Connect with us

The Conversation

Will AI revolutionize drug development? Researchers explain why it depends on how it’s used

Published

on

theconversation.com – Duxin Sun, Associate Dean for Research, Charles Walgreen Jr. Professor of Pharmacy and Pharmaceutical Sciences, University of Michigan – 2025-01-03 07:33:00

A high drug failure rate is more than just a pattern recognition problem.

Thom Leach/Science Photo Library via Getty Images

Duxin Sun, University of Michigan and Christian Macedonia, University of Michigan

The potential of using artificial intelligence in drug discovery and development has sparked both excitement and skepticism among scientists, investors and the general public.

“Artificial intelligence is taking over drug development,” claim some companies and researchers. Over the past few years, interest in using AI to design drugs and optimize clinical trials has driven a surge in research and investment. AI-driven platforms like AlphaFold, which won the 2024 Nobel Prize for its ability to predict the structure of proteins and design new ones, showcase AI’s potential to accelerate drug development.

AI in drug discovery is “nonsense,” warn some industry veterans. They urge that “AI’s potential to accelerate drug discovery needs a reality check,” as AI-generated drugs have yet to demonstrate an ability to address the 90% failure rate of new drugs in clinical trials. Unlike the success of AI in image analysis, its effect on drug development remains unclear.

Pharmacist searching through drawer of drug packages

Behind every drug in your pharmacy are many, many more that failed.

nortonrsx/iStock via Getty Images Plus

We have been following the use of AI in drug development in our work as a pharmaceutical scientist in both academia and the pharmaceutical industry and as a former program manager in the Defense Advanced Research Projects Agency, or DARPA. We argue that AI in drug development is not yet a game-changer, nor is it complete nonsense. AI is not a black box that can turn any idea into gold. Rather, we see it as a tool that, when used wisely and competently, could help address the root causes of drug failure and streamline the process.

Most work using AI in drug development intends to reduce the time and money it takes to bring one drug to market – currently 10 to 15 years and US$1 billion to $2 billion. But can AI truly revolutionize drug development and improve success rates?

AI in drug development

Researchers have applied AI and machine learning to every stage of the drug development process. This includes identifying targets in the body, screening potential candidates, designing drug molecules, predicting toxicity and selecting patients who might respond best to the drugs in clinical trials, among others.

Between 2010 and 2022, 20 AI-focused startups discovered 158 drug candidates, 15 of which advanced to clinical trials. Some of these drug candidates were able to complete preclinical testing in the lab and enter human trials in just 30 months, compared with the typical 3 to 6 years. This accomplishment demonstrates AI’s potential to accelerate drug development.

YouTube video
Drug development is a long and costly process.

On the other hand, while AI platforms may rapidly identify compounds that work on cells in a Petri dish or in animal models, the success of these candidates in clinical trials – where the majority of drug failures occur – remains highly uncertain.

Unlike other fields that have large, high-quality datasets available to train AI models, such as image analysis and language processing, the AI in drug development is constrained by small, low-quality datasets. It is difficult to generate drug-related datasets on cells, animals or humans for millions to billions of compounds. While AlphaFold is a breakthrough in predicting protein structures, how precise it can be for drug design remains uncertain. Minor changes to a drug’s structure can greatly affect its activity in the body and thus how effective it is in treating disease.

Survivorship bias

Like AI, past innovations in drug development like computer-aided drug design, the Human Genome Project and high-throughput screening have improved individual steps of the process in the past 40 years, yet drug failure rates haven’t improved.

Most AI researchers can tackle specific tasks in the drug development process when provided with high-quality data and particular questions to answer. But they are often unfamiliar with the full scope of drug development, reducing challenges into pattern recognition problems and refinement of individual steps of the process. Meanwhile, many scientists with expertise in drug development lack training in AI and machine learning. These communication barriers can hinder scientists from moving beyond the mechanics of current development processes and identifying the root causes of drug failures.

Current approaches to drug development, including those using AI, may have fallen into a survivorship bias trap, overly focusing on less critical aspects of the process while overlooking major problems that contribute most to failure. This is analogous to repairing damage to the wings of aircraft returning from the battle fields in World War II while neglecting the fatal vulnerabilities in engines or cockpits of the planes that never made it back. Researchers often overly focus on how to improve a drug’s individual properties rather than the root causes of failure.

Diagram of airplane with clusters of red dots on the wing tips, tail and cockpit areas

While returning planes might survive hits to the wings, those with damage to the engines or cockpits are less likely to make it back.

Martin Grandjean, McGeddon, US Air Force/Wikimedia Commons, CC BY-SA

The current drug development process operates like an assembly line, relying on a checkbox approach with extensive testing at each step of the process. While AI may be able to reduce the time and cost of the lab-based preclinical stages of this assembly line, it is unlikely to boost success rates in the more costly clinical stages that involve testing in people. The persistent 90% failure rate of drugs in clinical trials, despite 40 years of process improvements, underscores this limitation.

Addressing root causes

Drug failures in clinical trials are not solely due to how these studies are designed; selecting the wrong drug candidates to test in clinical trials is also a major factor. New AI-guided strategies could help address both of these challenges.

Currently, three interdependent factors drive most drug failures: dosage, safety and efficacy. Some drugs fail because they’re too toxic, or unsafe. Other drugs fail because they’re deemed ineffective, often because the dose can’t be increased any further without causing harm.

We and our colleagues propose a machine learning system to help select drug candidates by predicting dosage, safety and efficacy based on five previously overlooked features of drugs. Specifically, researchers could use AI models to determine how specifically and potently the drug binds to known and unknown targets, the level of these targets in the body, how concentrated the drug becomes in healthy and diseased tissues, and the drug’s structural properties.

These features of AI-generated drugs could be tested in what we call phase 0+ trials, using ultra-low doses in patients with severe and mild disease. This could help researchers identify optimal drugs while reducing the costs of the current “test-and-see” approach to clinical trials.

While AI alone might not revolutionize drug development, it can help address the root causes of why drugs fail and streamline the lengthy process to approval.The Conversation

Duxin Sun, Associate Dean for Research, Charles Walgreen Jr. Professor of Pharmacy and Pharmaceutical Sciences, University of Michigan and Christian Macedonia, Adjunct Professor in Pharmaceutical Sciences, University of Michigan

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Read More

The post Will AI revolutionize drug development? Researchers explain why it depends on how it’s used appeared first on theconversation.com

The Conversation

George Washington, a real estate investor and successful entrepreneur, knew the difference between running a business and running the government

Published

on

theconversation.com – Eliga Gould, Professor of History, University of New Hampshire – 2025-03-10 07:50:00

President George Washington delivers his first inaugural address in April 1789 in New York City.
Painting by T.H. Matteson, engraving by H.S. Sadd, via Library of Congress

Eliga Gould, University of New Hampshire

During his three presidential campaigns, Donald Trump promised to run the federal government as though it were a business. True to his word, upon retaking office, Trump put tech billionaire Elon Musk at the head of a new group in the executive branch called the Department of Government Efficiency.

DOGE, as Musk’s initiative is known, has so far fired, laid off or received resignations from tens of thousands of federal workers and says it has discovered large sums of wasted or fraudulently spent tax dollars. But even its questionable claim of saving US$65 billion is less than 1% of the $6.75 trillion the U.S. spent in the 2024 fiscal year, and a tiny fraction of the nation’s cumulative debt of $36 trillion. Because Musk’s operation has not been formalized by Congress, DOGE’s indiscriminate cuts also raise troubling constitutional questions and may be illegal.

Before they go too far trying to run the government like a business, Trump and his advisors may want to consider the very different example of the nation’s first chief executive while he was in office.

A man stands while behind him a man sits at a desk.
Elon Musk, left, and Donald Trump have undertaken an effort both describe as seeking to run government more like a business.
Andrew Harnik/Getty Images

The first businessman to become president

Like Trump, George Washington was a businessman with a large real estate portfolio. Along with property in Virginia and six other states, he had extensive claims to Indigenous land in the Ohio River Valley.

Partly because of those far-flung investments, the first president supported big transportation projects, took an active interest in the invention of the steamboat, and founded the Patowmack Company, a precursor to the builders of the Chesapeake and Ohio Canal.

Above all, Washington was a farmer. On his Mount Vernon estate, in northern Virginia, he grew tobacco and wheat and operated a gristmill. After his second term as president, he built a profitable distillery. At the time of his death, he owned nearly 8,000 acres of productive farm and woodland, almost four times his original inheritance.

Much of Washington’s wealth was based on slave labor. In his will, he freed 123 of the 300 enslaved African Americans who had made his successful business possible, but while he lived, he expected his workers to do as he said.

President Washington and Congress

If Washington the businessman and plantation owner was accustomed to being obeyed, he knew that being president was another matter.

In early 1790, near the end of his first year in office, he reflected on the difference in a letter to the English historian Catharine Macaulay. Macaulay had visited Mount Vernon several years before. She was eager to hear the president’s thoughts about what, in his reply, he described as “the last great experiment for promoting human happiness by reasonable compact.”

The new government, Washington wrote, was “a government of accommodation as well as a government of laws.”

As head of the executive branch, his own powers were limited. In the months since the inauguration, he had learned that “much was to be done by prudence, much by conciliation, much by firmness. Few, who are not philosophical Spectators,” he told his friend, “can realise the difficult and delicate part which a man in my situation (has) to act.”

Although Washington did not say why his situation was delicate, he didn’t need to. Congress, as everyone knew, was the most powerful branch of government, not the president.

The previous spring, Congress had shown just how powerful it was when it debated whether the president, who needed Senate confirmation to appoint heads of executive departments, could remove such officers without the same body’s approval. In the so-called Decision of 1789, Congress determined that the president did have that power, but only after Vice President John Adams broke the deadlock in the upper house.

The meaning of Congress’ vote was clear. On matters where the Constitution is ambiguous, Congress would decide what powers the president can legally exercise and what powers he – or, someday, she – cannot.

When it created a “sinking fund” in 1790 to manage the national debt, Congress showed just how far it could constrain presidential power.

Although the fund was part of the Treasury Department, whose secretary served at the president’s pleasure, the commission that oversaw it served for fixed terms set by Congress. The president could neither remove them nor tell them what to do.

Inefficient efficiency

William Humphrey, a member of the Federal Trade Commission, was unconstitutionally fired by Franklin Roosevelt in 1933.
Library of Congress

By limiting Washington’s power over the Sinking Fund Commission, Congress set a precedent that still holds, notably in the 1935 Supreme Court case of Humphrey’s Executor v. U.S.

To the displeasure of those, including Trump, who promote the novel “unitary executive” theory of an all-powerful president, the court ruled that President Franklin D. Roosevelt could not dismiss a member of the Federal Trade Commission before his term was up – even if, as Roosevelt said, his administration’s goals would be “carried out most effectively with personnel of my own selection.”

Like the businessman who currently occupies the White House, Washington did not always like having to share power with Congress. Its members were headstrong and independent-minded. They rarely did what they were told.

But he realized working with Congress was the only way to create a federal government that really was efficient, with each branch carrying out its defined powers, as the founders intended. Because of the Constitution’s checks and balances, the United States was – and is – a government based on compromise between the three branches. No one, not even the president, is exempt.

To his credit, Washington was quick to learn that lesson.The Conversation

Eliga Gould, Professor of History, University of New Hampshire

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Read More

The post George Washington, a real estate investor and successful entrepreneur, knew the difference between running a business and running the government appeared first on theconversation.com

Continue Reading

The Conversation

5 ways schools have shifted in 5 years since the COVID-19

Published

on

theconversation.com – Rachel Besharat Mann, Assistant Professor in Education Studies, Wesleyan University – 2025-03-10 07:48:00

Students sit in pop-up tents during wind ensemble class at Wenatchee High School on Feb. 26, 2021 in Wenatchee, Wash..
David Ryder/Getty Images

Rachel Besharat Mann, Wesleyan University and Gravity Goldberg, Wesleyan University

The U.S. educational landscape has been drastically transformed since the COVID-19 pandemic shuttered school campuses five years ago.

Access to high-quality teachers and curriculum developed by teachers is shrinking, for example. Likewise, there has been a loss of emotional support for students and a decline in the school use of technology and social media.

As education scholars focused on literacy practices in schools, here are five ways we believe the COVID-19 pandemic – and the rapid shift to remote learning and back – has transformed education:

1. Teachers are leaving, and those staying are stressed

At the start of the 2024-2025 school year, 82% of U.S. public schools had teaching vacancies.

Schools have tried to adapt by expanding class sizes and hiring substitute teachers. They have also increased use of video conferencing to Zoom teachers into classrooms.

A teacher sits at home in front of a computer monitor.
A teacher works from her home due to the COVID-19 outbreak on April 1, 2020, in Arlington, Va.
Olivier Doulier/AFP via Getty Images

Teacher retention has been a problem for at least a decade. But after the pandemic, there was an increase in the number of teachers who considered leaving the profession earlier than expected.

When teachers leave, often in the middle of the school year, it can require their colleagues to step in and cover extra classes. This means teachers who stay are overworked and possibly not teaching in their area of certification.

This, in turn, leads to burnout. It also increases the likelihood that students will not have highly qualified teachers in some hard-to-fill positions like physical science and English.

2. Increase in scripted curriculum

As of fall 2024, 40 states and Washington had passed science of reading laws, which mandate evidence-based reading instruction rooted in phonics and other foundational skills.

While the laws don’t necessarily lead to scripted curriculum, most states have chosen to mandate reading programs that require teachers to adhere to strict pacing. They also instruct teachers not to deviate from the teachers’ manual.

Many of these reading programs came under scrutiny by curricular evaluators from New York University in 2022. They found the most common elementary reading programs were culturally destructive or culturally insufficient – meaning they reinforce stereotypes and portray people of color in inferior and destructive ways that reinforce stereotypes.

This leaves teachers to try to navigate the mandated curriculum alongside the needs of their students, many of whom are culturally and linguistically diverse. They either have to ignore the mandated script or ignore their students. Neither method allows teachers to be effective.

When teachers are positioned as implementers of curriculum instead of professionals who can be trusted to make decisions, it can lead to student disengagement and a lack of student responsiveness.

This form of de-professionalization is a leading cause of teacher shortages. Teachers are most effective, research shows, when they feel a sense of agency, something that is undermined by scripted teaching.

3. Improvements in teen mental health, but there’s more to do

Many of the narratives surrounding adolescent mental health, particularly since the pandemic, paint a doomscape of mindless social media use and isolation.

However, data published in 2024 shows improvements in teen reports of persistent sadness and hopelessness. Though the trend is promising in terms of mental health, in-school incidences of violence and bullying rose in 2021-22, and many teens report feeling unsafe at school.

Other reports have shown an increase in feelings of loneliness and isolation among teens since the pandemic.

4. Crackdown on students’ technology use in schools

COVID-19 prompted schools to make an abrupt switch to educational technology, and many schools have kept many of these policies in place.

For example, Google Classroom and other learning management systems are commonly used in many schools, particularly in middle school and high school.

These platforms can help parents engage with their children’s coursework. That facilitates conversations and parental awareness.

But this reliance on screens has also come under fire for privacy issues – the sharing of personal information and sensitive photos – and increasing screen time.

And with academia’s use of technology on the rise, cellphone usage has also increased among U.S. teens, garnering support for school cellphone bans.

A child wearing a face mask looks at a laptop computer.
A student attends an online class at the Crenshaw Family YMCA on Feb. 17, 2021, in Los Angeles during the COVID-19 pandemic.
Patrick T. Fallon/AFP via Getty Images

But banning these devices in schools may not help teens, as smartphone use is nearly universal in the U.S. Teens need support from educators to support them as they learn to navigate the complex digital world safely, efficiently and with balance.

In light of data surrounding adolescent mental health and online isolation – and the potential for connection through digital spaces – it’s also important that teens are aware of positive support networks that are available online.

Though these spaces can provide social supports, it is important for teens to understand the strengths and limitations of technology and receive authentic guidance from adults that a technology ban may prohibit.

5. Students and adults need social emotional support

Students returned to in-person schooling with a mix of skill levels and with a variety of social and emotional needs.

Social and emotional learning includes self-awareness, self-management, social awareness, relational skills and decision-making.

These skills are vital for academic success and social relationships.

Teachers reported higher student needs for social and emotional learning after they returned to in-person instruction.

While some of this social and emotional teaching came under fire from lawmakers and parents, this was due to confusion about what it actually entailed. These skills do not constitute a set of values or beliefs that parents may not agree with. Rather, they allow students to self-regulate and navigate social situations by explicitly teaching students about feelings and behaviors.

A teacher and student are separated by plexiglass as they sit across from each other at a desk.
A teacher provides instruction to a student at Freedom Preparatory Academy on Feb. 10, 2021, in Provo, Utah.
George Frey/Getty Images

One area where students may need support is with cognitive flexibility, or the ability to adapt to current situations and keep an open mind. Classroom instruction that engages students in varied tasks and authentic teaching strategies rooted in real-life scenarios can strengthen this ability in students.

Besides allowing students to be engaged members of a school community, cognitive flexibility is important because it supports the skill development that is part of many state English language arts and social studies standards.

Social and emotional learning and cognitive flexibility are key components that allow students to learn.

Due to vague or confusing state policies, many schools have stopped teaching social and emotional learning skills, or minimized their use.

This, coupled with teacher stress and burnout, means that both adults and children in schools are often not getting their social and emotional needs met.

Message of mistrust

While we described five shifts since the start of the pandemic, the overall trend in K-12 schools is one of mistrust.

We feel that the message – from districts, state legislators and parents – is that teachers cannot be trusted to make choices.

This represents a massive shift. During the initial phase of the COVID-19 lockdown, teachers were revered and thanked for their service.

We believe in teacher autonomy and professionalism, and we hope this list can help Americans reflect on the direction of the past five years. If society wants a different outcome in the next five years, it starts with trust.The Conversation

Rachel Besharat Mann, Assistant Professor in Education Studies, Wesleyan University and Gravity Goldberg, Visiting Assistant Professor in Education Studies, Wesleyan University

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Read More

The post 5 ways schools have shifted in 5 years since the COVID-19 appeared first on theconversation.com

Continue Reading

The Conversation

Daylight saving time and early school start times cost billions in lost productivity and health care expenses

Published

on

theconversation.com – Joanna Fong-Isariyawongse, Associate Professor of Neurology, University of Pittsburgh – 2025-03-07 13:55:00

Daylight saving time kicks in on March 9, 2025, but some say it leads to more heart attacks, depression and car accidents.
Lord Henri Voton/E+ via Getty Images

Joanna Fong-Isariyawongse, University of Pittsburgh

Investigations into the 1986 Space Shuttle Challenger disaster revealed that key decision-makers worked on little sleep, raising concerns that fatigue impaired their judgment. Similarly, in 1989, the Exxon Valdez oil spill resulted in a massive environmental catastrophe. The official investigation revealed the third mate, in charge of steering the ship, was running on too little sleep, among other problems.

While these specific disasters were not caused by daylight saving time, they are conclusively linked to fatigue, based on postaccident investigations and reports. They underscore the well-documented dangers of sleep deprivation and fatigue-related errors. Yet a vast body of research shows that every year, the shift to daylight saving time needlessly exacerbates these risks, disrupting millions of Americans’ sleep and increasing the likelihood of accidents, health issues and fatal errors.

Imagine a world where one simple decision – keeping our clocks aligned with the natural cycle of the Sun – could save lives, prevent accidents and improve mental well-being. It’s not just about an hour of lost sleep; it’s about how small disruptions ripple through our health, our workplaces and even our children’s futures.

I’m a neurologist who specializes in sleep health. I’ve seen firsthand the negative impacts of poor sleep; it has enormous personal and economic consequences.

Yet despite overwhelming research supporting better sleep policies – such as delaying school start times to align with adolescent biology and the adoption of permanent standard time – these issues remain largely overlooked in public policy discussions.

Sleep deprivation comes with real costs

Chronic sleep deprivation does more than leave people tired. It costs an estimated US$411 billion annually in lost productivity and health care costs. Poor sleep leads to workplace mistakes, car accidents and long-term health issues that strain businesses, families and the economy as a whole.

Fortunately, there’s a fix. Smarter sleep policies – such as permanent standard time and later school start times – can boost efficiency, improve health and save lives.

In a classroom setting, students take an exam.
Sleep-deprived teens have lower test scores and graduation rates.
skynesher/E+

Up before dawn

Teenagers are the most sleep-deprived age group in the U.S. Multiple studies and surveys show that anywhere from 71% to 84% of high school students report getting insufficient sleep.

This is largely due to early school start times, which force teens to wake up before their biological clocks are ready. If you have a teenager, you probably see it every day: The teen struggling to wake up before sunrise, rushing out the door without breakfast, then waiting in the dark for the school bus.

More than 80% of public middle and high schools in the U.S. start before 8:30 a.m., with 42% starting before 8 a.m. and 10% before 7:30 a.m. As a result, some districts have bus pickups as early as 5 a.m.

Teenagers are going through a natural shift in their circadian rhythms by about two hours. This shift, driven by hormones and biology, makes it hard for them to fall asleep before around 11 p.m. The bodies of teens aren’t wired for these schedules, yet schools and society have designed a system that forces them to function at their worst.

Declining scores, drowsy driving and depression

Sleep-deprived teens have lower grades and test scores, more car crashes caused by drowsy driving, more alcohol and drug use and higher rates of depression, anxiety suicide and aggressive behavior, including carrying weapons.

Along with the health benefits, studies have found that moving school start times to 8:30 am or later could add $8.6 billion to the economy within two years, partly by increased graduation rates.

While concerns about increased transportation costs exist, such as the need for additional buses or drivers due to staggered school start times, some districts have found that optimizing bus routes can offset expenses, making the change cost-neutral or even cost-saving. For instance, a study in Boston found that reorganizing bus schedules using advanced algorithms reduced the number of buses needed and improved efficiency, which allowed high school students to start later and better align with their natural sleep cycles. This change not only supported adolescent sleep health but also saved the district $5 million annually.

YouTube video
Studies show that daylight saving time does not reduce energy use.

More heart attacks, car wrecks and suicide

Every March, most Americans shift their clocks forward for daylight saving time. Studies show this change disrupts sleep and leads to measurable adverse outcomes, including a significant increase in heart attacks. These effects linger for days after the shift, as sleep-deprived workers struggle to adjust.

The mental health impact is also severe. Suicide rates increase in the weeks following the switch, particularly for those already vulnerable to depression.

Unlike daylight saving time, standard time follows the body’s natural circadian rhythm, which is primarily regulated by exposure to sunlight. Our internal clocks are most stable when morning light exposure occurs early in the day, signaling the body to wake up and regulate key biological functions such as hormone production, alertness and metabolism. In contrast, daylight saving time artificially extends evening light, delaying the body’s release of melatonin and making it harder to fall asleep at a biologically appropriate time.

Studies have found that adopting permanent standard time could prevent up to 5,000 suicides annually by reducing seasonal depression, decrease errors, injuries and absenteeism in the workplace and make roads safer, potentially preventing 1,300 traffic deaths each year.

Times are changing

The U.S. tried permanent daylight saving time in 1974. It was so unpopular that Congress repealed it within nine months.

Russia tried it too, in 2011, but switched back three years later. The United Kingdom dropped permanent daylight saving time in 1971 after three years, and Portugal in 1996 after four. All of these countries found that the switch caused widespread public dissatisfaction, health concerns, more morning car accidents and disrupted work schedules. No country is currently on year-round daylight saving time.

These examples provide real-world evidence that permanent DST is undesirable due to public dissatisfaction, safety concerns and negative health effects – all three countries attempted it and ultimately reversed course. Since 2022, there has been renewed debate, largely driven by former U.S. Sen. Marco Rubio’s Sunshine Protection Act, which aims to make DST permanent.

However, the name is misleading because it doesn’t “protect” sunshine but rather eliminates critical morning light, which is essential for regulating circadian rhythms. Major health organizations, along with the National Safety Council, strongly oppose permanent DST due to its well-documented risks.

There are signs that suggest the U.S. is finally waking up to these problems. Out of 13,000 school districts, 1,000 have independently adopted later school start times. California and Florida have enacted laws requiring high schools to start no earlier than 8:30 a.m. California’s mandate went into effect in 2022, and Florida’s is set to begin in 2026.

Permanent standard time and later school start times are not radical ideas. They’re practical, evidence-based solutions based on human biology. Implementing these changes nationally would require congressional action. However, current federal law already allows states to adopt permanent standard time, as Arizona and Hawaii have done, setting a precedent for the rest of the country.The Conversation

Joanna Fong-Isariyawongse, Associate Professor of Neurology, University of Pittsburgh

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Read More

The post Daylight saving time and early school start times cost billions in lost productivity and health care expenses appeared first on theconversation.com

Continue Reading

Trending