If We Cannot Solve Vaccine Equity Globally, We Cannot Escape Variants Locally
Bhramar Mukherjee
John D. Kalbfleisch Collegiate Professor and Chair of Biostatistics, School of Public Health
In the last two years I, like many of my friends, have lived, breathed, and sighed on the epidemiology of SARS-CoV-2. When our team started modeling the trajectory of this insidious virus in India in March 2020, none of us expected the pandemic to last this long. We ended 2020 on an optimistic note, with multiple vaccines showing promising efficacy results in clinical trials and receiving emergency use authorisations. The rollout of the vaccines brought tears of joy to millions of eyes as we welcomed 2021.
Those moments will always represent the triumph of science and humanity to me. However, after this uber-uncertain vaccine-versus-variant roller-coaster of 2021, first Delta and then Omicron, we have learned some key lessons as we embark on the journey in 2022.
No country can possibly get ahead of the other. There is no quick, solitary win here.
Here are my top five lessons from 2021, followed by a summary of what I think 2022 may bring.
1. We Need Global Collaboration to End This Pandemic
It is a synchronized relay race of the whole world together, if there is something like a relay marathon. No country can possibly get ahead of the other. There is no quick, solitary win here. If we cannot solve the issue of vaccine equity globally, we cannot escape the variants locally.
A colleague of mine aptly says that, if we replace the “i” in illness with “we,” it becomes wellness. In this time of contagion, our own potential outcome does affect the outcome of others. Due to this spillover effect, we must make individual sacrifices for the collective good.
2. The Need for Robust and Transparent Public Health Data Systems and Large Cohorts with Integrated Data
Many of the key papers on COVID have come out of the UK, Israel, and Denmark due to their national health/insurance systems database or population-based cohorts. Integrating testing, vaccination, sequencing databases with clinical data were key to identifying new variants and characterizing their immune escape, transmissibility, and lethality properties. These multi-platform, cross-talking databases also gave us real-time data on vaccines and their effectiveness.
After two years of COVID, I have yet to see the daily hospitalization data across India.
In the early days of COVID, they primed us about who is at the highest risk of hospitalisation due to COVID. When I started working on the India data, the paucity of data was eye opening. It is nearly impossible to find national-level, age-sex disaggregated counts of COVID deaths. Many deaths are unregistered and go undercounted. After two years of COVID, I have yet to see the daily hospitalization data across India. Some data sources exist on government websites, but it takes incredible amounts of effort to web-scrape and collate them into a usable format for running models.
My conclusion is that we have many excellent mathematical and statistical models on COVID, but very few countries have had reliable and accessible data to inform/train those models. It is time to invest in such systems, as this is certainly not the last public health crisis the world is going to face. A robust data system bolstered by data transparency can evoke public confidence and trust in policies. Data denial and data opacity do not help—not even the government.
3. Embracing the Notion of Uncertainty and the Humility of Incomplete Knowledge
To adequately model COVID we needed a model for virus transmission, a model for virus mutation, a model for public health interventions and their effects, a model for vaccine escalation and effectiveness, and finally a model for human behavior. Identifying different components of these models and tying them together is nearly impossible.
As modelers we are used to working behind the scenes.
Thus, whatever we predicted was tenable only in the short term. Long-term, wide-ranging predictions were meaningless. Even within a 30-day horizon, one had to be modest about how little we knew and how rapidly things on the ground could change. The same is true in daily life—decisions about a trip, a gathering, or a classroom lecture need to change overnight. Elasticity and adaptability are key here.
4. Public Engagement, Scientific Communication, and Pace of Research
Harnessing the information tsunami to identify and deliver key messages was challenging, even for scientists. As modelers we are used to working behind the scenes. To be in the public eye was a huge leap for me, but many did this very successfully.
The rapid pace of research and the urgency to influence and guide policy meant seeking alternative modes of communication and often settling for a “good” analysis as opposed to an “ideal” analysis. It led to not just writing peer-reviewed papers but using social media, newspaper articles, blogs, and Medium articles to communicate.
Scientists had an opportunity to directly engage with the public in real time. This is a completely different format of communication than we are used to. The trolling and politicization were also something new. When I first started talking about the degree of undetected infections and deaths in India, I was highly criticized. Months later, data came in from seroprevalence surveys and excess death studies that substantiated the model assertions. When many were predicting a third wave in fall 2021, our models never did, and I went on television and say that.
We have struggled with charting a path that strikes a balance between alarmism and denial.
You must nuance your messages, as such predictions can also induce a false sense of security and lead to relaxed adherence to public health measures. Sticking to scientific truth is crucial.
5. Striking a Balance
Throughout this pandemic, we have struggled with charting a path that strikes a balance between alarmism and denial. Even now, some are proclaiming a cataclysmic Omicron wave while others are saying Omicron is clinically mild and will lead to endemicity without causing a spike in deaths. There have been so many COVID curveballs thrown at us that a middle road is extremely important to define—for individual decision-making and for public policy.
Some scientists proclaimed that, by the end of 2021, COVID would be endemic like the flu. That was premature and unwise. Millions of people worldwide are dealing with long COVID, many countries in Africa have a vaccination rate under 20%, and almost no country in the world has a high degree of the population with booster doses, which seems to be the only thing working to offer a reasonable level of protection against Omicron.
The situation after two years of an intense, arduous fight is still quite bleak. Our reservoir of hope, resilience, and grit is depleting over time. However, we must think about joy and mental health and find a way to live with COVID.
I personally am trying to adapt to a dial-up, dial-down modality of life in concert with the virus incidence curve. This disrupts our regular rhythm and pattern of life, but we must evolve and accept the randomness of this bizarre period. We must also grieve and recognize our loss instead of being in a mad rush to get over the pandemic. The second wave in India led to so many untimely deaths that we now need to heal as a country.
For me, as an immigrant scientist sitting here in the US, watching my family suffer through COVID in India was painful. Every day I process the trauma of last summer and try to contribute to the fight against COVID.
As I try to wrap my thoughts around COVID 2022, the one certain thing is that it is not over. The immune escape properties of Omicron are making me very nervous about India. The high degree of past natural infections combined with vaccination protected India in late 2021. But if you consider the peak timing of past infections in April-May 2021 and the fact that almost no one has booster doses yet, the predictions for next year are bleak.
It is important to know how many people have had “three hits”—one infection and two vaccines—to arrive at an accurate prediction. With waning immunity and breakthroughs, a very large number of infections can happen quickly, and even a small fraction of that needing hospital admissions will cause the fragile health care system to crumble. Instead of being complacent and feeling exceptional, we should plan for the worst and scale up healthcare resources, including new treatments for COVID, so that we face the Omicron wave as best as we can.
This is not the last variant. This is not the last pandemic. We should closely track hospital admissions and take stringent measures if capacity runs the risk of being full. Quality masks and rapid tests should become common practice at gatherings.
As a statistician, I feel we need to do more work in quantifying the effect of interventions that were rolled out in bundles through worldwide natural experiments. What works and what does not work for India—for example, enforcing social distancing—is difficult. But perhaps it is not difficult to distribute high-quality masks and to mandate them in indoor spaces. We need to think about boosters and scale up vaccine acquisition so that both first- and second-dose vaccination and boosting can continue in parallel. We must be nimble and transparent with data so that cluster infections, breakthroughs, and new variants can be identified quickly and shared.
The following curve reminds us that we never know—sitting in the valley of the curve—when the next tidal wave is going to crush us. In the long run, my prediction is that we likely will have to go through cycles of boosters every six months until the world gets vaccinated and boosted.
Endemic equilibrium is not in sight until global vaccine equity becomes a core value everyone assents to. Till then, may science, solidarity, and truth prevail in the face of adversarial uncertainty and political pandemonium.
About the Author
Bhramar Mukherjee is John D. Kalbfleisch Collegiate professor and chair of Biostatistics and professor of Epidemiology and Global Public Health at the University of Michigan School of Public Health. She is also research professor and core faculty member at the Michigan Institute of Data Science (MIDAS). Mukherjee serves as associate director for Quantitative Data Sciences in the Rogel Cancer Center and is associate workgroup director for Cohort Development at MichiganPrecision Health.
Mukherjee’s global work focuses on applying biostatistical methods to reduce health disparities and help serve vulnerable populations, specifically women’s health. Her interests in global health equity align with the Center themes of empowering women and communities and providing technical solutions. A central theme in her research has been to develop inferential methods for epidemiological data by using Bayesian, frequentist, and hybrid methods. Her collaborative interests focus on genetic and environmental epidemiology.
Mukherjee and her team have been modeling the SARS-CoV-2 virus trajectory in India for the last one year which has been covered by major media outlets like Reuters, BBC, NPR, The New York Times, The Wall Street Journal, Der Spiegel, Australian National Radio, The Times of India, and The Washington Post.