A nonprofit division of Sevadar Foundation Inc.

For clinicians  ·  For facilities  ·  For families  ·  For staff

Deprescribing and falls: what the evidence actually shows

The evidence base

Two things are true at once here, and collapsing them is how this subject gets misread in both directions. Sedating medications demonstrably impair gait, balance and reaction time — that is mechanism, and it is printed on their labels. But five independent evidence syntheses have tested whether a deprescribing program reduces fall rates, and none of them found it. This page holds both halves.

The short version

The drugs raise fall risk. Removing them has not been shown to lower it. Both statements are supported and neither cancels the other. The first is about exposure and rests on mechanism, labeling and large observational data. The second is about an intervention and rests on a handful of small, short trials that were mostly not designed to detect a change in falls at all. This page sets out exactly what was measured, by whom, and what they found.

Why the assumption is so intuitive

Because the individual drug associations are strong and real. Benzodiazepine users have roughly a one-third higher risk of hip fracture than non-users, pooled across 33 observational studies covering 169,660 hip-fracture cases, and the risk is concentrated in current rather than past users. Z-drugs are no safer on this measure — pooled relative risk 1.90 against 1.52 for benzodiazepines, with the highest risk in newly prescribed patients. Older adults with arthritis starting an opioid had a fracture hazard ratio of 4.9 compared with those starting an NSAID. Antipsychotic initiation in a nationwide cohort of over 1.5 million Danes aged 65 and over was associated with roughly two- to three-fold fracture incidence rate ratios in the first 30 days.

Every one of those is an association between taking a drug and falling. It is a reasonable hypothesis that removing the drug reverses it. It is a hypothesis that has been tested.

The five syntheses

1. The dedicated meta-analysis of withdrawing fall-risk-increasing drugs

Lee and colleagues pooled five randomized trials covering 1,305 participants aged 65 and over, four in the community and one in long-term care, with 6 to 12 months of follow-up. Withdrawing fall-risk-increasing drugs changed neither the rate of falls (rate ratio 0.98, 95% CI 0.63 to 1.51) nor the incidence of falls (relative risk 1.04, 95% CI 0.86 to 1.26) nor the rate of fall-related injuries (rate ratio 0.89, 95% CI 0.57 to 1.39). No trial evaluated fall-related fractures or hospitalizations at all.

Their conclusion, in their words: there is “a paucity of robust high-quality evidence to support or refute that a FRID deprescribing strategy alone is effective at preventing falls or fall-related injury in older adults,” and it “may result in little to no difference in the rate or risk of falls as a sole falls reduction strategy.”

2. The largest synthesis: 118 trials, 417,412 patients

A 2025 systematic review and meta-analysis covering 118 randomized trials and 417,412 patients aged 65 and over, in the community and in long-term care facilities, found no effect on injurious falls — standardized mean difference 0.01, 95% CI −0.12 to 0.14. The same analysis found no substantial difference in non-serious adverse drug reactions, quality of life, medical visits, emergency department admissions, hospitalizations, or all-cause mortality.

What it did find was a reduction in the number of medications prescribed — standardized mean difference −0.25, 95% CI −0.38 to −0.13, which the authors describe as approximately half a medication per patient.

3. The umbrella review of 27 systematic reviews

A 2026 umbrella review synthesized 27 systematic reviews of drug-class-targeted deprescribing. Reductions in medication use were most consistent for benzodiazepines and other psychotropics. Effects on anticholinergics and fall-risk-increasing drugs were limited. In the authors’ words, clinical outcomes including mortality, falls and cognition were “inconsistently reported and rarely improved,” most of the underlying reviews were rated critically low quality, and certainty of evidence was generally low or very low.

4. Cochrane on antihypertensive withdrawal: nobody measured it

The 2025 Cochrane review of withdrawing antihypertensive drugs in older people pooled six randomized trials and 1,073 participants aged 50 and over. Falls was a pre-specified secondary outcome. The review reports, plainly, that no study reported falls. Not a null result — an absence. Its future-research section calls for trials that measure “clinically important outcomes such as adverse drug events, falls, and quality of life.”

This matters because antihypertensives are among the most commonly deprescribed drugs in the name of falls prevention. The review also found blood pressure was higher after withdrawal, by a mean of 9.75 mmHg systolic.

5. Cochrane on falls in care facilities — the setting-specific answer

The 2025 Cochrane review of interventions to prevent falls in older people in care facilities is the largest and most directly relevant. It covers 104 randomized trials and 68,964 participants. On medication specifically:

  • Medication optimization as a single intervention may make little or no difference to the rate of falls (rate ratio 0.92, 95% CI 0.75 to 1.13; 13 trials, 4,314 participants; low-certainty evidence) and probably makes little or no difference to the risk of falling (relative risk 0.96, 95% CI 0.89 to 1.03; 12 trials, 6,209 participants; moderate-certainty evidence).
  • Medication review and deprescribing specifically: “We are uncertain of the impact of medication review/deprescribing on falls outcomes” — rate ratio 0.94, 95% CI 0.76 to 1.18, 12 trials, 4,125 participants, very low-certainty evidence.

What does reduce falls in a care facility

The same Cochrane review answers that question, and the answer is not nothing.

  • Exercise. As a single intervention, active exercise probably reduces the rate of falls (rate ratio 0.68, 95% CI 0.51 to 0.91; 14 trials, 2,215 participants; moderate certainty) and the risk of falling (relative risk 0.86, 95% CI 0.75 to 1.00; 13 trials, 2,408 participants; moderate certainty). In residents with cognitive impairment, it may reduce the risk of falling (relative risk 0.72, 95% CI 0.57 to 0.91; 4 trials, 451 participants; low certainty).
  • But only while it continues. After the intervention period, if exercise is not sustained there is no effect on the rate of falls (rate ratio 1.02, 95% CI 0.78 to 1.32; 7 trials, 1,354 participants) — and that is rated high-certainty evidence. Falls prevention through exercise is a standing program, not a course.
  • Tailored multifactorial programs with staff engagement. Delivered according to individual residents’ circumstances and with facility staff engaged, these probably produce a large reduction in the rate of falls (rate ratio 0.61, 95% CI 0.54 to 0.69; 7 trials, 3,553 participants; moderate certainty). All of those programs included medication optimization as one component alongside exercise and environmental assessment.
  • Vitamin D, in populations with low vitamin D levels, probably reduces the rate of falls (rate ratio 0.63, 95% CI 0.46 to 0.86; 5 trials, 4,603 participants; moderate certainty) while probably making little difference to the risk of falling.

The tension in the guidelines, stated honestly

The 2022 World Guidelines for falls prevention contain a narrative sentence asserting that medication review and deprescribing of fall-risk-increasing drugs “can significantly reduce fall risk.” That sentence is stronger than anything the guideline actually recommends.

Its graded recommendations all place deprescribing inside a program. It recommends assessing falls history before prescribing a fall-risk-increasing drug (grade 1B), using a validated structured tool to identify such drugs during a medication review (1C), that medication review and appropriate deprescribing “should be part of multidomain falls prevention interventions” (1B), and that in long-term care the falls strategy should always include rational deprescribing (1C). None of them endorses deprescribing on its own as a falls intervention. Exercise, in the same table, is graded 1A.

So the defensible statement is this one, and it is the statement this site uses: medication review belongs in a multidomain falls program, and deprescribing on its own has not been shown to reduce falls in randomized trials. Both halves are true and neither cancels the other.

Where the “two-thirds reduction” number comes from

The 2012 Cochrane review of falls prevention in the community reports that gradual withdrawal of psychotropic medication reduced the rate of falls — rate ratio 0.34, 95% CI 0.16 to 0.73 — but not the risk of falling. That figure is quoted constantly. It comes from one trial with 93 participants, and it is the same trial that Lee and colleagues identify as the only positive study among their five, and the only one that used blinded study capsules. It does not survive pooling with the other four.

Why this matters more than a point of academic accuracy

Because a facility or a family who expects a fall rate to drop because a medication list got shorter will conclude, when it does not, that the whole exercise was pointless — and will stop doing something that was worth doing for its own reasons. And because a promise that cannot be kept is a bad foundation for a conversation with a person who is being asked to change a medication they rely on.

There are good reasons to review and reduce inappropriate prescribing: it is safe, it reduces pill burden, it removes drugs that are no longer treating anything, and it sometimes uncovers sedation that was being read as decline. Falls prevention is not on that list, and saying so plainly is what makes the rest of the list credible.

Common questions

So should medications be reviewed at all if it does not prevent falls?

Yes. The randomized evidence supports medication review as safe and effective at what it directly does — reducing the number of medications and the number of potentially inappropriate ones. It also belongs inside multidomain falls programs, which do work. What it does not do on its own is reduce falls. The deprescribing step.

What actually reduces falls in a nursing home?

On the current Cochrane evidence: sustained exercise, and tailored multifactorial programs delivered with facility staff engagement — which include medication optimization as one component. Vitamin D helps where levels are low. Exercise stops working when it stops. What adopting the framework involves.

Do some medications still increase fall risk?

Yes, and the associations are strong — benzodiazepines, Z-drugs, opioids and antipsychotics all show elevated fracture risk in observational data, concentrated in the period just after starting. That is an argument about how carefully these drugs are started and monitored, which is a different argument from whether stopping them later reverses the risk. The drug-class library.

Back to the evidence base.

The mechanism: what these drugs do to standing and walking

Before any argument about interventions, the physical effect. Sedating medications impair the systems that keep an older person upright — alertness, reaction time, postural control and gait. This is not an inference from fall statistics; it is on the labels.

What the FDA-approved labeling says

The label for a commonly prescribed benzodiazepine hypnotic states plainly, under Warnings: “Because Restoril can cause drowsiness and a decreased level of consciousness, patients, particularly the elderly, are at higher risk of falls.” Its Geriatric Use section is more specific still: “the risk of the development of oversedation, dizziness, confusion, ataxia and/or falls increases substantially with larger doses of benzodiazepines in elderly and debilitated patients.”

Note what that is. A fall warning naming the elderly, with a named mechanism — oversedation, dizziness, confusion, ataxia — in FDA-approved labeling. That is a regulatory finding, not an association pulled from a database.

The same pattern runs through the other sedating classes. Gabapentin’s labeling reports somnolence, dizziness and ataxia at greater rates than placebo and warns that the drug “may cause significant driving impairment.” Z-drug labeling carries a boxed warning for complex sleep behaviors occurring while not fully awake, and the FDA notes patients may not remember them. Opioid labeling warns of sedation and psychomotor impairment.

Why the criteria sets name these drugs

The AGS Beers Criteria advises avoiding benzodiazepines and Z-drugs in older adults specifically because of falls and fractures, alongside cognitive impairment and delirium. That is the stated rationale, not a side note — the criteria exist because these associations were considered strong enough to warrant a default of avoidance.

And the observational signal is large and consistent

Covered in full below, but in summary: hip fracture risk roughly one-third higher in benzodiazepine users and nearly doubled in Z-drug users; fall risk roughly 3.8 times higher in the 24 hours after a benzodiazepine is started in nursing home residents; fracture hazard nearly five-fold in the two weeks after an opioid is started in older adults with arthritis; fracture incidence rate ratios of about 2 to 3 in the first 30 days after an antipsychotic is started.

Reading the null correctly

So why is the trial evidence null? Because the trials were not built to answer this question. Deprescribing trials are typically small, run for six to twelve months, and use a medication count as the primary endpoint — falls, where measured at all, are usually a secondary outcome in a study underpowered to detect a change in one. Falls are also multifactorial: vision, footwear, flooring, blood pressure, muscle strength, cognition and continence all contribute, so removing one contributor from a frail person rarely moves a fall rate on its own.

Absence of evidence from underpowered trials is not evidence that the drugs do not cause falls. The exposure evidence is why the drug should not be there without a current reason. The weak trial evidence is why nobody should promise that removing it will prevent a fall.

Why this matters: what a hip fracture does to a nursing home resident

Most falls do not cause a fracture. But when one does, in this population, it is frequently catastrophic — and the figures usually quoted come from community-dwelling older adults, who are not the people this site is about.

In nursing home residents

A retrospective cohort of 60,111 Medicare beneficiaries living in US nursing homes who were hospitalized with a hip fracture between July 2005 and June 2009 found that 36.2% were dead within 180 days. Among those who had not been totally dependent in locomotion beforehand, 53.5% had died or become newly totally dependent in locomotion within 180 days. Function declined across all seven activities of daily living assessed.

That is the figure to hold: at six months, for a resident who could still move themselves before the fracture, it is close to a coin flip whether they are dead or can no longer move themselves at all.

And cognition compounds it. In the same cohort, residents with very severe cognitive impairment had a relative risk of 1.66 (95% CI 1.56 to 1.77) for death or new total dependence compared with those whose cognition was intact — which is the point where this page joins the rest of the site. Cognitive decline and care burden.

For contrast, in a general hip fracture population

A review of 833 hip fracture patients at an Irish urban trauma center found 20.5% had died within one year. A prospective cohort of 480 hip fracture patients in the Netherlands found only 29% had returned to their pre-fracture level of instrumental activities of daily living at twelve months — roughly seven in ten never got back to where they started.

The comparison is the argument. Roughly a fifth of a general hip fracture population is dead at one year; more than a third of nursing home residents are dead at six months. The same fracture is a far worse event in a facility resident, and that is before counting the ones who survive without their mobility.

The limit of the argument

What this does and does not license. It means a sedating medication with no current indication sits upstream of the most consequential event that can happen to a resident. It does not mean that deprescribing prevents hip fractures — no trial has shown that. It removes an upstream risk factor, which is a smaller and more defensible claim, and a sufficient one.

References

  1. Lee J, Negm A, Peters R, Wong EKC, Holbrook A. Deprescribing fall-risk increasing drugs (FRIDs) for the prevention of falls and fall-related complications: a systematic review and meta-analysis. BMJ Open. 2021;11(2):e035978. PMID 33568364 · DOI 10.1136/bmjopen-2019-035978.
  2. Persaud N, Workentin A, Rizvi A, et al. Interventions to Address Potentially Inappropriate Prescribing for Older Primary Care Patients: A Systematic Review and Meta-Analysis. JAMA Network Open. 2025;8(6):e2517965. PMID 40577011 · DOI 10.1001/jamanetworkopen.2025.17965.
  3. García Álvarez Á, Riera Serra P, Ricci-Cabello I, et al. Effectiveness and safety of focused deprescribing interventions targeting specific drug classes in older adults across clinical settings: an umbrella systematic literature review. Age and Ageing. 2026;55(8):afag243. PMID 42607278 · DOI 10.1093/ageing/afag243.
  4. Gnjidic D, Langford AV, Jordan V, et al. Withdrawal of antihypertensive drugs in older people. Cochrane Database of Systematic Reviews. 2025;3(3):CD012572. PMID 40162571 · DOI 10.1002/14651858.CD012572.pub3.
  5. Dyer SM, Kwok WS, Suen J, et al. Interventions for preventing falls in older people in care facilities. Cochrane Database of Systematic Reviews. 2025;8(8):CD016064. PMID 40832852 · DOI 10.1002/14651858.CD016064.
  6. Montero-Odasso M, van der Velde N, Martin FC, et al. World guidelines for falls prevention and management for older adults: a global initiative. Age and Ageing. 2022;51(9):afac205. PMID 36178003 · DOI 10.1093/ageing/afac205.
  7. Gillespie LD, Robertson MC, Gillespie WJ, et al. Interventions for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews. 2012;2012(9):CD007146. PMID 22972103 · DOI 10.1002/14651858.CD007146.pub3.
  8. Hopewell S, Adedire O, Copsey BJ, et al. Multifactorial and multiple component interventions for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews. 2018;7(7):CD012221. PMID 30035305 · DOI 10.1002/14651858.CD012221.pub2.
  9. Guirguis-Blake JM, Perdue LA, Coppola EL, Bean SI. Interventions to Prevent Falls in Older Adults: Updated Evidence Report and Systematic Review for the US Preventive Services Task Force. JAMA. 2024;332(1):58–69. PMID 38833257 · DOI 10.1001/jama.2024.4166.
  10. Poly TN, Islam MM, Yang HC, Li YJ. Association between benzodiazepines use and risk of hip fracture in the elderly people: A meta-analysis of observational studies. Joint Bone Spine. 2020;87(3):241–249. PMID 31778821 · DOI 10.1016/j.jbspin.2019.11.003.
  11. Donnelly K, Bracchi R, Hewitt J, Routledge PA, Carter B. Benzodiazepines, Z-drugs and the risk of hip fracture: A systematic review and meta-analysis. PLoS One. 2017;12(4):e0174730. PMID 28448593 · DOI 10.1371/journal.pone.0174730.
  12. Miller M, Stürmer T, Azrael D, Levin R, Solomon DH. Opioid analgesics and the risk of fractures in older adults with arthritis. Journal of the American Geriatrics Society. 2011;59(3):430–438. PMID 21391934 · DOI 10.1111/j.1532-5415.2011.03318.x.
  13. Torstensson M, Leth-Møller K, Andersson C, Torp-Pedersen C, Gislason GH, Holm EA. Danish register-based study on the association between specific antipsychotic drugs and fractures in elderly individuals. Age and Ageing. 2017;46(2):258–264. PMID 27932365 · DOI 10.1093/ageing/afw209.
  14. SpecGx LLC. RESTORIL (temazepam) capsules — FDA-approved prescribing information (Warnings; Precautions; Geriatric Use). DailyMed, U.S. National Library of Medicine. 2026. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=bc56f7fd-1aaf-48ff-8aa7-d467e59e1015.
  15. Parke-Davis, Division of Pfizer Inc. NEURONTIN (gabapentin) capsules, tablets and oral solution — FDA-approved prescribing information. DailyMed, U.S. National Library of Medicine. 2026. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=ee9ad9ed-6d9f-4ee1-9d7f-cfad438df388.
  16. U.S. Food and Drug Administration. Certain Prescription Insomnia Medicines: New Boxed Warning — Due to Risk of Serious Injuries Caused by Sleepwalking, Sleep Driving and Engaging in Other Activities While Not Fully Awake. Drug Safety Communication, April 30, 2019. fda.gov. 2019. https://www.fda.gov/safety/medical-product-safety-information/certain-prescription-insomnia-medicines-new-boxed-warning-due-risk-serious-injuries-caused.
  17. By the 2023 American Geriatrics Society Beers Criteria Update Expert Panel. American Geriatrics Society 2023 updated AGS Beers Criteria for potentially inappropriate medication use in older adults. Journal of the American Geriatrics Society. 2023;71(7):2052–2081. PMID 37139824 · DOI 10.1111/jgs.18372.
  18. Neuman MD, Silber JH, Magaziner JS, Passarella MA, Mehta S, Werner RM. Survival and functional outcomes after hip fracture among nursing home residents. JAMA Internal Medicine. 2014;174(8):1273–1280. PMID 25055155 · DOI 10.1001/jamainternmed.2014.2362.
  19. Ferris H, Merron G, Coughlan T. 1 year mortality after hip fracture in an Irish urban trauma centre. BMC Musculoskeletal Disorders. 2023;24(1):487. PMID 37312089 · DOI 10.1186/s12891-023-06605-5.
  20. Moerman S, Mathijssen NMC, Tuinebreijer WE, Nelissen RGHH, Vochteloo AJH. Less than one-third of hip fracture patients return to their prefracture level of instrumental activities of daily living in a prospective cohort study of 480 patients. Geriatrics & Gerontology International. 2018;18(8):1244–1248. PMID 30004174 · DOI 10.1111/ggi.13471.

Bibliographic records on this page were retrieved from PubMed. Every reference links to its DOI or PubMed record.

Written and medically reviewed by Dr. Gurpreet Singh Padda, MD, MBA, MHP, founder and medical director of DWARAA. Last reviewed . This page is educational. DWARAA does not prescribe, deprescribe, diagnose or treat.