“Pay-at-risk” became a political flash point in Alberta last month when Health Minister Fred Horne fired the Alberta Health Services board when it didn’t agree to withhold the at-risk part of the compensation package for about 100 executives.

Alberta Health Services (AHS) had introduced pay-at-risk for health care executives in 2009.

With pay-at-risk — also known as pay for performance, variable pay, performance linked pay, and performance based compensation — a certain percentage of remuneration is not guaranteed, but is instead dependent on meeting specific objectives.

The issue had also sparked controversy in Ontario in 2011 when, faced with a public outcry, the province reneged on an agreement to pay “performance incentive awards” to a number of eHealth Ontario employees.

In both provinces, pay-at-risk was eventually honoured (click here and here), because of contractual agreements.

Widely accepted in the private sector

Although pay-at-risk is widely accepted in the private sector, where payouts typically come from company profits, it is more controversial for publicly financed hospitals, where compensation is a budgeted item, and most funding is from the public purse.

And it can be particularly controversial in times of economic restraint and pressure, when hospital staff may be laid off and the pay-at-risk portion of a health care chief executive officer could well exceed the average citizen’s annual earnings. For example, AHS chief executive office Chris Eagle was reportedly due $108,000 in pay-at-risk this year.

It doesn’t help that pay-at-risk is sometimes erroneously referred to as a bonus. In the former, part of an executive’s total compensation can be withheld if certain conditions are not met, whereas a bonus is an “add on” to the total compensation package.

Likely because of the controversy, Healthy Debate was unable to find someone  inside the Alberta government to comment on the role of pay-at-risk for health care executives in that province, but the situation in Ontario sheds some light on the dynamics of pay-at-risk.

Ontario legislation mandates “performance based compensation”

In Ontario’s 2010 Excellent Care For All Act (ECFA), the province mandated that all hospitals set aside a certain portion of chief executive officers’ pay as “performance based compensation” and tie the amount of payout to the extent to which CEOs meet performance improvement measures spelled out in the hospitals’ quality improvement plans (QIPs).

The Act did not specify what percentage of total compensation should be put “at-risk”, although the 2011 Manley report (Report of the Independent Expert Panel on Executive Compensation in the Hospital Sector) recommended between 10% and 30%.

Professor Adalsteinn Brown, Director of the Institute of Health Policy Management and Evaluation at the University of Toronto, says the amount of pay that is placed  at-risk for Ontario hospital executives is not important — even 1% would be okay in his opinion. (Some argue that 5% or less is really just a token, and that performance evaluations at the time of annual salary reviews could serve the same purpose.)

Pay-at-risk “helps people focus on a specific set of goals”

But pay-at-risk is important because it is linked to improving quality — a key goal of the health care system — and “it helps people focus on a specific set of goals,”  says Brown.

This is a step forward because “in reality the system is not doing a lot of quality improvement, in reality it is not giving a lot of data back to people…and in reality people are being paid for program growth rather than program improvement.”

In 1999, research that Brown co-authored found evidence that the salaries of CEOs of public hospitals in Ontario were “largely unrelated” to the hospitals’ financial performance.

However, research on the effectiveness of pay-at-risk in public sector hospitals is scarce and inconclusive (for example click here and here). As well, there is debate about the control that CEOs can actually have over quality improvement, notes Moishe Greengarten, associate director for the Hay Group health care consulting.

And key issues include being able to identify criteria that actually advance outcomes and, at the same time, avoid unintended consequences and/or use the wrong measures, says Greengarten. To that end, he says there needs to be more research and an evaluation of the impact of pay-at-risk in Ontario public hospitals.

So how is pay for performance playing out in Ontario?

The ECFA followed hard on the heels of Ontario’s 2010 Broader Public Service Accountability Act, which froze executive compensation. (The freeze on hospital CEO salaries was extended in a later amendment.)

Before ECFA, 25% of Ontario hospitals had incentive pay plans for their chief executive officers, according to the Ontario Hospital Association.

In context of freeze, pay-at-risk “carved out of existing compensation”

For the many large hospitals in Ontario that had already implemented pay-at-risk, (some in the early 1990s) and, in many cases, developed their own QIPs, compliance with ECFA meant they had some added transactional costs.

But for smaller hospitals, all this was new and somewhat burdensome largely  because, in the context of the province’s public sector compensation freeze, pay for performance targets had to be “carved out of existing compensation,” as the OHA put it in a series of explanatory slides.  In other words, CEO base pay had to be reduced in order dedicate a portion to pay-at-risk.

To help its members, the OHA produced a guide for implementing a CEO compensation framework.

Anthony Dale, the OHA’s interim chief executive officer, acknowledged that pay at risk is somewhat controversial among OHA members, adding that the percentage put at-risk likely ranges from about 1% to about 30%.

Asked if hospital CEOs regularly receive a very large proportion of their pay-at-risk, Dale says he has no data on this. Brown, who was formerly a public servant in Ontario, says in his experience it is “pretty rare” for a full payout to be made.

And Tom Closson, who received pay-at-risk when he was CEO of two large Ontario hospitals, says he often did not receive the full payout. “Of course I always did my best” before pay-at-risk was implemented, he added, “but it certainly helps if CEOs have skin in the game.”

The ECFA did not specify QI measures that are common to all hospitals, yet an alignment of at least some QI goals across the province is desirable, says Dale of the OHA.

OHA lists recommended core quality improvement indicators

To that end, the OHA has provided a list of “core recommended indicators for hospital QIPs.” The list includes reducing  the rate of hospital acquired infections,  the incidence of new pressure ulcers, the use of physical restraints,  and rates of death and complications related to surgical care.

The document also lists the indicators that specialized hospitals, for example hospitals for rehabilitation and mental health, have identified and developed for their particular circumstances.

Physician says pay-at-risk may encourage overly modest goals

Dale says the Ministry of Health and Long-term Care “has taken the view that you start with indicators that are straightforward and less complex, and then accelerate—as change happens the goals should become more stretch goals.”

Still, Greengarten says an evaluation should be undertaken of the widespread introduction of pay-at-risk among Ontario Hospitals.

One Ontario physician with an interest in quality improvement, who wished to remain anonymous, says there are several potential problems with the way that pay-at-risk has been implemented in many Ontario hospitals. He noted that there is usually no independent audit of the data used to determine whether a particular target has been achieved. As a result, good performance in some cases may simply be an artifact of poor quality data.

In the absence of independent auditing, an executive team whose pay is tied to performance may be unlikely to insist on reliable, accurate data. The physician also noted that pay-at-risk may encourage hospital executives to set overly modest goals. For example, it would be foolish for a CEO to tie his or her pay-at-risk to a 100% staff influenza vaccination target, even though that should be a goal for all hospitals.

For his part, Brown noted that transparency about pay-at-risk and how QIP targets are addressed is important because “when you make things public, performance improves.” Overall, he sees the ECFA as a positive development, but adds that what the Ontario health care system needs now is “a focus on performance, a common set of measures of performance, putting that information out as transparently as possible, and getting some agreement on big goals.”

As well, the push for quality improvement needs to go beyond hospitals into the rest of the health care system, he says.