Utilisation Review

Utilisation Review in Health Insurance and the NHS

Utilisation review (UR) is the scrutiny of individual admissions, hospital days, procedures and prescriptions against written criteria, with coverage or payment depending on the result. Health economists treat it as a non-price way of rationing insured care, whereas cost sharing works through the price the patient pays. Each review can avoid spending on low-value care, but it adds administrative cost for payer and provider and can delay or block care that was worth giving. Whether a review programme pays is therefore a cost-effectiveness question, faced by US insurers and NHS commissioners alike.

Moral hazard and the case for reviewing decisions one by one

Insurance lowers the price of care at the point of use, so some care is used whose value is below its cost, the problem of moral hazard in health insurance. Brot-Goldberg and colleagues note that economists have mostly looked to price instruments such as higher cost sharing, and that recent evidence suggests cost sharing often screens out high-value care in low-income households. Review acts on the prescriber and the specific decision instead.

In their account, a prior authorisation form lets the prescriber pass information about the patient's suitability to the insurer, easing an information asymmetry so that denials can be aimed at low-value use. The effort of the form also works as an ordeal in the sense of Nichols and Zeckhauser, signalling that the prescriber judges the treatment worth the trouble. Patients who would have been approved anyway still bear the paperwork and any delay, while the gains come from marginal patients whose low-value care is deterred, so the size of both groups decides whether a requirement pays.

Prospective, concurrent and retrospective review

Review is classified by when it happens, because timing decides what the reviewer can still change. Prospective review, such as prior authorisation or pre-admission certification, makes coverage conditional on approval before the service. Concurrent review takes place during an episode, for example as continued-stay review that approves fewer hospital days than the doctor requested. Retrospective review comes afterwards, through claims checks and audits that deny or recover payment, and changes what providers expect next time.

US Medicare writes these stages into its hospital rules. Under 42 CFR 482.30, a hospital taking part in Medicare must, with limited exceptions, have a UR plan covering the medical necessity of admissions, the duration of stays and professional services, including drugs. Admissions may be reviewed before, at or after admission, and the committee must include at least two doctors, and no review may be done by anyone with a direct financial interest in the hospital or professional involvement in that patient's care.

How the payment system decides what is worth reviewing

The case for review depends on who bears the cost of an extra unit of care. Under fee-for-service or per-day payment the payer bears the cost of every extra day, so continued-stay review is an obvious target. Under per-case payment through a diagnosis-related group the hospital bears the cost of a longer stay itself. The Medicare rule follows this logic, since hospitals paid under the prospective payment system need review only stays and services they reasonably assume to be outliers on length of stay or cost.

Under capitation the plan keeps any money that a denial saves. The Office of Inspector General (OIG) of the US Department of Health and Human Services describes the possible incentive to deny services and payment to raise profits as a central concern about the capitated model used in Medicare Advantage. Review run by the party that keeps the money saved needs scrutiny for wrongful denial as well as for overuse.

Criteria, coverage rules and the reviewer's decision

A review applies written criteria, often built from appropriateness criteria, to the documentation sent for one patient. Clinical appropriateness and contractual coverage are separate questions, since a service can be clinically reasonable yet excluded from a benefit. The OIG found that some Medicare Advantage organisations denied requests using clinical criteria not contained in Medicare coverage rules, for example requiring an x-ray before approving more advanced imaging. Criteria stricter than the formal benefit narrow the effective benefit without any change to the policy itself.

Evidence on hospital use, spending and readmissions

The classic evaluations come from US employer insurance in the 1980s. Feldstein, Wickizer and Wheeler analysed claims for 222 insured groups in 1984 and 1985 under compulsory programmes with financial penalties for non-compliance. Review reduced admissions by 12.3%, inpatient days by 8.0%, hospital expenditure by 11.9% and total medical expenditure by 8.3%, with a savings-to-cost ratio of about 8 to 1, although health status was not studied. Groups with high admission rates before review saw a 34% reduction in patient days.

Later data from a managed fee-for-service plan showed where restriction fell. Among 49,654 adult patients reviewed between 1989 and 1993, fewer than 1% were denied care at admission, and concurrent review accounted for 83% of the reduction in requested days. Among 2,443 privately insured psychiatric patients reviewed over the same years, each day cut from the requested stay raised the adjusted odds of readmission within 60 days by 3.1%. Savings on the reviewed admission can therefore return through readmissions and lost health.

Recent studies use stronger designs. Brot-Goldberg and colleagues used the random assignment of low-income Medicare Part D beneficiaries to default plans and found that a prior authorisation restriction cut use of a drug by 26.8%, with about half of marginal beneficiaries moving to a related drug and half to none. Drug spending fell by USD 96 per beneficiary-year against about USD 10 in paperwork costs. Shi found that every dollar Medicare spent on monitoring for unnecessary spending generated USD 24 to 29 in savings, mostly by deterring future care, with no evidence of harm to the marginal patient.

Denials, appeals and the error rate

Denial and appeal figures are easy to misread, because an overturn rate describes only the self-selected minority of denials that are appealed. The OIG's clinical review of a random sample of denials by 15 large Medicare Advantage organisations in June 2019 estimated that 13% of prior authorisation denials met Medicare coverage rules, closer to a true error rate. Published appeal and overturn figures, and how many wrong denials are corrected, belong to appeals design.

Net monetary benefit of a review programme

Gross savings overstate what a programme is worth. A fuller account sets avoided spending against substitute care, review and appeal costs, and health lost through wrongful denial or delay, expressed as net monetary benefit.

$$\text{NMB} = n_D (c - s) - N a - n_A p - \lambda \, \Delta Q$$

where $n_D$ is the number of procedures not performed because of review, $c$ is the procedure cost, $s$ is the cost of substitute care for each untreated patient, $N$ is the number of requests, $a$ is the combined payer and provider cost per request, $n_A$ is the number of appeals, $p$ is the cost per appeal, $\lambda$ is the money value of one QALY, and $\Delta Q$ is the QALYs lost through denials and delays.

Worked example with one elective procedure under prior approval

All numbers are illustrative. A commissioner applies prior approval to a procedure costing GBP 3,000, receives 2,000 requests a year and denies 300.

1. Sort the denials. 240 are correct, as the procedure would bring each patient only 0.02 QALY. The other 60 are wrongful denials of patients who would gain 0.25 QALY each. Of these, 20 appeal and are treated after a delay costing 0.02 QALY each, and 40 go without.

2. Avoided spending. 240 + 40 = 280 procedures are not performed, and each untreated patient receives conservative care costing GBP 500, so the net saving is 280 × (3,000 − 500) = GBP 700,000.

3. Review costs. Handling at GBP 60 per request costs 2,000 × 60 = GBP 120,000, and appeals at GBP 200 each cost 20 × 200 = GBP 4,000.

4. Health lost. Correct denials lose 240 × 0.02 = 4.8 QALYs, delays 20 × 0.02 = 0.4 and standing wrongful denials 40 × 0.25 = 10, a total of 15.2 QALYs. At an illustrative GBP 30,000 per QALY this is 15.2 × 30,000 = GBP 456,000.

5. Combine. NMB = 700,000 − 120,000 − 4,000 − 456,000 = GBP 120,000.

6. Test the error rate. A correct denial is worth 2,500 − 0.02 × 30,000 = GBP 1,900 and a standing wrongful denial 2,500 − 0.25 × 30,000 = minus GBP 5,000. Turning one into the other lowers NMB by GBP 6,900, so 120,000 / 6,900 = 17.4, or about 17 such switches, removes the gain.

7. Interpretation. A savings-to-cost ratio would set GBP 700,000 saved against GBP 124,000 of review and appeal costs, a ratio of about 5.6 to 1, yet the net monetary benefit is GBP 120,000 and would fall to zero at about 57 standing wrongful denials in 300 instead of 40. Deterrence of requests never submitted is left out, though Shi found it the main source of audit savings.

Prior approval and criteria-based access in the NHS

Although NHS commissioners do not adjudicate insurance claims, they control access to elective procedures in similar ways. In 2019 the NHS in England launched the Evidence-Based Interventions (EBI) programme, with guidance on 17 surgical procedures, four to be stopped and 13 restricted to patients meeting specific criteria. Conefrey and colleagues analysed 306 local commissioning policies. Policies for restricted procedures relied on criteria-based access (48%) or prior approval (33%), those for stopped procedures most often named an Individual Funding Request (74%), and only 43% of restricted access policies matched the national recommendations.

Anderson and colleagues used a difference-in-difference design and concluded that EBI did not accelerate disinvestment for its procedures up to February 2020. Prior approval works as prospective utilisation review applied to decommissioning low-value procedures, and local differences in its criteria feed clinical practice variation in who is treated.

Utilisation review and neighbouring activities

Utilisation management is the wider programme, which Wickizer and Lessler describe as including prospective review, case management and physician gatekeeping. Claims adjudication applies contract and billing rules to decide payment, with a medical necessity check as one step. Clinical audit compares practice with standards to improve quality and does not decide payment for a patient. Cost containment is the broader aim, which budgets, prices or capitation can pursue without case-level review.

Limitations and common misreadings

Most errors in judging review come from counting only what the payer sees, or from reading one statistic alone. The same few misreadings recur.

1. Counting gross savings. Payer savings-to-cost ratios omit provider time, substitute care and lost health.
2. Reading approval rates. A high approval rate fits both a requirement that deters low-value requests and one that mainly adds paperwork.
3. Treating shorter stays as efficiency gains. The readmission evidence shows part of the saving can return later.
4. Generalising old evidence. The large 1980s effects were greatest in groups with high admission rates, a poor guide where such use is already low.

Sources

• Anderson M, Molloy A, Maynou L, Kyriopoulos I, McGuire A, Mossialos E. Evaluation of the NHS England evidence-based interventions programme: a difference-in-difference analysis. BMJ Quality & Safety. 2023;32(2):90-99. https://doi.org/10.1136/bmjqs-2021-014478
• Brot-Goldberg ZC, Burn S, Layton T, Vabson B. Rationing medicine through bureaucracy: authorization restrictions in Medicare. NBER Working Paper 30878; January 2023. https://www.nber.org/papers/w30878
• Code of Federal Regulations. 42 CFR 482.30, Condition of participation: Utilization review. Current text, accessed 28 October 2026. https://www.ecfr.gov/current/title-42/chapter-IV/subchapter-G/part-482/subpart-C/section-482.30
• Conefrey C, Farrar N, Coyle M, et al. Delivering a national de-adoption programme: a documentary analysis of local commissioning policy compliance with England's Evidence-based Interventions programme (EBI). BMC Health Services Research. 2025;25(1):995. https://doi.org/10.1186/s12913-025-13012-0
• Feldstein PJ, Wickizer TM, Wheeler JR. Private cost containment: the effects of utilization review programs on health care use and expenditures. New England Journal of Medicine. 1988;318(20):1310-1314. https://doi.org/10.1056/NEJM198805193182006
• Office of Inspector General, US Department of Health and Human Services. Some Medicare Advantage organization denials of prior authorization requests raise concerns about beneficiary access to medically necessary care. OEI-09-18-00260; April 2022. https://oig.hhs.gov/oei/reports/OEI-09-18-00260.pdf
• Shi M. Monitoring for waste: evidence from Medicare audits. Quarterly Journal of Economics. 2024;139(2):993-1049. https://doi.org/10.1093/qje/qjad049
• Wickizer TM, Lessler D. Do treatment restrictions imposed by utilization management increase the likelihood of readmission for psychiatric patients? Medical Care. 1998;36(6):844-850. https://doi.org/10.1097/00005650-199806000-00008
• Wickizer TM, Lessler D. Effects of utilization management on patterns of hospital care among privately insured adult patients. Medical Care. 1998;36(11):1545-1554. https://doi.org/10.1097/00005650-199811000-00004
• Wickizer TM, Lessler D. Utilization management: issues, effects, and future prospects. Annual Review of Public Health. 2002;23:233-254. https://doi.org/10.1146/annurev.publhealth.23.100901.140529