by Dr. Helmy Haja Mydin. First published in The News Straits Times 14 April 2014

HOW should doctors be rewarded? The traditional fee-for-service model, where doctors are rewarded/paid on volume of activity, is coming under increasing fire.

There is an argument that remuneration should be given on the basis of patients’ outcomes, that is, doctors be rewarded for a job well done. However, the criteria for defining “a job well done” are less than obvious, especially when the complexity of modern medicine is taken into account.

In the pay-for-performance (P4P) model, doctors are given financial incentives for better clinical outcomes. For example, general practitioners in the United Kingdom are given financial incentives if they are able to identify and treat patients with high cholesterol levels. The rationale of such a model is to promote good clinical practice, as income is linked to the quality of care provided.

There are a number of advantages to the P4P model. There is evidence that it decreases unwanted variations in healthcare provision, which not only improves quality and safety, but helps curb the ever escalating cost of healthcare.

A study by the Institute of Medicine in the United States also noted that the traditional model can disincentivise doctors as it would be in the financial interest of for-profit hospitals to encourage recurrent admissions — unfortunately, not all doctors are motivated by altruism or professionalism.

While the ultimate aim of the P4P model is to stimulate quality care, a secondary consequence may develop if data collection is made public. Patients will be empowered as they will have insight into centres of excellence, allowing them to choose between healthcare providers.

Implementing P4P

The keystone of any change in a system is data collection — the old management adage of you can only manage what you measure rings true. Significant human and financial resources will be necessary for data collection, analysis and storage. The single most important first act is identifying the type of data that is relevant.

For each specialty or hospital, standards will need to be identified and used as benchmarks to evaluate progress. The Ministry of Health already does this: its “Technical Specification: Performance Indicators for Medical Programme” identifies numerous outcomes that are deemed to be representative of quality care. These include clinical outcomes (e.g. 90 per cent of patients with waiting time of > 90 minutes to see the doctor at a specialist clinic).

The most important step would be for the clinicians to buy into this concept and to practise in a culture that values continuous monitoring and auditing.

This will involve additional training for both junior and senior medical staff alongside the rolling out of a robust IT support network that will allow analysis and dissemination of data.

Once data can be obtained and verified robustly, results can be used within the framework of a P4P model and made available to the wider public.


The United Kingdom’s Society for Cardiothoracic Surgery provides analysis of heart surgery outcomes. The details of surgery performed and mortality rates for specific procedures for all hospitals and most surgeons are made available to the public.

However, the Society makes it a point to modify outcome scores based on the complexity of cases. Collecting and analysing data on clinical performance is not as straightforward — for example, surgeons who have higher levels of deaths may have been operating on complex patients who were refused surgery elsewhere. For this reason, efforts at introducing P4P must be tampered with caution as surgeons may avoid the severely ill patient.

A possible unintended consequence of P4P might be to encourage “tunnel vision”, where aspects not covered by performance indicators are neglected. It is also important to acknowledge that important aspects of medicine are not easily quantified — a doctor’s communication skills and the role of a palliative care team in easing the passing of one’s death are such examples.

The ministry should also bear in mind that psychologists have noted that characteristics of individual doctors (including intrinsic motivation, professionalism, and altruism) may influence their response to incentives.

Performance data in the context of an incentive programme will also trigger these factors as one would hope that many doctors would be driven to improve themselves and enhance their reputation.

Incentives matter in healthcare, and subjecting doctors to a P4P model would help develop meaningful and lasting quality improvement initiatives. Transparency will allow patients to identify institutions of excellence, and will also enable publicly-run hospitals to compete with private institutions for patients in a real market environment.

– – –

Dr. Helmy Haja Mydin is a founding Associate of IDEAS

Leave a comment