Medication Risk
Most polypharmacy on a personal injury claim does not arrive in one prescription. It accumulates one logical step at a time, as each medication's side effect is treated with the next. Understanding that sequence is the difference between managing a medication list and managing the process that built it.
What is a prescribing cascade?
The term was coined by Rochon and Gurwitz in 1997 and revisited by the same authors in the Lancet in 2017. A prescribing cascade begins when an adverse drug reaction is misinterpreted as a new medical condition. A second drug is prescribed to treat that condition. The second drug carries its own adverse effects, which may in turn be misinterpreted and treated, and the original cause is never addressed because it has been given a different name. Australian Prescriber's 2011 review put it plainly: the new medicine is prescribed "in the mistaken belief that a new medical condition requiring treatment is present".
The classic examples are textbook material. A calcium channel blocker causes ankle swelling and a diuretic is added. An antipsychotic or metoclopramide causes tremor and stiffness and levodopa is added. A thiazide raises uric acid and allopurinol is added. An ACE inhibitor causes a cough and a cough suppressant or a course of antibiotics follows. Each second drug is a reasonable response to the diagnosis as written. The error is upstream, in the diagnosis.
What does a cascade look like on a personal injury claim?
Injury claims have their own set of cascades, built from the drugs that injury attracts. A typical sequence over 12 months of a back injury claim runs like this.
| Step | Prescribed | Adverse effect | Interpreted as | Next prescription |
|---|---|---|---|---|
| 1 | Oxycodone for acute pain | Constipation, nausea | Gastrointestinal complaint | Laxatives, then a proton pump inhibitor for reflux |
| 2 | Oxycodone continued past 90 days | Sedation, low mood, poor sleep | Adjustment disorder or depression secondary to injury | Sertraline or mirtazapine |
| 3 | Mirtazapine | Daytime drowsiness, weight gain | Fatigue, deconditioning | Sleep hygiene referral; sometimes a stimulant or a switch that adds rather than replaces |
| 4 | Pregabalin added for "neuropathic" pain | Dizziness, cognitive slowing, peripheral oedema | Vestibular problem; cognitive impairment; possible cardiac cause | Prochlorperazine; neuropsychology referral; frusemide |
| 5 | Oxycodone dose escalation as pain "worsens" | Opioid-induced hyperalgesia | Disease progression | Higher dose, imaging, pain specialist |
By step five the claimant is on six or seven medications, has three new diagnoses on file, is further from work than at step one, and every additional item can be defended as a response to a documented condition. Nobody in the chain did anything unreasonable given the information in front of them. The chain itself is the problem.
What does the evidence say about the harm?
The cascades are not hypothetical. A 2024 population cohort from Ontario followed 39,347 older adults started on a calcium channel blocker and found that 4.8 per cent had a diuretic added within 90 days. That cascade group had a 21 per cent higher rate of emergency presentations and hospitalisations than patients who were not cascaded, which the authors read as harm beyond the mere addition of a second drug. A 2025 Irish study of 533,464 community-dwelling adults over 65 measured the prevalence of expert-defined cascades and found the sequences most relevant to injury claims running at 2.5 to 3.2 per cent of the exposed population: antidepressant followed by a sleep medication, benzodiazepine followed by an antipsychotic, and alpha-blocker followed by a vestibular sedative. Australian data put the broader picture in context, with adverse medicine events implicated in more than 190,000 hospital admissions a year.
In the injury population specifically, Monash researchers reported in 2026 that antidepressant, benzodiazepine and gabapentinoid dispensing rises after a back or neck injury claim begins, and earlier Monash work showed high-risk early opioid prescribing predicts persistent use. Those are the raw materials of the cascade in the table above.
Why are claims particularly prone to cascades?
Four structural features of compensation claims make cascades more likely than in ordinary primary care. Prescribing is fragmented: a GP, a pain specialist, a psychiatrist and an emergency department may each add a drug without a shared view of the list. Symptoms are expected: nobody is surprised when an injured worker reports pain, poor sleep or low mood, so the possibility that a drug is causing them is rarely raised. Diagnoses have consequences: a new diagnosis on a claim unlocks treatment funding and, for secondary conditions, entitlements, so there is a pull toward naming rather than questioning. And nobody is paid to subtract: every participant in the claim is engaged to add treatment, and deprescribing is nobody's brief unless the insurer makes it someone's.
How do you detect a cascade in a claim file?
Detection is a timeline exercise. Build a single chronology of every medication start, stop and dose change from dispensing records, RTPM data where available, and the treating practitioner's notes. Overlay every new symptom, investigation and diagnosis with its date. Then look for the signature: a new diagnosis within a few months of a medication change, where the diagnosis matches a known adverse effect of that medication, and where the medication was not reduced or stopped before the new treatment was added. Tools such as ThinkCascades and the consensus list of potentially inappropriate prescribing cascades exist for older adults; on a claim, a pharmacist applies the same logic to the drugs actually present. The five causality questions used to test any individual symptom are set out in the pillar guide on side effects versus new diagnoses.
How do you reverse one?
Reversal runs the cascade backwards. Identify the earliest drug in the chain that could plausibly explain the first misread symptom. Reduce or stop it under supervision and observe whether the downstream symptoms improve. If they do, the downstream drugs are candidates for withdrawal in turn. This is slower and less dramatic than adding a treatment, and it requires a treating practitioner willing to test a hypothesis. It also produces the only outcome that reduces both claim cost and claimant harm at the same time. The insurer's role is to make the question unavoidable: before funding the next diagnosis, ask whether the existing medications have been considered as the cause and whether a reduction has been tried.
The drug-by-drug reference for which symptoms to look for is in the companion article on side effects mistaken for new conditions, and the special case of psychological symptoms and secondary injury is covered in the article on medication-induced psychological symptoms.
Key Takeaways
- A prescribing cascade starts when a side effect is diagnosed as a new disease and treated with a second drug. Each step is locally reasonable; the sequence is the harm.
- On injury claims the typical chain runs opioid to constipation and low mood, antidepressant to sedation, gabapentinoid to dizziness and oedema, and opioid escalation to hyperalgesia read as progression.
- Population studies show cascaded patients have around 20 per cent more serious adverse events, and relevant cascades affect 2 to 5 per cent of exposed patients.
- Claims are cascade-prone because prescribing is fragmented, symptoms are expected, diagnoses unlock funding and nobody is engaged to subtract.
- Detection is a timeline exercise: medication changes against symptom onset, looking for a new diagnosis that matches a known adverse effect of a recent change.
- Reversal means reducing the earliest plausible culprit under supervision before funding treatment for the downstream diagnosis.
Frequently Asked Questions
What is the most common prescribing cascade on a workers compensation claim?
The opioid cascade. Opioids cause constipation, sedation, low mood and poor sleep, and with prolonged use, hyperalgesia. Each of those is routinely treated as a separate condition with laxatives, antidepressants, sleep medication and, for hyperalgesia, a higher opioid dose, when the underlying driver is the opioid itself.
Is a prescribing cascade always a prescribing error?
Not at the level of the individual decision. Each prescriber usually responds reasonably to the diagnosis in front of them. The error is that the diagnosis was made without considering existing medications as the cause. That is why cascades are best detected by someone reviewing the whole medication history rather than any single consultation.
Can an insurer stop a prescribing cascade?
The insurer cannot direct prescribing, but it decides what it funds. Declining to fund a new treatment line until the treating practitioner has documented that existing medications were considered and a reduction trialled is a legitimate reasonably-necessary decision, and an independent pharmacist review provides the clinical basis for it.
How long does it take to reverse a cascade?
Weeks to months. Each drug in the chain needs a supervised reduction and an observation period to see whether the downstream symptom improves. The process is slower than adding treatment but it is the only approach that reduces medication burden, claim cost and claimant harm together.
Primary sources: Rochon and Gurwitz, The prescribing cascade, BMJ 1997, and The prescribing cascade revisited, Lancet 2017; Kalisch et al, The prescribing cascade, Australian Prescriber 2011; Rochon et al, Journal of the American Geriatrics Society 2024; Annals of Family Medicine 2025 prescribing cascade prevalence study; Tefera et al, British Journal of Clinical Pharmacology 2026; Di Donato et al, early high-risk opioid prescribing in Australian workers, CNS Drugs 2025.
