric Psychiatry 2019, 27 (1): 32–41. PMID: 30409547
• Agarwal, S.D., Landon, B.E. Patterns in outpatient benzodiazepine prescribing in the United States. JAMA Open 2019, 2(1):e187399. PMID: 30681713.
• Martin, P., Tamblyn, R., Benedetti, A., Ahmed, S., Tannenbaum, C. Effect of a pharmacist-led educational intervention on inappropriate medication prescriptions in older adults. The D-PRESCRIBE randomized clinical trial. JAMA 2018, 320(18): 1889-1898.

Cited in these papers:

• Tannenbaum, C., Martin, P., Tamblyn, R., Benedetti, A., Ahmed, S.  Reduction of inappropriate benzodiazepine prescriptions among older adults through direct patient education: The EMPOWER cluster randomized trial.  JAMA internal medicine 2014, 174 (6), 890-898. PMID: 24733354

Dr. Mark Rapoport is an associate professor in the geriatric psychiatry division of the department of psychiatry at the University of Toronto, a clinical scientist at Sunnybrook Health Sciences Centre, and past President of the Canadian Academy of Geriatric Psychiatry. His main areas of research are traumatic brain injury in the elderly and the risk of motor vehicle collisions associated with neurological and psychiatric diseases and their treatments. After serving as an Assistant Editor, Dr. Rapoport became the Deputy Editor for the Research and Practice section of the IPA Bulletin in 2016. He regularly writes on recent advances in the field.

Excerpted as reprint from the IPA Bulletin, Volume 36, Number 1
IPA Members can download the full PDF issue here.