While Chapter 2 discussed preparedness efforts and planning more broadly over the past 2 decades and throughout the COVID-19 pandemic, key facets of those plans and ongoing response efforts include disease surveillance and the ability to procure the right amounts of medical countermeasures and other medical supplies. This chapter highlights the current state and future directions for both global and domestic disease surveillance and discusses various areas for improvements to the medical countermeasure stockpiles and medical supply chains more broadly ahead of the next emergency.
Vivian Singletary, director of the Public Health Informatics Institute, facilitated a discussion with speakers around the current state of surveillance, how things have changed since the onset of the COVID-19 pandemic, key challenges in making improvements for the future, and what elements of a better system are necessary to inform a stronger response in future emergencies.
Janet Hamilton, executive director at the Council of State and Territorial Epidemiologists, said, “We can probably all remember where we were during those initial case investigations in December 2019 when COVID-19 began emerging.” The U.S. infrastructure is set up to investigate small
numbers of cases well, but it became quickly overwhelmed by the volume of tests and cases in March 2020. Although she acknowledged the number of investments that had been made to try and move data more rapidly across laboratory and surveillance systems, this capacity varied at the state level, with some states enabling sharing of data across systems better than others, resulting in gaps in actionable information for the federal government on how to manage a response. Electronic laboratory reporting had also made numerous strides as the pandemic went on, she said, and great progress had been made on case-based surveillance systems and thinking about the needs for different locations.
Hamilton went back to the state of things in December 2019, reminding people that public health departments were likely going to find out about new cases of suspected COVID-19 from a physician phone call. But what happens after that is important. Early in the response, health departments were very closely assessing who could even get a test, she recounted. Unless they fit the right profile, most sick individuals were not offered tests, as they had to be processed in Atlanta, so health departments were quickly overwhelmed by having to triage a person with an interesting history or symptoms and encountering a bottleneck of getting them the test and having it processed. From there, she asked, “How do we triage the volumes of calls and volumes of unusual information coming in via phone calls and faxes?” Large numbers of laboratories processing tests were not connected online to deliver information to health departments at scale, so all the information had to be shared individually with health departments. Even once we had that information, she said, we were not set up to process the volume of data and turn it around and deliver actionable information to decision makers. It is easy to talk about data like a monolith, said Hamilton, but data mean many things to many different people. She suggested looking at what questions need to be asked and then determining whether systems are collecting the information in the right way to deliver answers to those questions.
Building on Hamilton’s comments, Eric Blank, senior advisor for Public Health Systems at the Association of Public Health Laboratories (APHL), said one of the things that struck him in the early days was that the United States does have some strong surveillance systems, but they are very siloed. In the last few emergencies in the United States, no one was looking for Zika or coronaviruses, he pointed out. So, while there were some surveillance elements in place, they were not broad enough to catch what was emerging. Once surveillance elements were in place, said Blank, we were so focused on what was happening in China that we did not think about what could have been coming from Europe. Blank also noted that the available systems were not used very wisely. While the infrastructure is not ready to predict the next emerging threat, Blank said there is at least a framework
developed at this point to broaden the ability to see what might be coming in the future.
Caitlin Rivers, senior scholar at the Johns Hopkins Center for Health Security and associate director at the Center for Forecasting and Outbreak Analytics, Centers for Disease Control and Prevention (CDC), said, “Looking across the last 2 years of the pandemic, we see a lot of highs and lows with surveillance.” There has been immense progress across the board, she said, but there is no single area where the United States excels. However, she was impressed with the speed of detection of the initial cluster in Wuhan. Shortly after detection, the genome was published online for the world, said Rivers. Detection in other countries varied, she noted, as in the United States many struggled to understand where the disease was circulating, but there have been enormous advances in tracking cases from hospitalizations to deaths. Rivers highlighted three specific innovations in detection:
Rivers commented on how the importance of high-quality, timely public health data is more widely recognized now than at the start of the pandemic. She argued that there is an opportunity to capitalize on the current situation and use it to build a future where people are paying attention to the public health status in their area and can use that information to make decisions about their lives. Many public health agencies are using data infrastructure that is decades out of date. Data modernization initiatives are in further development, Rivers noted, but continued sustained investment is necessary to keep these systems operational and ready to respond to future events.
Blank added that for the first time genomic sequencing has come to play a significant role in surveillance and clinical care. APHL manages the APHL Information Messaging System (AIMS) platform, which allows anyone performing a COVID test to report results at the jurisdictional and federal levels. In this sense, Blank noted, APHL is at the forefront of data
modernization, both in trying to connect with CDC and by expanding the movement of information. Hamilton said, “The next 5 to 10 years can be transformational, but we need concrete and strong leadership to advocate for systemic data modernization and bring health care and public health together to move data across systems.”
It is painful as an epidemiologist, she said, when you are lacking data for decision making, but you know it exists and would help you get answers. She called out the immense investment of $40 billion in health care for electronic health records (EHRs), but so far just $1 billion has been set aside for a national public health system delivering data and updating systems.
Rivers said one of the main concerns right now is that many of the data streams that the United States relies on at every level to manage the pandemic are tied to the public health emergency declaration. When that emergency declaration ends, she explained, the visibility at the federal level of many of those data streams—including hospitalizations and lab reporting—can be lost. Blank added that whatever system is built moving forward needs to be sustainable. We have learned on the lab side that we can build a nice system today, he said, but if we do not have the money to manage it and sustain it, we will be talking again in 2 years to wonder what was accomplished. Hamilton said, “The challenge ahead is that public health, health care, and the Centers for Medicare & Medicaid Services (CMS) have not been convened to identify the questions that need to be answered or to identify how to effectively leverage collective abilities to ensure that information is able to move into and across the public health system.”
We can learn a lot of lessons from COVID-19, added Blank, but there are other emerging threats, beyond a transmissible respiratory virus, that should be incorporated into future discussions and thinking.
Rivers agreed with other speakers about the importance of identifying the right questions and the right data needed. She called for a community-wide discussion to outline what the exact elements would include. In terms of non-negotiables, she said surveillance needs to be electronic, cross-sectional, and delivered in a timely fashion to those who need it. Blank suggested standardization as a necessary critical effort. Our current system is split by disease agent, said Blank, but we need to think about using standardized data sets and language to transmit information, even standardizing
how we upload sequencing data. Currently, each country does it in their own way, he said, which makes sharing and comparing across countries difficult. Hamilton added that there are some guiding principles that should be employed, agreeing that surveillance needs to be electronic, but she also suggested examining approaches from an enterprise level, and prioritizing interoperability.
Another opportunity Hamilton suggested is bringing together health care and public health sectors. There are five key pillars from a public health surveillance perspective, she said, so we can answer questions not just about the most severe infections but also about the continuum of infections. She also called for investment in a nationally notifiable disease system, starting at the state and local level. “We need automated, electronic case reporting,” she argued, “where it comes directly from the EHR to the public health department, with no action needed by the clinician.” The effective delivery of lab data could also be much improved, she noted. Currently, when the order is placed at the lab, it does not supply the full level of information (e.g., demographic indicators) that public health needs when it gets the results. She acknowledged small improvements in information that goes with the results, but there are still major gaps. Finally, she advocated for investment in an electronic vital records system nationwide, as many states and localities are still using paper or systems that are not connected. All of these components are foundational for us to ensure that we are not missing things, said Hamilton.
Rivers shared some innovations from their colleagues in the United Kingdom, highlighting specifically their use of the REal-time Assessment of Community Transmission (REACT) survey, where study coordinators mailed out test kits to random households throughout the country, asking residents to swab themselves regardless of symptoms and mail it back (Riley et al., 2020). This is one way that could help supplement a narrower case-based surveillance system, she said, and help us understand what is happening in communities. Hamilton added that a number of health departments have moved into a new space and are obtaining electronic case reporting for COVID-19 and other conditions under public health surveillance from health care. Looking forward, she saw electronic reporting as a critical component. For example, Idaho has been able to identify many cases of multisystem inflammatory syndrome in children by using electronic reporting. Many of these cases are seen across the border in Utah, so without this infrastructure it would be difficult to track them. She also emphasized the importance of funding the core and foundation of data and surveillance infrastructure in conjunction with innovation. Finally, Blank said he was
impressed with South Korea’s overall approach to the pandemic. South Korea had an integrated approach across the board, from testing to case identification to mitigation strategies. But to support all of those things, the sector will still rely on strong data and information systems.
John Wiesman of the University of North Carolina at Chapel Hill said another essential policy needed is a unique patient identifier tied to a database where one can find basic information about a specific patient, so the same details and preexisting conditions do not need to be collected every time they have a visit. This would streamline much of patient care, especially during a response, he noted, but he asked if it was realistic. Hamilton replied that this has been talked about for years, but now is the time it is really needed. She noted that the workshop presenters have not yet discussed long COVID, but it is a condition that will be a huge issue when considering burden and disability going forward. Basic patient information and an aggregated patient record with all of their previous visits would be really important to have in a centralized database. She agreed with Wiesman that this is absolutely a recommendation that requires action.
Jeff Duchin for Public Health–Seattle & King County noted that long COVID may represent a long-term burden for many people, and asked how best to think about what information is needed to capture this burden. Blank reiterated some of their conversation around collecting appropriate information and moving it well within a system. Rivers added a caution that when thinking about the future of public health surveillance, people need to ensure that systems are not overfocused on elements related to COVID-19, or it would be easy to miss other issues that might arise from different diseases, such as birth defects from Zika.
Andrew Pavia, Infectious Disease Society of America, asked, “How do we solve the problem of scalability? For the last three infectious disease emergencies, CDC has not been able to scale up and get diagnostics quickly enough. What is the plan for the next time around?”
Blank said it has been an ongoing effort to do that with CDC, and there has been recent progress on how scaling was accomplished in that initial verification and validation stage when trying to understand what threats are emerging. The other aspect to consider is the handoff, he said. This went from zero to 60 with COVID-19, but with Zika the process was much different. When thinking about a plan like that, added Hamilton, there needs to be focus on not just getting a test result but also how those data flow back into the public health system. So, while we supported developing partnerships with academia and the private sector to scale efforts and be
able to process more tests, she said their results and data do not necessarily flow back to the public health department.
Regarding electronic case reporting, Pavia said there have been some successes but they were in spite of, not because of, the EHRs. “Are EHR companies any closer to developing standardized scalable ways of reporting? And how are we going to collect accurate data to allow us to do modeling with at home testing?” he asked. Hamilton replied that Utah has been at the forefront of electronic reporting and has had good engagement with EHR vendors to elevate this issue. We are on the forefront of incentive payments with CMS to help this, she said, but we are still missing certification standards to be able to truly deliver on that ability. Regarding modeling, Rivers added that it has gotten wobbly with people shifting to more usage of home testing but noted the potential for triangulating with waste-water testing and hospitalizations to build a better picture for surveillance and decision making. But moving beyond this crisis, she believes the future is in random sampling, so that is an area her department is working on.
In response to a question about the partnerships and institutions that need investment now to address the next pandemic, Rivers highlighted the new WHO Hub for Pandemic and Epidemic Intelligence as an emerging key player. It is in part focused on data and analytics and is part of a type of network emerging internationally. Hamilton said that the United States is even struggling domestically, as this area is very complex and difficult to execute. Looking at the International Health Regulations from 2005, she saw a need for a convening to specifically look at those key actors and players. Currently there is not a good way to gather case data from other countries.
Lastly, Stephen Kutz, representing tribes in Washington State, said that tribes are concerned about data sovereignty in these conversations. How can health emergency responders ensure the right data are collected but privacy and sovereignty are protected, he asked. Hamilton noted that this is an area that has not been addressed well, nor at the level of granularity and detail that tribes may need to address issues. This is another space where a lot of growth is needed, she admitted.
Before facilitating the panel, Monique K. Mansoura, executive director for Global Health Security and Biotechnology at the MITRE Corporation, provided a framing for the session. She first highlighted the responsibility of those working in the medical countermeasures enterprise to develop and deliver on measures such as vaccines and treatments, which have been necessary to respond to the COVID-19 pandemic. She also noted that there
is a broad array of threats for which medical countermeasures need to be designed, developed, and delivered, emphasizing that discussions on medical countermeasures will have to be very particular about what threats are being discussed. Mansoura also highlighted the framing of “investing” and not “funding” medical countermeasures to reduce risks. This section summarizes speakers’ reflections and discussions on medical countermeasures, including supply chain considerations, effective medical countermeasures delivery, multiagency coordination and governance of the Public Health Emergency Medical Countermeasures Enterprise (PHEMCE), and needs for supply chain resilience.
Pinar Keskinocak, William W. George Chair and professor at the H. Milton Stewart School of Industrial and Systems Engineering at Georgia Tech, outlined a typical supply chain for most products, spanning several components and including many raw materials and locations. Supply chains of most medical products are global and complex, she stated, and the goal is to deliver the right product to the right customer at the right time—trying to align supply and demand. To achieve this goal during normal times, the focus would typically be on efficiency and minimizing costs, which typically requires high utilization of capacity and labor. And while this highly optimized, lean supply chain approach looks good in normal times, when unexpected events happen (e.g., a demand shift, a reduction in supply, or a failure in coordination) the limited resilience would lead to a mismatch between demand and supply. To increase responsiveness, she explained, supply chains need to have more visibility and flexibility, on both the supply and demand sides. For example, visibility on the demand side requires surveillance mechanisms as well as testing and other ways to assess the current state, to make projections, and to plan for the future. When there is a huge demand surge for certain products (e.g., during pandemics), it is important to take certain actions, including nonpharmaceutical public health interventions, to reduce demand until supply ramps up.
Most medical and pharmaceutical supply chains do not have the characteristics of visibility, resilience, or flexibility, and there is a very limited buffer in the system for high levels of uncertainty, explained Keskinocak. Raw materials might be coming from all over the world, and visibility for that is often quite low. While onshoring has been proposed as a potential solution to help address this, Keskinocak explained that it would likely reduce flexibility, create more issues, and increase costs. She argued for a more hybrid approach. There are also additional challenges during emergencies, such as the lack of guidelines on allocating scarce resources and last-mile challenges in distribution.
The more complex the supply chain, the more potential for failure without proper measures to ensure resilience. She shared a few recommendations from another recent National Academies study on building resilience into nations’ medical product supply chains, spanning awareness, mitigation, preparedness, and response (NASEM, 2022). The report called for increased visibility not just for the main product but also for ancillary supplies and other components that may affect the product downstream. She highlighted the needs for capacity buffering, which can help certain products where capacity is a cost-effective complement to stockpiling and can help increase flexibility, as well as investing in flexible capacity to produce different products. Having a portfolio of vaccines, including some that might have a lower efficacy but might be cheaper and easier to produce and distribute (i.e., broader reach), can help slow down or stop the spread of disease, especially when the scale is so vast (Kim et al., 2021, 2022). In summary, designing and operating flexible medical supply chain demands investment, Keskinocak stated, but the financial and human costs of not making these essential investments are even higher.
Jarrod Goentzel, founder and director of the Massachusetts Institute of Technology Humanitarian Supply Chain Lab, explained that his center has been around for more than 50 years. The Humanitarian Supply Chain Lab works closely with the private sector and the Federal Emergency Management Agency (FEMA). He shared another relevant study from the National Academies studying supply chain strengths and vulnerabilities in areas affected by 2017 hurricanes that provided recommendations for key commodities distribution and restoration of utilities. The study released a report, Strengthening Post-Hurricane Supply Chain Resilience, in 2020 (NASEM, 2020). He shared the four primary recommendations from the report for achieving supply chain resilience:
Focusing on COVID-19, he shared some observations of the personal protective equipment (PPE) supply chain adaptation. There was a lot of adaptation of manufacturing capacity, he said, as many pivoted their operations to produce needed materials in response to COVID-19. Specifically, the Manufacturing Emergency Response Team in Massachusetts learned great lessons on how the government can be a catalyst for adaptation. The private sector also initiated some adaptations, and Goentzel gave examples of the Boston-based athletic apparel company New Balance using its assets to make masks, as well as other retailers working with manufacturers and suppliers to convert their routine production to PPE for health workers (Sheffi, 2020).
There were also new manufacturers that emerged in response to the massive demand for masks and other PPE, but once Chinese companies began exporting masks below cost, the customers for American manufacturers disappeared, as they were no match for low-price foreign products, he said, resulting in most of those businesses folding. Most success in production capacity was using adjacent production capacity (i.e., those that do not usually serve the medical sector) to meet the exponential growth in demand, as efforts to add nearshore capacity may not be sustainable.
Looking at inventory and the role of supply chain in shaping demand, he noted that there were hospital stockpiles and national stockpiles, but the working capital to maintain these stockpiles was not sustainable. Hospitals are instead considering direct relationships with producers and having more of a vendor management approach. This idea to go directly to suppliers and not rely on distributors is potentially a direction forward. Finally, he said crisis standards of care were implemented to mitigate the exponential growth in demand.
At the stockpile and demand management level, he shared a study supported by the Administration for Strategic Preparedness and Response (ASPR) that his team conducted from July 2020 to July 2021 to determine appropriate PPE stockpiles for a state-level public health agency. He outlined the basic framework starting with demand planning (see Figure 3-1). The ability to turn an epidemiological model into a PPE consumption model was an important step they developed, Goentzel shared. They wanted to see what the internal capacity was to provide supply. Beyond that was residual demand, where larger facilities might need additional supply to support satellite locations.
His team developed a simulation taking a collection of epidemiological models that could translate information to create a predicted consumption of PPE that remains open source. They also talked with facilities, including hospitals, long-term care centers, and behavioral health facilities, to get a better understanding of the supply-and-demand constraints. Supply plans varied by facility type, he noted, whereas skilled nursing facilities had low supplies prior to COVID, and hospitals and emergency medical services had
much more on hand. If we had come into the pandemic with the April 2021 stockpile levels, Goentzel said, it would have given facilities much more time to ramp up to the adaptive capacity needed. Essentially, the larger the PPE stockpile, the later stockouts begin, allowing more time to place replenishment orders, he explained. At the same time, there was the effect of strong supplier relationships, whereby if facilities had better supplier allocation, better relationships with suppliers for vendor management or procurement, then needs would not be as great, as facilities could source directly from supplies, he said. Goentzel explained,
In the case where a higher level of supplier engagement is combined with the stockpiles levels equivalent to after COVID began, there are scenarios where even if we had the same COVID epidemic curve, we would not need a central or a state-level stockpile to be relied upon because there would be enough capacity between supplier relationships and facility stockpiles to have met the surge.
Finally, his team also looked at demand levers in different aspects of care. Some levers were fairly intuitive, such as reusing masks to reduce demand; however, others were more surprising. For example, decreasing patient diagnostic test turnaround time from 2 days to 1 day decreases the need for N95 mask use in skilled nursing facilities by 22 percent. Cohorting can also decrease N95 mask use by as much as 95 percent in skilled nursing centers. These analyses were compiled to produce a preparedness checklist (see Box 3-1).
Based on ongoing work with FEMA and private-sector companies, Goentzel’s team implemented some of the recommendations for freight systems affected by the pandemic. In this process they were able to develop, distribute, and receive peer reviews from experts to validate and generate new hypotheses about what is going on and how things could be improved. The government could use this type of process more broadly, to work with the private sector, Goentzel explained, and more cooperative problem solving ahead of crises would build relationships. He outlined three emerging areas in the future of health emergency supply chains:
Gigi Gronvall, senior scholar at the Johns Hopkins Center for Health Security, highlighted a recent National Academies study that she was involved in, covering the PHEMCE (NASEM, 2021). The purpose of PHEMCE, Gronvall stated, is to be the nation’s coordinating body across multiple government agencies involved in the medical countermeasures development process. PHEMCE is an interagency effort to advance national preparedness and medical countermeasure coordination, she explained, that has had a troubled history and mixed track record, so the National Academies report included recommendations on how to rebuild PHEMCE for a successful future.
One of the main recommendations from the report that is germane to this workshop, said Gronvall, is that the mandate for PHEMCE needs to be reaffirmed, especially given the varied health threats facing the nation. Similarly, the committee also recommended establishing an advisory committee of nonfederal and private-sector partners and stakeholders to ensure transparency and garner the right expertise. “I cannot overstate the importance of making sure this work is done with the least restrictions possible from a security perspective,” she noted. She recounted the stories of staff in New York City hospitals receiving ventilators that they were not properly trained on, and hospital administrators often did not know what type of ventilator they were receiving until it arrived. All of this is unnecessary, she said, and counter to having a well-prepared public health workforce.
The U.S. health care system was also unprepared for the high demand for diagnostic tests, Gronvall continued, saying that her team started providing as much information as they could find on availability of tests. “But at this point in our technological evolution, people expect to have information about their own bodies, whether from a Fitbit, flu test, or COVID-19 test,” Gronvall noted. This is the direction that society is moving, and public health needs to prepare for it, she said. Lastly, Gronvall mentioned that people often forget about the need for validation for diagnostic tests. Right
now, it is the Wild West with serologic tests hitting the market, Gronvall noted, and several governments are spending millions of dollars on tests that were not even giving reliable results.
A participant asked how to consider transparency in the Strategic National Stockpile and what the role should be among industry stakeholders in contributing to this transparency. Gronvall replied that you need to do a little digging to know what goes into the stockpile, but it is possible to find statements from private companies that announce when the government has purchased large orders of anthrax or smallpox vaccine, which demonstrates that this type of information is already available and does not seem to be a security risk, she explained. Goentzel said people should not just consider the Strategic National Stockpile, as all available stockpiles are part of the buffer to buy time, so increasing transparency and coordination across stockpiles and the actors involved helps to understand the level of supply involved. Increasing transparency and coordination among stockpiles would help to align the planning processes and build them more efficiently.
Keskinocak added that the stockpile is only one component of preparedness capabilities in meeting the demand for a particular product. Discussing visibility, we need to understand what our capability is, she said. For many medical products, there really is not good visibility at the moment. There is also a need to maintain the stockpile and ensure that there is fuel, or that there are other supplies to run the products and devices, or that materials are not expired. Mansoura commented that without this transparency and visibility it is difficult to know what and where the gaps are. Details matter, she said. If we are talking about PPE it can go from individual to community to hospital levels. She advocated for looking beyond just measures and checking assumptions before moving forward and asking such questions as, do we have deployment logistics arranged? Will supplies be equitably allocated? She suggested that the challenge right now is to think about what was missed. That question needs to be addressed now so things can be changed before the next emergency, Mansoura noted.
Dan Hanfling, vice president at In-Q-Tel, asked panelists how disease surveillance can be integrated with calibrating the supply chain. “How do we bring those communities closer together to make better decisions?” he asked. Keskinocak noted that typical supply chains are built on the assumption that there is demand out there; it might be shaped, but mostly the goal is to meet the existing demand to maximize profits. But in a pandemic, what we do and how we allocate resources have a huge effect on what we see in terms of demand and what is done going forward, she said. Most
of the context of endogeneity is not incorporated in basic supply chain practice but is very important, Keskinocak added. When there is a disaster, we just assume that we will come in and save the day by distributing what is needed, but she argued for expanding the view beyond the simple preparedness-and-response mentality. This needs to start at local and state levels, and it needs engagement from the private sector, she argued, and it needs to start with basic access to health care.
Goentzel added that many companies are unable to track or visualize their supply chains. Relying on quantitative data and clear signals for surveillance may not be as viable for public health and health care, especially during pandemic, but people can still be monitoring and providing insight to see how the system is changing and adapting, he noted. His team in particular has been trying to facilitate a group of stakeholders and operators across the supply chain to collectively share up-to-date information about what is happening on the ground. Goentzel stated that there are also modeling opportunities to better understand disease trajectory, supply chain resilience, and the effect of delivering medical countermeasures on epidemiological models and vice versa.
Gronvall noted that this is another reason why the National Academies’ committee recommended an advisory board for the PHEMCE. There are too many varieties of emergencies that governments might face, that require shared discussions in an environment with some transparency to the responder community, she explained. Shared discussions allow insight into why decisions were made, as there is no way to prepare fully for every type of emergency or disaster. Mansoura added that the challenge of what is the right target and wrong investment is a difficult one. We paid for PPE and did not use it for years until we did, she said, commenting on the investments made over the years since the inception of the Office of the Assistant Secretary for Preparedness and Response (ASPR, now the Administration for Strategic Preparedness and Response) and the Biomedical Advanced Research and Development Authority (BARDA) in the late 2000s. This is a challenge for those of us responsible for preparedness, she said, with obligations to engage the entire user system and help communicate that the assumptions underlying drivers for decision making were always estimates. It will take courage and leadership when you are the defender of an investment in preparedness, Mansoura argued, or supply chain resilience when the investment is being made among budgetary restrictions and geopolitical considerations. She advocated for more visibility, communication, evidence-based decision making, and courage from leaders to say that this is what we are doing and why. This process should build credibility and trust among the public.
This page intentionally left blank.