Paying for Quality: What Pay for Performance in Primary Care Actually Changed

Pay for performance ties part of a practice's income to hitting measured quality targets, such as recording a blood pressure or offering a vaccination. Large national schemes improved documentation and some intermediate outcomes and narrowed a few inequalities, but the gains were generally modest and uneven, and did not clearly translate into large reductions in death. When incentives were later removed from certain measures, performance mostly held steady, which is itself a clue about what the money was and was not doing.

What pay for performance is

Pay for performance is a financing idea imported into clinical care. Rather than paying only for activity, a payer ties a portion of income to whether measurable markers of quality are met, such as the share of patients with a recorded blood pressure or an offered immunization. The premise is straightforward: what gets measured and paid for gets done.

The appeal is obvious, and so is the risk. Any target creates an incentive to hit the target, which is helpful when the measure captures real quality and troubling when the measure is a thin proxy for it. The evidence is where that tension gets tested.

What the largest schemes measured

The most studied example is a national primary-care scheme that attached a meaningful slice of practice income to a long menu of clinical indicators across common conditions like diabetes, heart disease, asthma, and mental illness. Indicators covered processes, such as taking a measurement, and intermediate outcomes, such as bringing a number into a target range.

Because the scheme was large, near-universal, and generated detailed records, it became a natural laboratory for asking what financial incentives do to the quality of everyday care. A systematic review pulling that literature together offers a balanced picture rather than a simple verdict.

What improved

The review found genuine gains. Recording of clinical information improved, some intermediate outcomes moved in the right direction, and there were signs that gaps between more and less advantaged groups narrowed for certain measures, alongside reductions in some mortality rates linked to incentivized care.

Those are not trivial achievements. Better documentation supports safer follow-up, and narrowing inequality is a goal that unincentivized systems often struggle to reach. On its own terms, the scheme moved several needles it was designed to move.

The limits and unintended effects

The same review is candid about the costs. Attention appeared to drift toward the activities that were counted and paid for, with less focus on aspects of care that sat outside the scheme. Clinicians and observers described a pull toward tick-box exercises and pressure to meet targets that could feel disconnected from an individual patient's needs.

There is also the deeper problem of the proxy. Recording a measurement is not the same as improving a life, and a scheme can post strong numbers on its indicators while the outcomes that ultimately matter move only a little. None of this means incentives failed, but it means their headline figures need reading with care.

What happened when incentives were withdrawn

A revealing natural experiment came when payments were removed from a set of indicators. If the money was continuously driving the behavior, performance should have fallen once it stopped. Instead, an analysis of practice records found that recorded performance across most of these measures stayed largely stable after the incentive ended.

That stability is informative. It suggests much of the improvement had become embedded in systems, templates, and routines rather than depending on a live payment for each action. It also hints that a scheme might do more good by periodically retiring settled indicators and pointing the incentive at new gaps.

How to read incentive evidence

For a careful reader, pay for performance is a case study in separating what a policy measures from what it achieves. The first questions to ask of any such scheme are whether its indicators capture real quality or convenient proxies, and whether improvement on those indicators tracks the outcomes patients actually care about.

The second set of questions is about durability and displacement: does the effect persist without the payment, and does chasing the measured targets quietly crowd out the unmeasured care? Held to those questions, the evidence supports a measured conclusion. Financial incentives can nudge documented behavior and reduce some inequities, but they are a blunt instrument whose benefits are real, modest, and easy to overstate.