Canada faces an acute shortage of doctors and medical equipment. But simply hiring more people or buying more machines won’t fix our healthcare backlog. In this episode of The Executive Summary, professor Opher Baron explains how — from an operations management perspective — we got into this mess, and why smarter planning and better use of data are just as essential as addressing our shortages if we really want to dig out of Canada’s healthcare deficit.
Canada faces an acute shortage of doctors and medical equipment. But simply hiring more people or buying more machines won’t fix our healthcare backlog. In this episode of The Executive Summary, professor Opher Baron explains how — from an operations management perspective — we got into this mess, and why smarter planning and better use of data are just as essential as addressing our shortages if we really want to dig out of Canada’s healthcare deficit.
Show notes:
[0:00] Canadians are in a healthcare crisis.
[0:12] Meet Opher Baron, a professor of operations management at the Rotman School, and an expert in hospital queues. He doesn’t believe more doctors and medical equipment will get us out of our healthcare crisis.
[1:34] A recognition that our healthcare system – and its problems – is nuanced and complex, and this episode is just 15 minutes.
[2:07] Canada has a supply shortage: too few doctors and too little equipment to meet the population demand.
[2:42] This shortage is felt across all sectors of healthcare, but acutely in family care and emergency care.
[3:12] The provincial governments are taking steps to remediate the problems, but it won’t happen quickly.
[4:20] So how did we get here? In short: Poor planning.
[5:31] Hospitals are chaotic, and that makes it difficult to plan.
[7:21] Let’s look at how one issue – continuity of care, coupled with physician incentives – can affect how quickly patients move through an emergency room. Simply, doctors are more likely to take patients at the beginning of their shift than at the end of it.
[9:57] How do you address these types of issues? First, you need to be measuring the right key performance indicators…which we’re not. And you can’t fix a problem, if you can’t identify the problem.
[11:49] Let’s start by recognizing that healthcare professionals aren’t operations management experts.
[12:33] Opher has worked with several hospitals to test new operational efficiencies, including Erie Shores.
[13:58] With the advent of AI, there’s also a lot of opportunity in what Opher calls “digital twin” environments, which will allow hospital administrators to test changes virtually before implementing them in real life.
[15:13] We can’t just throw money at the problem and expect we’ll fix our healthcare system. “I hope that once people understand better what is the important data for them and what they can do with the right data…we can improve the efficiency of our systems, and given the current resources, investing a little bit more kind of in the right places.”
Be sure to check out the Executive Summary back catalogue. We tackle everything from whether we can fix our broken online review system to how extreme heat negatively impacts companies' bottom lines.
Megan Haynes: Canada has a healthcare crisis.
On average, Canadians are waiting 31 weeks to see a specialist, 16 weeks for an MRI scan and upwards of 22 hours in the emergency department if they need a bed in the hospital.
These delays really come down to shortages – we simply don’t have enough healthcare professionals and equipment to handle the patient load.
Opher Baron: Well, one of the reasons why we find ourselves in long lines, long waits and shortage of family doctors, for example, shortage of capacity in emergency departments, surgeries and so on, is because it hasn't been planned well. I’m Opher Baron. I'm a Distinguished Professor of Operations Management in the Rotman School of Management, University of Toronto. I'm also the CEO of SiMLQ, and I've been at Rotman for over 21 years, in the Operations Management Group.
MH: Opher specializes in supply chains – looking specifically at chains that are filled with randomness. Over the years, he’s worked with a number of hospitals to better understand how they can better manage and plan around the unique randomness of a healthcare setting. He’s an interesting person to chat with when it comes to dealing with our current healthcare crisis.
In media and policy discussions, the loudest solutions are often “more.” Hire more nurses. Train more doctors. Buy more MRI and X-ray machines. Build more beds. But, as Opher says, we didn’t get into this crisis overnight, and hiring more, building more and buying more all takes time. There are, however, changes we can make today that can have an impact. The question becomes: are we equipped to make them?
Welcome to The Executive Summary. I’m Megan Haynes, editor of the Rotman Insights Hub.
Musical interlude
MH: I want to start this episode recognizing that the issues facing Canadian healthcare are complex, nuanced and not easily addressed. There are, among other problems, serious divides between things like city versus rural care and systemic issues accessing healthcare for marginalized groups. There are shortages of beds in long-term care homes, not to mention a lack of funding for things like preventative care and mental health initiatives.
However, we only have 15 minutes, so we are going to home in on the specific issue of resource management.
In Canada’s medical system, we have what Opher calls a supply shortage. In very simplistic terms, if you think of patients as the demand – we go to hospitals and doctors’ offices with ailments – you can think of the medical professionals – doctors, nurses, radiologists – and devices – the MRI machines, beds, etc. – as the supply.
We don’t have enough supply to deal with an aging and growing population. One study estimates that by 2030, we’ll be short about 44,000 physicians – and primary care doctors are in the shortest supply. Today, roughly six million Canadians are without a primary care physician.
OB: You can't listen to the news today without hearing about the shortage of family physicians in Ontario, in Canada in general, and in the world more broadly.
MH: This has a really big impact on all elements of our healthcare system – but perhaps most acutely on hospital emergency rooms. The Canadian Institute for Health Information estimates that one in seven visits to an ER could have been managed by a primary care provider.
Politicians, hospitals, nursing unions and medical associations have been calling for strategies to reduce these gaps. Ontario launched two new medical schools, with goals of graduating at least 500 more doctors a year. Provinces have also made headway in reducing provincial barriers to practice and work.
And concerted campaigns have taken place abroad with the hopes of attracting international talent.
But what if this isn’t enough?
OB: We need to start planning our workforce ahead much better, right? The shortage in nurses leads to lots of burnout. COVID kind of highlighted this in a tremendous fashion. So those are things that are, you know, some of them, putting money into it is helpful, but we need something bigger than that. We need to be able to use our data more effectively. We need to plan our capacity more effectively, hopefully some better support for people to prevent them from being sick, right?
MH: After all, nursing takes two to four years of schooling and training; medical school takes four years, followed by up to three to seven years as a resident. That means those first graduates from the new Ontario programs won’t be working for nearly a decade.
So how did we get to this state? Opher doesn’t mince words.
OB: Well, one of the reasons why we find ourselves in long lines, long waits and shortage of family doctors, for example, shortage of capacity in emergency departments, surgeries and so on, is because it hasn't been planned well. When our leadership is here for three or four years, they're very myopic about this – decisions that are 15 years down the road. Investment in this is hard to convince the current regime to do, because they're going to pay for it, the results someone else is going to see. And I need to be elected in a couple of years again, so I cannot waste money and time and effort on those non-myopic things, which is another issue.
MH: So, let’s be real here: investing in training new talent, in particular, is a great place to start, and is incredibly important. But it’s a long-term solution that probably needed to be tackled years ago.
We need to acknowledge that we’re not going to solve our doctor and nurse shortage quickly. That doesn’t mean hospitals and governments are completely without the tools to make wait times, particularly at emergency rooms, more efficient today.
Musical interlude
MH: Hospitals broadly, but emergency rooms specifically, are really chaotic places. We don’t necessarily mean that these are places without order and reason; we mean it more from the perspective that these are places besieged by unpredictability.
OB: So let's think about emergency departments. You go to an emergency department when you need to see a doctor; it's not always life-or-death emergency, right? If somebody breaks a leg or you have a minor car accident, you also go there. If I'm the head of the emergency room, I have uncertainties in how many patients would arrive during each hour during the day. I have uncertainty in what is going to be the characteristics of these different patients – in terms of their acuity level, their ages, their comorbidities, how they arrive, walking or through ambulance.
If there's a snowy day, I'm going to see a higher demand because people slip on the ice. On the other hand, I also have some uncertainty in the supply, so in my resources. Beds and MRI machines are there, but a doctor got stuck on the road with a flat tire – they show up an hour and a half later. A nurse had an issue. He doesn't show today. So all of those things are some of the variability in the supply, which is, quote-unquote, within my responsibility because those are the doctors and nurses that are working in the emergency department.
OB: So if you think about all of this randomness, all of this uncertainty, it creates mismatches between the capabilities that we have and the needs that we have.
MH: Not only are hospitals chaotic places where sometimes the supply – medical professionals – isn’t set up to meet the patient demand, but emergency departments – and healthcare broadly – struggle with other challenges; the main supply is people, and people aren’t uniform widgets.
They come with their own habits, perceptions, motivations and goals. Take the issue of incentives and continuity of care. Traditionally, if you go to an emergency room, there’s an expectation that the doctor who sees you at the start of your visit will be the doctor who sticks with you until you’re discharged. That’s called continuity of care.
However, in Opher’s research, they found that this sometimes leads to an interesting phenomenon. Doctors in the emergency room would see lots of patients at the beginning of their shift, but as the shift went on, they become less likely to pick up new patients, especially as they inched towards the end of their shift.
OB: We are two doctors. We're taking care of patients. And my shift ended. I did the first assessment. You have to do the second assessment because I go home. First of all, we don't like it because of continuity of care. I know these patients. I know what I had in mind when I ordered this diagnostic and this test, and you can read it – if you can read my handwriting, which is terrible, because I'm a doctor, after all. This issue of continuity of care is important, so doctors don't like to move the treatment of a patient in.
MH: If you’re expected to see a patient through to discharge, it becomes more challenging to pick up new patients when you know your shift is almost over. This becomes compounded by the other challenge: payment.
Different hospitals pay their staff differently, but generally, doctors in Canada are paid under a fee-for-service model. That means, while some doctors receive a salary, they are often paid based on how many patients they treat. According to the Canadian Medical Association, more than 95% of doctors in Canada receive at least part of their compensation through this model.
If you’re not there to see a patient through to the end of treatment, you may have to split the payment with another doctor who takes over your caseload, and as a result, there might be less financial incentive to pick up that patient that you know you can’t see through to the end.
This isn’t to say that doctors only see patients at the beginning of their shifts or won’t hand off patients because doing so means they won’t get paid. But what Opher’s research shows is that lots of emergency rooms have a system that disincentivizes doctors from taking a consistent load of patients throughout their shift, instead encouraging them to front-load their patient roster.
As a patient, that means…
OB: If I arrive at the beginning of the shift, I can be lucky in terms of the fact that I'm being seen quickly. If I arrive at the end of the shift, I'm kind of being unlucky, because I need to wait until the beginning of the next shift, essentially. So this creates additional variability in the waiting time. The waiting time is not flat anymore.
MH: So can hospitals address this issue?
OB: They can start thinking about how to change their internal incentives, or how to change their behaviour, in order to reduce the impact of the shift effect and how it changes their ability to meet KPIs.
MH: How could hospitals shift their pay structure to be an incentivizer? Or can they create a system where the decision is pulled from the doctor’s hands? These are big changes that can’t be done until they tackle the next challenge: data collection.
OB: There's a lot of data collected generally in healthcare as well. It is not always the right one to support operational planning. So much of the data that is being collected today is motivated by financial incentives and accountability.
MH: In Ontario, for example, part of a hospital’s scorecard includes things like how long you waited in an ER before being assessed, how quickly you were offloaded from an ambulance, and how many people left the ER without being seen at all. This isn’t wrong per se, but for operations management practitioners, it’s not enough information.
OB: And what you are measuring is, say, ambulance offload time, which is one of the important KPIs in the emergency department. You know when the ambulance arrived and you know when the patient went from ambulance to, say, triage, but you don't know when the nurse started taking care of these ambulance patients in order to kind of move them to triage. So you don't have this time stamp, which implies you don't know how much time your resources spend on this activity. This is something that is missing.
Musical interlude
MH: So, what can we do about it? Well, first, let’s recognize that doctors and healthcare administrators aren’t operations management professionals.
OB: So I talked about my wonderful knowledge in theoretical models of randomness and uncertainty. You don't want me to treat you in a hospital bed. I have no idea how to do that. The medical doctors, the nurses, the support staff that work in the hospital, they know how to do their work, but many of them don't have the right training in terms of running an operation. So every time I have a doctor in my MBA classes, we start to talk about queuing theory and waiting time, you can see their eyes open: “Yeah, this is important. I want to understand what's going on here.” Unfortunately, not many doctors are going through such training.
MH: Better, more robust data collection – which doesn’t necessarily need to align with provincial metrics – is important as well. Because we need to understand if our incentive models, our scheduling structures, our flow is working correctly, policymakers and hospitals need the right information to make sure they can better plan for the future.
They don’t have to tackle the challenge alone, and many researchers and academics – like Opher – are working with hospitals now on figuring out some of these optimization challenges.
He’s done work with several across Ontario, including a recent partnership with Erie Shores, which was struggling to optimize patient flow in its emergency department. Erie Shores wanted to know, was there a better way to assess, process and help people who, say, break a bone, but don’t necessarily need to stay in the hospital? Historically, planning for resources around these non-acute needs is based on previous year’s data.
Again, very simplistically, if 10 people broke their leg in 2023, then you might expect 12 people to break their leg in 2024. But what happens if skateboarding becomes super popular and suddenly 20 people have broken their leg in one year? How do you staff for, and change your ER plans, on the fly?\
OB: When you have to choose different interventions, different actions and evaluate what is their impact on performance, you want to have an evaluation of which project is better to choose between project one and project two in the context of managing waiting times – thinking about the emergency department, elective surgeries, MRI waits, right? All of these wonderful things that we hear about, and when we experience our waits are really long. The way to get an evaluation of what is the impact of different interventions, I would say, requires simulation.
MH: This is where AI comes in – and it’s an area Opher is particularly excited about. In July 2024, he and his colleagues launched SiMLQ, an AI simulator. It’s essentially digital versions of the hospital setting.
In these digital simulations, admins can pull a lever on a change they want to make – say, testing new scheduling systems – to simulate what would happen to the broader environment. What knock-on effects could it have? Would it actually save the hospital money? Would it help patients get seen faster, or might there be issues down the pipeline?
OB: “Oh, it's a snowy day right now. My digital twin knows exactly how many people are on the floor. It knows exactly how many nurses and doctors are on the floor with their diagnostics. It has a good ability to see what would happen on this snowy day where demand is a little bit higher, and then I can decide if I want you to come an hour earlier, two hours earlier – what is the benefit in terms of performance to the system?”
MH: Now, building and training these AI systems takes time too; so, like training doctors and nurses, AI won’t magically solve hospital staffing issues. But it’s worth remembering that they are a tool we can use while we wait for those thousands of doctors and nurses to make their way through training and bureaucracy. Because in the end, you can’t just throw money at healthcare and hope that fixes the systemic and chronic shortages we’ve built over decades.
OB: Just like anything else, it's using the money in the right fashion. Adding 25% budget to all hospitals is not necessarily going to solve the problem, right? You need to invest in the right places. You need to think about it better. And also, as I mentioned, in terms of strategic planning, we need to start planning our workforce ahead much better.
Investing in the right places – the ROI on these different investments is not always known ahead of time. So that is something that requires in-depth learning and more structured decision-making. I hope that once people understand better what is the important data for them and what they can do with the right data, we will have better data and we can push and improve the efficiency of our systems better, and, given the current resources, invest a little bit more in the right places.
Musical outro
MH: This has been Rotman Executive Summary, a podcast bringing you the latest insights and innovative thinking from Canada’s leading business school. Special thanks to Professor Opher Baron.
Join us next month as we chat with professor Minlei Ye on why we need to take a serious look at auditor bias.
If you’re just tuning in for the first time, check out some of our earlier episodes. We tackle everything from how to deal with burnout to whether AI can really be regulated. Make sure you subscribe on Apple, Spotify, YouTube or Amazon. And please consider giving this episode and the series a five-star rating — it’s hugely helpful in getting the word out.
And if you want more bold ideas in less time, check our Visiting Experts, a monthly podcast featuring sharp conversations with today’s top business thinkers, including Malcolm Gladwell, Oliver Burkeman and Amy Edmondson. It’s available wherever you get your podcasts.
This episode was written and produced by Megan Haynes. It was recorded by Dan Mazzotta and edited by Avery Moore Kloss.
Thanks for tuning in.