'By supporting doctors in making decisions, I help patients' - Hester Lingsma


Hester Lingsma is active on measuring outcomes on varies levels. During the webcast ‘Uncertainty as the new normal’ on 28th October she tells more about collecting data and a prognostic model to stimate the likelihood that someone should be admitted to intensive care (ICU).

Register for the webcast here

How did you come into contact with value-based healthcare?

Over the past years, I have been active on this topic on varies levels. My first introduction to measuring outcomes was during the open days before choosing a study. Marc Berg, then a lecturer at Erasmus University, explained that in certain hospitals in the Netherlands, you had 10times more chance to die from appendectomy. By now I know through my own research that this was probably 100% coincidence, but thinking about whether or not there is quality differences in Dutch healthcare and how to measure them immediately fascinated me, and still does. During my graduation research, I investigated whether hospitals that had good outcomes in CVA patients also better adhered to the evidence-based guidelines. This turned out not to be the case, the patient outcomes of each hospital were mainly determined by the disease severity of their patients, not by their 'quality' of care. During my promotion under Ewout Steyerberg I was given a lot of free space to continue with this topic measuring quality of care. Competition between healthcare providers is a pillar of the healthcare system. This competition cannot function without reliable information about quality. But after decades of research, we still don't have how to properly measure that quality.

What are you doing right now?

I currently lead the medical decision-making research group of the Department of Public Health at the Erasmus MC. After my promotion, I continued to work there. We have a multidisciplinary team of about 15 colleagues; doctors, psychologists, econometrists, statisticians and epidemiologists. We do research based on mostly clinical datasets where we focus on optimizing treatment decisions for individual patients. So less on, for example, the effect of policy decisions. We evaluate clinical interventions and create prognostic models for individual patients, always in collaboration with clinicians.

In the research on value-based healthcare, my research group works closely with the value-based healthcare team at Erasmus MC. There is an increasing need and motivation in Erasmus MC to evaluate the implementation of value-based healthcare scientifically. We encourage stakeholders to pay attention to a good evaluation and think along about suitable research methods and designs. For example, whether it is possible to create a control group. At present, for example, larger projects under way are evaluating the roll-out of the PROMIS-10 at Erasmus MC. This goes in phases per department, allowing us to use all departments, some short, some longer, also as control group. Effect parameters we plan to look at are process measures such as consultation duration, referrals to other departments and multidisciplinary consultation. In addition, together with the Departments of Neurology and Radiologist, we are doing a multicenter study on the effect of performance feedback to caregivers of CVA patients, the PERFEQTOS study. It's a stepped wedge cluster randomized trial. This means that every few months a number of hospitals switch to the intervention group in randomised order. As the intervention, they receive information about their processes and outcomes in CVA patients via dashboard, also compared to other hospitals. They need to form an improvement team and work on possible improvement actions. We investigate whether all this actually improves the processes and outcomes of care.

What is/was the influence of COVID-19?

As a result of COVID-19, we have been asked to create a prognostic model that can be used in the emergency room. With simple clinical parameters such as age and respiratory rate, we can estimate the likelihood that a patient should be admitted to intensive care (ICU) and the likelihood that a patient dies. This model is currently being tested by emergency physicians. In addition, we have been working on a decision-model on prioritizing deferred care. Based on literature and data, we made an estimate of how much health loss results from postponement of specific surgeries. You could than argue that the surgeries with most health loss associated with postponement should be prioritized. But that's only one ethical perspective. You could also say the most sick should be prioritized, or health care personnel. Being transparent in your arguments is essential when making such difficult decisions.

Do you have an inspiring example/ person?

I take my pleasure and inspiration from working with doctors who dare to take a close look at their own actions. Those who are not afraid to question the contribution of the treatments they give. By looking critically with each other 'What do we do?' and 'Does that make the patient better?' that gives me energy.

What's your mission?

With my research I hope to be able to contribute to the optimization of decisions that doctors make in order to ultimately improve the quality of life of patients. I don't directly help the patient, but I help doctors to help patients as best I can.

What do you think the future of care looks like?

In the future I expect more attention to be paid to prevention. The current healthcare system is financially unsustainable if we focus only on curing patients. These people are already sick. In addition, I expect that we will learn faster because there is more and more data available because we measure continuously. But measuring alone doesn't make us any wiser. Learning from data can be done under certain conditions, and thinking about this is what we epidemiologists do. We must keep thinking about what the question and how and with what data we can get an answer to it.

Do you have any tips for others?

In terms of value-based healthcare implementation and all such initiatives, I would say that we need to think about how to evaluate them. See if a pre- or post-measurement is possible, and measure that where you expect effect, for example on process or organization parameters. Provide a control group and ideally a randomized control group. We provide a lot of care, without knowing what this contributes to the patient's health. And especially with value-based healthcare initiatives, where real damage will not happen quickly, it is easy to think that everything you do has a positive contribution. But you can only spend money and resources once.  Let us take the responsibility of continuing to critically evaluate our actions in care and continue to learn.


Hester Lingsma | Associate Professor of Medical Decision Science, Department of Social Health, Erasmus MC |  

Continue reading


X (voorheen Twitter)

Volg Linnean online


Over ons

Het Linnean Initiatief is een landelijk netwerk van koplopers en aanjagers van waardegedreven zorg. 

Lees meer


Ook deelnemen aan het Linnean-netwerk?
Sluit u aan bij het netwerk.


Stichting Linnean Initiatief
KvK: 75738007

Neem vooral contact met ons op voor meer informatie, tips of suggesties.