Older adults with dementia: tramadol fell after the opioid guideline, gabapentinoids crept up

Design

Interrupted time series of a nationally representative longitudinal cohort

Population

3,246 US community-dwelling adults aged 65 or over with dementia

Primary outcome

Quarterly opioid and gabapentinoid prescribing before and after the 2016 CDC guideline

Effect

Tramadol −27.1% (guideline-associated); gabapentinoids 9.5% to 11.5% (no guideline-linked trend change)

This interrupted time series used a nationally representative US survey linked to Medicare claims to follow 3,246 community-dwelling adults aged 65 or over with dementia, median age about 81, from 2010 to 2022. It asked whether the 2016 CDC guideline advising against opioids for most chronic pain changed what they were prescribed.

Prescribing of conventional opioids fell by 26.5% over the period, but the decline was not linked to the guideline. Tramadol fell by 27.1%, and that fall was associated with the guideline. Gabapentinoid prescribing rose from 9.5% to 11.5%, with no guideline-linked change in trend. Buprenorphine use stayed near zero.

The picture is one of substitution rather than simple reduction. Gabapentin and pregabalin carry their own risks in people with dementia — sedation, dizziness, falls and confusion — and their benefit outside neuropathic pain is limited. The study cannot say whether pain control got better or worse.

• Review gabapentinoid prescriptions in older adults with dementia, especially those not for neuropathic pain.
• Ask carers about drowsiness, confusion and falls after any analgesic change.
• Use regular paracetamol and non-drug measures as the base of pain management.
• Look for behavioural signs of untreated pain in people who cannot describe it.
• Consider low-dose buprenorphine patches as an option some guidelines support in older adults.

Why it matters

Reducing opioids in frail patients may simply move the risk to gabapentinoids rather than remove it.

Don't overread it

This tracks prescriptions, not pain or harm, so it cannot show whether patients were worse off.

The statistics, in plain English

An interrupted time series compares the trend before and after an event. A change in trend whose confidence interval crosses zero, as for conventional opioids here, means the decline was already happening and the guideline did not measurably alter it.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.