Potential problematic prescribing in older women

Prescription of estrogen therapy in older women could be a red flag for use of other medications linked to dementia, a recent study found.

Three researchers from the University of California, San Francisco (UCSF), Mary M. Hunter, PhD, Nadra E. Lisha, PhD, and Alison Huang, MD, MPhil, MAS, used 2006-2016 data from the National Ambulatory Medical Care Survey to evaluate the association between estrogen therapy and coprescription of other potentially inappropriate medications in women ages 65 years and older. The first category included estrogen alone and estrogen plus progestin but excluded prescriptions labeled as vaginal estrogen, while the second included first-generation antihistamines, antispasmodics, benzodiazepines, nonbenzodiazepine receptor agonists, opioids, and skeletal muscle relaxants.

The study found a link between prescription of estrogen therapy and the use of at least one sedating potentially inappropriate medication (adjusted odds ratio, 1.6) as well as two or more types of sedating medications (adjusted odds ratio, 2.1). In addition, polypharmacy, defined as five or more medications, was seen in 39.8% of visits not involving estrogen therapy, 66.5% of visits involving estrogen plus progestin, and 76.9% involving estrogen alone. Estrogen therapy was also strongly associated with superpolypharmacy, defined as 10 or more medications.

The results were published Feb. 26 as a research letter by the Journal of the American Geriatrics Society. Dr. Huang, professor of medicine, urology, and epidemiology and biostatistics, director of research for general internal medicine, and director of the UCSF Women's Health Clinical Research Center, recently spoke to I.M. Matters about the findings.

Q: Why did you do this particular study?

A: Estrogen therapy is most risky or problematic in older menopausal women, as opposed to younger menopausal women. We have reason to think from randomized trials that the risks of estrogen therapy are greatest in older age groups for a variety of reasons, among them increased rates of stroke and dementia with systemic estrogen therapy in older women.

At the same time, there are questions about when older women do use estrogen therapy, what are the factors that go into those decisions? Does this have to do with differences in the kind of somatic and psychological symptoms that women have if they need estrogen therapy or want estrogen therapy at older ages? Is this related to differences in physician prescribing patterns?

Estrogen therapy in older women is described in the American Geriatrics Society Beers Criteria as a potentially inappropriate medication in older women. We were just interested in whether estrogen therapy prescribing clustered with prescribing of other types of medications that are considered higher risk in older women, particularly the more sedating medications, which have also been linked with dementia in older adults.

Q: Were the study results what you expected?

A: We did find overlap between estrogen therapy prescribing and prescribing of sedative medications, strong antihistamine medications, opioid medications—a variety of classes of sedating medications considered potentially inappropriate for older adults. I think we suspected this might be the case, but we had never seen it demonstrated or confirmed anywhere else, so these were new findings to us and, I think, to the literature in general.

What's tricky is that from this study, we can only get sort of a broad overview of these prescribing patterns, and we can really only speculate as to the reasons why we see the overlap. It could be that older women are having more symptoms or health issues that are resulting in them needing or wanting estrogen therapy and also needing or wanting more sedating medications prescribed to them in older age. It could be that the clinicians who are prescribing both types of medications are clinicians who are more comfortable prescribing medications even if they're higher risk in older age groups. It could be other sort of shared differences in health status or health conditions. From these data, we can't tell what's causing it.

It's possible that all of these things are true.

I think it's also possible that these visits are singling out patients who just have greater engagement with the health care system at large. You could say people who are prescribed one type of medication are just in general more likely to be prescribed other kinds of medications. It's a marker of interacting with the health care system and with clinicians, and more opportunities for medications of all kinds to be prescribed, potentially both good and bad.

Q: What else was noteworthy about your findings?

A: We did see that prescribing estrogen therapy did go along with polypharmacy. It wasn't just that the women were prescribed more sedating medications; they were prescribed many more medications in general. Any of those medications could be medications that are important for their health, that could be preserving their functioning and quality of life or helping to prevent death. But they also could include medications that are potentially problematic or that increase the risk of drug-drug interactions or side effects.

For there to be an association with not just any sedating medications but also simultaneously taking five or more medications, simultaneously taking 10 or more medications, that does raise some concern. Are women in older age who are being prescribed estrogen therapy also older women who are more at risk from polypharmacy in general?

Q: The Beers Criteria flag medications that are potentially inappropriate in older patients, but they can also be prescribed appropriately in this population, correct?

A: That's right. The Beers Criteria are really part of an effort to raise awareness of medications that have greater risk in older adults—not that they should never be prescribed, but that clinicians should think carefully before prescribing them. There's also concern that when older patients are simultaneously being prescribed multiple potentially inappropriate medications that have similar risks, that this is where we may see the greatest adverse effects. It's one thing to be prescribed one medication that could potentially increase your dementia risk, but if you're being simultaneously prescribed two or three or four or five medications, then the combined results could be serious for the patient.

Q: Are there some older women for whom estrogen therapy is indicated?

A: I definitely think that there are some women for whom the benefit-to-risk tradeoff would make sense, but probably it's a smaller number, and I think it's awfully important for older women and clinicians to properly think about and face the risks.

One potential area of concern is that there are some folks who are really hoping or believing that in younger menopausal women, estrogen therapy could somehow be protective of cognition. We don't have good randomized trial data to show that, but there are people who hope or suspect, "Oh, if estrogen is taken right around the time of menopause, could it mean that a woman is less likely to develop dementia later on?"

However, we do have quite good randomized trial data saying that when systemic estrogen is administered in the older menopausal years, dementia rates go up. There is a kind of challenge where women whose benefit-to-risk ratio may be more favorable for taking estrogen in early menopause have to really rethink what their risks are in older age.

Q: What are some take-home messages from your research for internal medicine physicians caring for and prescribing for older women?

A: I do think first that we should recognize that systemic estrogen therapy is a potentially inappropriate medication in women aged 65 and older. I do think that there are some women out there who may benefit from estrogen ... but at least older women should be made aware of the risks, including the randomized trial data demonstrating greater risk of stroke and dementia with systemic estrogen therapy if taken in older age.

Also, it's important for clinicians and women to think about the potential synergistic effects of taking multiple simultaneous medications that could increase dementia risk. Maybe if it's valuable for women to take one medication that increases risk, they can think about whether they can stop or get by without another medication that also increases that risk. Really, we want to think about what the cumulative effect is on women's cognitive function and their risk of disability, including cognitive disability, in older age.