How Much Flu Vaccine Do Pharmaceutical Companies Need to Produce?
When you picture the intersection between business and health care, you might see social impact – innovators who design new ways to make us all healthier.
Take, for instance, Medha Sharma, who plans to graduate with a Wharton School MBA degree in 2025. While Medha is studying to be a doctor, she is also in Wharton’s Health Care Management Program because she hopes to build a healthier world through innovation. Medha is moved by the idea that health outcomes are influenced by non-medical factors like education and economic stability – otherwise known as social determinants of health (SDOH). She founded a Social Determinants of Health Accelerator, funding medical students to work as consultants for SDOH-focused start-up companies that work to improve health challenges like addiction and chronic absenteeism in schools.
Wharton Global Youth will be exploring more about how the entrepreneurial mindset is making a difference in medicine – for both patients and practitioners.
Studying the Flu
But first, we want to stop by Wharton’s OIDD, better known as the Operations, Information and Decisions Department, where students are learning to make better data-driven decisions that have a very different impact on the business of health care – and how companies operate and manage public health.
Hummy Song, a Wharton associate professor of operations, information and decisions, as well as health care management, visited some 300 high school students learning in Wharton Global Youth’s summer programs on the school’s Philadelphia campus, to offer a look into her data-informed research about the pharmaceutical industry, which discovers, develops and manufactures drugs to treat illnesses.
The goal of OIDD, noted Dr. Song, is to give students frameworks, tools and analytical skills to make more informed, data-driven decisions as future business leaders and managers.
She dug into the data to illustrate the decision-making and operations of pharmaceutical companies that produce flu vaccines (think Sanofi), a core part of the health care industry that must balance business priorities like supply, demand, costs, and profits.
“Every year pharma companies are developing flu vaccines,” noted Dr. Song, whose research focuses on identifying ways to improve the performance of service systems, with a particular emphasis on the health care sector. “In the U.S., they’re going to pay attention to data from Australia or New Zealand, where the flu season comes much earlier. So, we’re looking at strains of mutated virus that are being captured and studied there and the U.S.-based pharma companies that develop vaccines for the U.S. market are going to use that to try to come up with a vaccine that’s going to be more up to date in terms of a strain that might show up in the U.S. Every year this cycle is happening and every year in addition to the R&D [research and development], pharma companies are going to have to make some decisions about how many doses of vaccine they need to produce.”
How do the numbers inform those decisions? Dr. Song offered a few examples:
Flu season at a glance. The Centers for Disease Control and Prevention or CDC releases data that helps to guide decision making in the health care industry, including the annual estimated number of flu-related illnesses, medical visits, hospitalizations and deaths. This helps pharmaceutical companies understand the scale and impact of the previous year’s flu season.
Vaccine effectiveness. “The idea behind having flu vaccines ultimately is to try to prevent serious illnesses and deaths,” noted Dr. Song. Data revealing how many flu-related illnesses, medical visits, hospitalizations and deaths were estimated to be prevented by flu vaccination in the previous season is important intel for pharmaceutical companies as they make decisions about vaccine production.
Who is getting a shot? Data tracks the vaccination coverage rates among different adult age groups over time. This gives the pharmaceutical companies insight into the demand patterns and trends for flu vaccines, which is crucial for forecasting expected demand.
“How much flu burden is there? How many people actually want the shot? This is the kind of data companies want to understand…to determine how many doses of flu vaccine they should be producing,” said Professor Song, adding that pharmaceutical companies have to decide on vaccine supply before they know the actual demand since production begins well before the arrival of flu season each year.
Dr. Song detailed other critical information (OIDD!) related to operations management, pricing and risk management that influences pharma decisions about vaccine production:
💊 The optimal decision is to supply exactly the amount of vaccine that matches the expected demand, which will maximize profits by avoiding unsold vaccines (overproduction) and lost sales (underproduction).
💊 You need to know the costs that the producer of vaccines will incur for producing each dose and also the vaccine producer’s revenue. So, in addition to demand, how much does it cost for me to produce these things? And if I sell them, how much am I going to make off of that?
💊 Companies need to consider the costs of underproduction (lost revenue per unsold dose) versus overproduction (cost per unsold dose) to determine the profit-maximizing supply level, which is typically slightly below the expected demand.
💊 Not every question decision-makers are faced with is about expected profits. Sometimes you’re going to have competing objectives that you need to weigh and balance, like the overall welfare of the population and preventing illnesses and deaths.
Aligning supply and demand in the health care industry can be challenging, prompting pharma companies to use data and analytical models to make smarter vaccine-production decisions. Critical ratio, an equation that helps optimize supply decisions in the face of demand uncertainty, is one framework that OIDD students learn to guide operations and production management.
Concluded Dr. Song: “The biggest compliment I hear from our students once they’ve learned about these things — especially as they worked on summer internships or maybe they are MBA students who were working at different companies – is that they say, I wish I knew about [data-driven decisions] because I would have not only made better decisions, but also been able to defend them better in meetings, and confidently would have been able to say: here’s the data I utilized, and here’s the reason why I think you should make this decision.”
Conversation Starters
How does data help to inform pharmaceutical companies about the challenge of supply and demand for flu vaccines?
What is OIDD? Does it sound like something you would study? Why or why not?
Take a look at the CDC’s estimated flu burden data. What interests you about this information? How might it inform your own decisions, in business or otherwise?
At its core, business is not about making money or purchasing assets. It’s not just an activity either. Business is a way of thinking. This article about the intersection of business and healthcare is powerful reading for our time – we must always remember that the global pandemic is not far behind us, and acknowledge how much mismanagement of resources took place in 2020. I personally was oblivious to the fact that the vaccine industry was its own subset of Big Pharma.
Dr. Song and her colleagues are employing data science to fix issues with the American vaccine delivery system. This is a fine way to apply data-driven decision making skills. OIDD is also critical to the survival of the millions when flu seasons reach their zenith. Statistical models around estimated flu burden can axe the death/severe illness rate, especially in lower income neighborhoods. People sometimes overgeneralize that Pharma companies lack concern for the public, but this article proves otherwise. On that note, I would also like to include my thoughts on a potential new horizon in healthcare for data and data-driven decision making.
While reading this article, I thought of an experience I had about three years ago. In November 2022, when I was 14, I was diagnosed with appendicitis and therefore, had to undergo a life-saving surgery. I ended up staying in the hospital for over a week in recovery. I remember, very well, the arduous process involved in finding the right doctors in a large hospital. I happen to live in a country that is praised worldwide for its universal and socialized healthcare (Canada). American readers might initially look at this with envy.
But there’s a hidden dark side to “FREE HEALTHCARE FOR ALL!”. I had to wait 9 hours in the hospital across a weekend to get treatment for an inflamed appendix, and I’ve heard in the news of wait times of 21 hours in some Canadian hospitals.
People literally die waiting for treatment. We need to fix this issue.
A 2017 study showed that 29% of Canadians reported waiting four or more hours in an ER – the worst of eleven “developed” countries. Our country’s hospitals aren’t streamlined and their management systems are, at least in my humble opinion, outdated. And it’s not just Canada – the US and other countries are also burdened by flawed or corrupt healthcare systems.
I get that not every problem can be solved simply by changing one’s outlook on things. Sometimes we just need more cardiologists or more nurses or more paramedics, and there’s nobody in the workforce to fill those roles. Sometimes public health authorities are strapped for cash. I, as a rising high school senior, happen to know that my country’s healthcare system is billions under budget. But implementing data-driven decisions (much like OIDD) in this setting can make an extraordinary difference (no matter how objectively small) – and that’s the one between a life and a death.
Perhaps, instead of responding reactively to patient complaints without a plan, hospitals could leverage the power of historical data like Dr. Song did to predict the number of vaccine takers. They might take every day’s average level of activity for 365 days and model the best heuristic solution using algorithms. Two professors, Dr. Xu and Dr. Chan (from Stanford and Columbia University Business Schools respectively) have already constructed such a model that is proven to alleviate ER wait times by up to 15%. Hospitals, using these, would notice what time of day happens to have the most patient volume. Perhaps patients could see a doctor in less time, instead of needing to wade their way through multiple layers of hospital bureaucracy.
Perhaps, to increase cooperation between medical staff, a hospital could make patients’ data easily accessible. Physicians would not need to start from scratch when dealing with a patient and his/her symptoms. I’m proposing a unified knowledge network. This would allow a patient to reach the right doctor and receive the best care at a personal level.
And maybe, just maybe, hospitals could achieve optimal or near-optimal levels of resource allocation. They could apply the economic principles of supply & demand. Or scarcity or incentive theory. Critical equipment such as medical oxygen, ventilators, hospital beds, and life-support machines are limited in number; these should be deployed where they are needed the most. Hospitals could divert patients who need these resources to survive, limiting the strain on one particular node in a large system. We saw global shortages on these items at the peak of COVID-19, and frankly, it is heartbreaking that your hospital’s ZIP code determined whether you lived or died at that time.
Dr. Song tells us one last thing in this interview, which is a piece of information that her students relayed back to her. She says that the use of data helps us become more confident in our decisions. I believe that this is the key takeaway from this article. Whether it’s delivering flu vaccines, or streamlining emergency room operations, or something much smaller like deciding which route to take to work, data gives people a clear roadmap. When things go wrong, we know where they break down. When our opponents question our calls, at least we have solid reasons.
With the help of data, we can – and should – run hospitals more effectively. 100 or 50 or 30 years ago, we had no choice in this matter. We were compelled to work with what we had. Now, however, technological infrastructure is stronger than ever before. And organizations like OIDD are showing us the endless possibilities. So let’s get started.
I’d like to end with a big thank you to Wharton Global Youth and Ms. Drake for publishing this article and leading the Comment and Win Contest!!