Summary

The next administration should institute a national White House Youth Innovation Showcase similar to the discontinued White House Science Fair to promote and provide new opportunities for increased K–12 participation in science, technology, engineering, and math (STEM). As a national platform to amplify and inspire scientific accomplishments by students of all backgrounds, the Showcase will help the next administration strengthen the U.S. STEM talent pipeline and pave the way for future growth in American science and technology industries. The Showcase will also provide an opportunity for the White House Office of Science and Technology Policy (OSTP) to facilitate public-private collaborations that provide resources for participating students and support regional initiatives to increase diversity in STEM fields. The next administration can use the announcement of the Showcase to reveal its STEM agenda, outlining its policies to support STEM education and a diverse STEM workforce while articulating how its STEM goals will support emerging technology industries and overall economic development.

With about $130 billion USD, the United States leads the world in federal research and development (R&D) spending. Most of this spending is distributed by science and technology agencies that use internal reviews to identify the best proposals submitted in response to competitive funding opportunities. As stewards of quality scientific research, part of each funding agency’s mission is to ensure fairness, transparency, and integrity in the proposal-review process. The selection process is a crucial aspect of ensuring that federal dollars are invested in quality research.

Manual proposal review is time-consuming and expensive, costing an estimated $2,000–$10,000 per proposal. This equates to an estimated $300 million spent annually on proposal review at the National Science Foundation alone. Yet at current proposal-success rates (between 5% and 20% for most funding opportunities), a substantial fraction of proposals reviewed are simply not competitive. We propose leveraging machine learning to accelerate the agency-review process without a loss in the quality of proposals selected and funded. By helping filter out noncompetitive proposals early in the review process, machine learning could allow substantial financial and personnel resources to be repurposed for more valuable applications. Importantly, machine learning would not be used to evaluate scientific merit—it would only eliminate the poor or incomplete proposals that are immediately and unanimously rejected by manual reviewers.

The next administration should initiate and execute a pilot program that uses machine learning to triage scientific proposals. To demonstrate the reliability of a machine-learning-based approach, the pilot should be carried out in parallel with (and compared to) the traditional method of proposal selection. Following successful pilot implementation, the next administration should convene experts in machine learning and proposal review from funding agencies, universities, foundations, and grant offices for a day-long workshop to discuss how to scale the pilot across agencies. Our vision is that machine-learning will ultimately become a standard component of proposal review across science and technology agencies, and improving the efficiency of the funding process without compromising the quality of funded research.

Challenge and Opportunity

Allocating research funding is expensive, time-consuming, and inefficient for all stakeholders (funding agencies, proposers, reviewers, and universities). The actual cost of reviewing proposals (including employee salaries and administrative expenses) has never been published by any federal funding agency. Based on our experience with the process, we estimate the cost to be between $2,000 and $10,000 per proposal, with the variation reflecting the wide range of proposals across programs and agencies. For the National Science Foundation (NSF), which reviews around 50,000 research proposals each year¹, this equates to an average of $300 million spent annually on proposal review.

Multiple issues beyond cost plague the proposal-review process. These include the following:

1. Decreasing proposal-success rates. This decline is attributable to a combination of an increase in the number of science, technology, engineering, and math (STEM) graduates in the United States3 and an increase in the size of average federal STEM funding awards (from $110,000 to about $130,000 in less than 10 years). Current success rates are low enough that the costs of applying for federal funding opportunities (i.e., from time spent on unsuccessful proposals) may outweigh the benefits (i.e., funding received for successful proposals).
2. Difficulties recruiting qualified reviewers.
3. Delayed decisions. For example, NSF takes more than six months to reach a funding decision for about 30% of proposals reviewed.
4. Increasing numbers of identical re-submissions. With proposal-success rates as low as 5%, the results of selection processes are often seen as representing “the luck of the draw” rather than reflective of fundamental proposal merit. Hence there is a growing tendency for principal investigators (PIs) to simply re-submit the same proposal year after year rather than invest the time to prepare new or updated proposals.

There is a consensus is that the current state of proposal review is unsustainable.Most proposed solutions to problems summarized above are “outside” solutions involving either expanding available research funding or placing restrictions on the numbers of proposals submitted that may be submitted (by a PI or an institution). Neither option is attractive. Partisanship combined with the financial implications of COVID-19 render the possibility of an increased budget for S&T funding agencies vanishingly small. Restrictions on submissions are generally resented by scientists as a “penalty on excellence”. Incorporating machine learning could improve the efficiency and effectiveness of proposal review at little cost and without limiting submissions.

Incorporating machine learning would also align with multiple federal and agency objectives. On January 4, 2011, President Obama signed the GPRA Modernization Act of 2010. One of the purposes of the GPRA Modernization Act was to “lead to more effective management of government agencies at a reduced cost”. One of NSF’s Evaluation and Assessment Capability (EAC) goals established in response to that directive is to “create innovative approaches to assessing and improving program investment performance”. Indeed, two of the four key areas identified in NSF’s most recent Strategic Plan (2018) are to “make information technology work for us” and “Streamlining, standardizing and simplifying programs and processes.” In addition, NSF recognized the importance of reviewing its processes for efficiency and effectiveness in light of OMB memo M-17-26.9 NSF’s Strategic Plan includes a strong commitment to “work internally and with the Office of Management and Budget and other science agencies to find opportunities to reduce administrative burden.“ These principles are also mentioned in NSF’s 2021 budget request to Congress, as a part of Strategic Goals (e.g., “Enhance NSF’s performance of its mission”) and Strategic Objectives (e.g., “Continually improve agency operations”). Finally, longterm goal outlined in the Strategic Plan is reducing the so-called “dwell time” for research proposals—i.e., the time between when a proposal is submitted and a funding decision is issued.

Incorporating machine learning into proposal review would facilitate progress towards each of these goals. Using machine learning to limit the number of proposals subjected to manual review is a prime example of “making information technology work for us” and would certainly help streamline, standardize, and simplify proposal review. Limiting the number of proposals subjected to manual review would also reduce administrative burden as well as dwell time. In addition, money saved from using machine learning to weed out non-competitive proposals can be used to fund additional competitive proposals, thereby increasing return on investment (ROI) in research-funding programs. Additional benefits include an improved workload for expert reviewers—who will be able to focus on reviewing the scientific merit of competitive proposals instead of wasting time on non-competitive proposals—as well as the establishment of a strong disincentive for PIs to resubmit identical proposals years after years. The latter outcome in particular is expected to improve proposal quality in the long run.

Plan of Action

We propose the following steps to implement and test a machine-learning approach to proposal review:

1. Initiate and execute a pilot program that uses machine learning to triage scientific proposals. To demonstrate the reliability of a machine-learning-based approach, the pilot should be carried out in parallel with (and compared to) the traditional method of proposal selection. The pilot would be deemed successful if the machine-learning algorithm was able to reliably identify proposals ranked poorly by human reviewers, and/or proposals rejected unanimously by review panels. NSF—particularly the agency’s Science of Science and Innovation Policy (SciSIP) Program—would be a natural home for such a pilot.
2. Showcase pilot results. Following a successful pilot, the next administration should convene experts in machine learning and proposal review from funding agencies, universities, foundations, and grant offices for a day-long workshop. The workshop would showcase pilot results and provide an opportunity for attendees to discuss how to scale the pilot across agencies.
3. Scale pilot across federal government. We envision machine learning ultimately becoming a standard component of proposal review across science and technology agencies, improving the efficiency of the funding process without compromising the quality of funded research.

Reducing the numbers of scientific proposals handled by experts without jeopardizing the quality of science funded benefits everyone—high-quality proposals receive support, expert reviewers don’t waste time on non-competitive proposals, and the money saved on manual proposal review can be reallocated to fund additional proposals. Using machine learning to “triage” large submission pools is a promising strategy for achieving such objectives. Preliminary compliance checks are already almost fully automated. Machine learning would simply extend the automation stage one step further. We expect that initial costs of developing appropriate machine-learning algorithms and testing algorithms in pilots would ultimately be justified by greater long-run ROI in research-funding programs. We envision a pilot that could benefit the government but also foundations that are increasingly shouldering research funding. Ideally the pilot would be experimented in two different set-ups: a government funding agency and a Foundation.

Summary

By bringing wireless communications technology to cars and trucks, we could prevent hundreds of thousands of car crashes every year, saving many lives. We could also reduce commute times, fuel consumption, air pollution and greenhouse gas emissions, and the cost of mobile Internet access. In the longer term, deployment of connected vehicle technology can lay groundwork for better autonomous (self-driving) vehicles. In 2021, the Federal Government should establish a task force to develop a coherent vision through open and inclusive processes, and provide leadership to achieve that vision.

Note: As a working paper, this draft is still under development. The author invites your feedback and comments, which can be sent to info@dayoneproject.org.

Breastfeeding can provide important health and financial benefits for new families. But insufficient healthcare coverage, underlying medical conditions, and economic obstacles can make breastfeeding difficult or impossible for many parents. In this memo, a three-pronged approach is proposed—facilitated by an interagency collaboration through the National Advisory Council on Maternal, Infant, and Fetal Nutrition—to transform infant nutrition. First, to increase breastfeeding rates in the United States, the Centers for Medicare & Medicaid Services (CMS) should alter reimbursement policy by reimbursing tele-lactation and nutrition support for all babies covered under Medicaid. Second, the government should partner with the private sector to launch a “Synthesizing Human Milk Grand Innovation Challenge” to catalyze new extramural R&D and innovation efforts to accelerate commercialization of breast-milk alternatives for those that cannot breastfeed. And finally, the government should enact paid parental leave policies to give parents financial flexibility and dedicated time after birth to breastfeed.

Challenge and Opportunity

To ensure that all babies begin their lives on equal footing, swift action should be taken to give as many babies as possible access to breastmilk and high-quality breastmilk alternatives. Though breastfeeding and breastmilk represent only 0.04% of the National Institute of Health (NIH) budget, access to breastmilk and infant nutrition are issues that affect the health and finances of all American families with very young children. For babies, access to breastmilk has been shown to protect against respiratory illnesses, ear infections, gastrointestinal diseases, eczema, and sudden infant death syndrome. For mothers, breastfeeding may help reduce postpartum blood loss and may lower risk of post-partum depression, Type 2 diabetes, rheumatoid arthritis, cardiovascular disease, breast cancer, and ovarian cancer. The U.S. Department of Agriculture (USDA) Economic Research Service has estimated that Medicaid would save at least $172.6 million every year if breastfeeding rates among women, infants, and children increased to medically recommended levels.¹ More broadly, one study highlighted by the American College of Obstetricians and Gynecologists (ACOG) estimated that increasing breastfeeding rates could save $3.6 billion annually in the costs of treating some childhood illnesses.²

While breastfeeding can provide important health and financial benefits for new families, not all babies can breastfeed. 1 in 8 mothers in the United States face lactation dysfunction, which means that they cannot produce enough breastmilk to provide sufficient infant nutrition.³ Medical conditions such as Insufficient Glandular Tissue (IGT), mastitis, postpartum depression and anxiety (PPD/A), and infant birth defects—to name just a few—present challenges to breastfeeding. Adoptive parents can only breastfeed in certain circumstances, and birth mothers may be confused about whether they can breastfeed while on certain medications, may dislike the process of breastfeeding, or face difficulty breastfeeding while transitioning back to work.

For these and other reasons, 75% of babies use infant formula instead of breastmilk to some extent by the time they are 6 months old. A 2007 report from the Department of Health and Human Services (HHS) Agency for Healthcare Research and Quality (AHRQ) found that existing formula-feeding solutions are associated with higher risks for chronic diseases including Type 2 diabetes, asthma, and childhood obesity.⁴ Formula feeding is also linked with higher rates of necrotizing enterocolitis (NEC) for premature infants. More research is needed to understand the underlying biochemical mechanisms of human breastmilk to develop infant formulas that better mimic breastmilk. In addition, infant formula is a major expense for the federal government. Infant formula is the single most expensive item that the federal Special Supplemental Nutrition Program for Woman, Infants, and Children (WIC) provides, and the program spends more on formula than any other food—a total of $927 million in FY 2010.

It is also important to note that paid parental leave is a critical part of the postnatal experience for mothers and babies. Increases in paid parental leave are consistently associated with better infant and child health, particularly in terms of lower infant mortality rates.⁵ Paid parental leave also gives parents the opportunity and flexibility to focus on breastfeeding, which can be extremely time-consuming. The children of educated, well-off mothers are more likely to breastfeed because they have access to paid parental leave, careers with access to breaks for breast pumping, and disposable income to hire support such as night nurses. However, according to a national survey of employers conducted by the Bureau of Labor Statistics (BLS), only 18% of private industry U.S. employees had access to paid family leave through their employers. Paid parental leave in the private sector is voluntary and more prevalent among managerial and professional occupations.

Plan of Action 

CMS, USDA, NIH, state WIC agencies, and the private sector should work together through the National Advisory Council on Maternal, Infant, and Fetal Nutrition to transform U.S. infant nutrition for the better. The following specific actions are recommended:

First, to increase breastfeeding rates in the United States, CMS should alter its reimbursement policy to reimburse bi-weekly tele-lactation and nutrition support appointments for any baby covered under Medicaid during the baby’s first three months of life. Currently, the Affordable Care Act requires private insurance plans and Medicaid expansion programs to cover maternity care—including prenatal screenings and lactation consultations—without cost sharing by the patient. But there is no federal requirement to reimburse for telemedicine. Advocates should encourage the Center for Consumer Information and Insurance Oversight (CCIIO) at CMS to expand mandatory maternal-health coverage to include telehealth and for CMS to implement this policy change. This can be done in collaboration with WIC, which already provides breastfeeding support through state agencies.

Second, the federal government should catalyze new R&D and innovation efforts to accelerate commercialization of high-quality breastmilk alternatives such as

• Organizing a national “Synthesizing Human Milk R&D Summit.” This Summit would bring together formula makers, academic researchers, clinicians, parent and infant advocacy groups, representatives of the public-health community, and government stakeholders. Government stakeholders include NIH, CMS, the Centers for Disease Control and Prevention (CDC), the U.S. Food and Drug Administration [FDA], the U.S. Surgeon General, and the U.S. Preventive Services Task Force (USPSTF). The goals for the event include (i) gathering input and commitments from stakeholders to lay the groundwork for a “Synthesizing Human Milk Grand Challenge” (see next bullet); (ii) identifying barriers to developing an infant formula that more closely mimics breast milk; and (iii) generating a white paper to summarize the current understanding of the underlying biochemical mechanisms of human milk.
• Launching a “Synthesizing Human Milk Grand Challenge.” Based on the insights gained by the Synthesizing Human Milk R&D Summit, NIH should launch a “Synthesizing Human Milk Grand Challenge” that awards cash prizes for innovations in development of human breastmilk alternatives. Industry stakeholders should be recruited to match NIH investments in the prize amounts.
• Dedicating extramural academic research funds from NIH as well as funds from the Small Business Innovation Research (SBIR) program to stimulate commercialization of high-quality breastmilk alternatives. These funds would help scale up innovations from the “Synthesizing Human Milk Grand Challenge” and would motivate additional R&D in academic settings to bring infant formula closer to human milk. In addition, a “synthesizing human milk” budget set-aside subpriority should be established as part of the SBIR NIH/NICHD program.
• Only purchasing validated infant formulas with federal funds. WIC should only buy infant formula solutions that demonstrate specific outcomes as outlined by an interagency committee convened by the USDA through the National Advisory Council on Maternal, Infant, and Fetal Nutrition. The committee should include representatives from NIH, CMS, CDC, USDA, FDA, USPSTF, and the American Academy of Pediatrics (AAP).

Third, in the longer term, the federal government should enact paid parental-leave policies that give parents financial flexibility and dedicated time after birth to breastfeed.

Summary

With 1,255 VA medical facilities serving over 9 million veterans each year, the VA — through its Veterans Health Administration — maintains the largest integrated healthcare system in the United States. The VA is a national leader in delivering quality health services and driving innovation in high-priority healthcare issues such as telehealth, precision medicine, suicide prevention, and opioid safety. Yet the VA remains an under-appreciated and underutilized health policy stakeholder, involved in minimal interactions with other federal health agencies and exerting limited influence on the private healthcare system. This is a mistake. The VA is a robust healthcare provider with innovative clinical and operational practices that should be firmly entrenched in the national health policy conversation.

As a remedy, we propose strategically coordinating and consolidating the healthcare innovation, demonstration, and implementation capacities of the VA and HHS in order to ensure care of the highest possible quality across urgent issues. Elevating the VA as a major healthcare policy stakeholder will demonstrate the value of government-run healthcare, promote best practices for building an effective and forward-thinking healthcare system, and advance the VA’s “fourth mission” of supporting national preparedness.

Summary

Pending bipartisan “Cures 2.0” legislation is intended to safely and efficiently modernize healthcare delivery in the wake of the novel coronavirus (COVID-19) pandemic. Such modernization is contingent on access to high-quality data to power innovation and guided decision-making. Yet over 80% of Americans feel that the potential risks of companies collecting their data outweigh the benefits. To ensure the success of Cures 2.0, provisions must be added that bolster public trust in how health data are used.

Addressing the largely unregulated activities of data brokers — businesses that collect, sell, and/or license brokered personal information — offers a budget-neutral solution to the public’s crisis of faith in privacy. Building a well-governed health-data ecosystem that the public can trust is essential to improving healthcare in the United States.

Summary

Fragmented federal program structures and laws create enormous barriers to effective coordination across government agencies and levels of government. The next administration can advance the nation’s health and economic well-being and improve the effectiveness of taxpayer investments by creating the enabling conditions for federal, state, and local decision-makers and managers to adopt modern data analytics tools and practices.

Note: As a working paper, this draft is still under development. The author invites your feedback and comments, which can be sent to info@dayoneproject.org.

Summary

The Biden-Harris Administration should aim to make the United States a world leader in privacy-preserving machine learning (PPML), a collection of new artificial intelligence (AI) techniques capable of providing the benefits of machine learning while minimizing data-privacy concerns. By some estimates, improvements to the speed, accuracy, and scale of AI could augment global GDP by 14%, or $15.7 trillion, by 2030. Yet Americans fear that expansion of AI will have moderate to severe negative consequences. They are particularly concerned about the privacy implications of how companies and agencies use personal data to generate new developments. To assuage these concerns, this proposal recommends targeted initiatives for the Biden-Harris Administration to bring PPML techniques to maturity, including

1. Investing in PPML research and development.
2. Identifying compelling opportunities to apply PPML techniques at the federal level.
3. Creating frameworks and technical standards to facilitate wider deployment of PPML techniques.

Summary

In the 20th century, the costly nature of surveillance made it easier to maintain constitutional guarantees protecting U.S. persons from mass surveillance. In the 21st century, digitization of our everyday lives and communications has sharply reduced surveillance costs—and indeed, changed the nature of surveillance itself.  The core responsibility of any President is to “preserve, protect and defend the Constitution,” but recently unsealed federal court rulings show that intelligence agencies such as the Federal Bureau of Investigation (FBI) and the National Security Agency (NSA) are routinely accessing the digital communications of U.S. persons and otherwise using digital surveillance in ways that violate Americans’ Fourth Amendment rights against “unreasonable searches and seizures.” To fulfill their oath of office, the next president should take concrete steps to reform federal operations with respect to digital surveillance. This is important not only for protecting basic American rights, but also for diplomatic relations with key foreign allies.  Instituting meaningful protections against government surveillance in the United States would have the significant diplomatic benefit of helping reestablish the credibility of American calls for other countries to adhere to high human-rights standards.

Summary

The organ-donation crisis is one of the most persistent, expensive, and yet solvable public-health challenges of our time. As of January 2020, nearly 115,000 Americans were waitlisted for an organ transplant.  The vast majority have kidney failure, which, as one of the rare conditions qualifying patients for Medicare, imposes billions of dollars of costs on taxpayers. In 2016 alone, taxpayers spent an alarming $113 billion on kidney disease — more than the entire budgets of the National Institutes of Health ($39 billion), the Department of Homeland Security ($44 billion), and the National Aeronautics and Space Administration (NASA, $21.5 billion) combined. The clear solution is to shorten the organ waiting list. For every Medicare patient who receives a kidney transplant, taxpayers save $250,000 in avoided dialysis costs.  This proposal presents a discrete set of actions for the federal government to take to quickly and decisively to address the organ-donation crisis.

Summary

The next Administration should accelerate federal basic and applied research investments over a period of five years to return funding to its historical average as a share of GDP.  While this ambitious yet achievable strategy should encompass the entire research portfolio, it should particularly seek to reverse the long-term erosion of collective investments in physical and computer science, mathematics, and engineering to lay the foundation for economic competitiveness deep into the 21st century. This proposal outlines a strategy and series of steps for the federal government to take to reinvigorate U.S. competitiveness by restoring research and development investments. 

Summary

State and local governments are not taking full advantage of data and technology innovation that could help address key priorities such as delivery of local public services, management and design of the built environment, and fulfillment of climate goals. Supporting innovation across these domains is difficult for state and local governments due to limited technical staff, procurement challenges, and poor incentives and mechanisms to develop and scale creative solutions. Civic research is a collaborative process for addressing public priorities and improving communities by connecting technical experts to policymakers and civic partners, creating a platform for evidence-based, research-informed action. This process relies on partnerships among universities, state and local agencies, and community organizations, and has proven successful in communities nationwide. This paper recommends seven actions the next administration can take to advance civic research nationwide.