(CNN)There’s no question that the impact of diseases varies drastically across the United States, depending on which state you live in.
Now, a study published Tuesday in the journal JAMA details just how wide those state-by-state differences are when it comes to how diseases, injuries and risk factors impact America’s youth, adults and older populations.
“There’s so many different levels of health in the US, so many different disparities across states and age groups,” said Dr. Christopher Murray, a professor of global health at the University of Washington and director of the Institute for Health Metrics and Evaluation, who was lead author of the study.
“The top five risk factors — diet, obesity, elevated blood pressure, tobacco and physical inactivity — explain an awful lot of the differences across states,” he said. “Why those causes are getting worse in some states and not getting worse in the other states, I think, deserves more investigation.”
‘Very divergent trends’
For the study, researchers analyzed data from the Global Burden of Disease study, a research tool that quantifies health impacts across countries and within the US using various data sources such as the Centers for Disease Control and Prevention, the National Health and Nutrition Examination Surveys and state inpatient databases.
Since 1990, the Global Burden of Disease study has collected and analyzed health data, with the most recent data coming from 2016. The data capture premature death and disability from more than 300 diseases and injuries in 195 countries, by age and sex.
The researchers examined the data from 1990 to 2016, taking a close look at state-by-state trends and calculating the probability of death among three age groups: 0 to 20, 20 to 55 and 55 to 90.
“It turned out that the trends in health — and in this case being measured by the chances of dying at different age groups — basically in all 50 states were improving in kids and adolescents and in people over age 55, but then in the ages between 20 and 55, you had the US going in two different directions,” Murray said.
“You had complete divergent trends in those middle age groups, and that’s sort of surprising because usually, people think whatever is driving health would affect everybody pretty much the same,” he said. “The fact that we get these very divergent trends by state and age, I think, is really quite unusual.”
In 1990, Hawaii had an estimated average of 78.5 years, followed by Utah at 77.9 years and Minnesota with 77.8 years.
Mississippi was the state with the lowest at 73.1 years, followed by Louisiana at 73.3 and South Carolina at 73.7.
Even though it’s not a state, the District of Columbia was included in the data and had the lowest of all at 68.4.
In 2016, Hawaii still had the highest life expectancy at birth, at 81.3 years. California climbed from 24th in 1990 to having the second-highest in 2016, at 80.9 years. Connecticut rose from seventh in 1990 to having the third-highest in 2016 at 80.8 years, the study showed.
Mississippi still had the lowest of all the states at 74.7 years, followed by West Virginia, which dropped from 45th in 1990 to 50th in 2016 at 75.3 years. Alabama ranked third at 75.4 years, the study showed. Meanwhile, life expectancy at birth appeared to improve in DC, as it moved from ranking lowest of all in 1990 to 36th in 2016.
Dr. Howard Koh, a professor at the Harvard T.H. Chan School of Public Health, and Dr. Anand Parekh, chief medical adviser at the Bipartisan Policy Center, co-authored an editorial published alongside the new study.
They wrote that “if each state were a country, Hawaii would rank 20th worldwide (after Ireland) whereas Mississippi would rank 76th (tied with Kuwait).”
The researchers also found that Mississippi ranked as having the highest probability of death for people between the ages of 0 and 20 between 1990 and 2016.
The probability of dying young, by state
Probability of death for this age group was interpreted as the probability in a particular state that a child would die before his or her 20th birthday. Data for the District of Columbia were not included in this analysis.
Overall, the change in probability of death from birth to 20, between 1990 and 2016, declined in all states, the study showed. The states with the most pronounced declines were South Carolina, Georgia, Alaska and New York. On the other hand, Maine had the lowest decline of probability.
The researchers pointed out that the nationwide declines were associated with improvements in neonatal disorders; other noncommunicable diseases, including congenital; and injuries, with slight increases from mental and substance use disorders.
In 2016, Mississippi, Alabama and Louisiana ranked in descending order as having the highest probability of death in that age group.
In descending order, California, New Jersey and then Massachusetts ranked at the bottom of the list with the lowest probability of death for that year.
The researchers also measured probability of death for adults between the ages of 20 and 55, interpreted as dying before age 55.
The health of young adults, by state
In descending order, West Virginia, Mississippi and then Alabama ranked as having the highest probability of death for adults 20 to 55 in 2016. In descending order, New York, California and Minnesota ranked as having the lowest in 2016.
Between 1990 and 2016, the largest reductions in probability of death were seen in New York and California, and the highest increases were in West Virginia and Oklahoma, the researchers found.
Decreases in the probability of death in the US might have been influenced by declines in the prevalence of HIV and AIDS across all states, as well as declines in road injuries and neoplasms or tumors, the researchers noted in the study.
Meanwhile, increases in the probability of death might have been influenced by a rise in the burden of drug use disorders, alcohol use disorders and chronic kidney disease, among other factors, the researchers noted.
“Mortality reversals in 21 states for adults ages 20 to 55 years are strongly linked to the burden of substance use disorders, cirrhosis, and self-harm,” the researchers wrote. “This study shows that the trends for some of these conditions differ considerably across different states.”
Cardiovascular disease is a concern in older adults
For adults 55 to 90, all states saw a considerable reduction in probabilities of death between 1990 and 2016.
The highest point decline between 1990 and 2016 was observed in California, which ranked as having the second-lowest probability of death for this age group in 2016. Hawaii had the lowest in 2016, and Florida had the third-lowest.
Hawaii was the only state in which the probability of death was less than 65% for adults 55 to 90, the study showed.
In descending order, Mississippi had the highest probability of death for that age group in 2016, and West Virginia had the second-highest, followed by Alabama in third. For this age group, the overall declines in probability of death were largely tied to reductions in the probability of dying from cardiovascular diseases, the researchers wrote.
A separate study, published in the journal JAMA Cardiology on Wednesday, found that the overall burden of cardiovascular disease improved for all states between 1990 and 2016 but disparities remained. In 2016, the greatest burden was concentrated in a band of states extending from the Gulf Coast to West Virginia, according to that study.
The JAMA study on life expectancy also explored causes of death for all ages nationwide.
What’s killing Americans
The study showed that, between 1990 and 2016, overall death rates in the US declined from 745.2 per 100,000 people to 578 per 100,000 people, but that doesn’t mean Americans are no longer facing many health risks.
“This study shows that high [body-mass index], smoking, and high fasting plasma glucose are the three most important risk factors in the United States, and that although smoking is decreasing, BMI and fasting plasma glucose levels are steadily increasing,” the researchers wrote.
“These two risk factors pose unique challenges in the United States given that unabated, they have the potential to change the health trajectory for individuals in many states. Levels of overweight and obesity increased during the study period,” they wrote.
Between 1990 and 2016, ischemic heart disease was the top cause of years of life lost in the US, and lung cancer ranked second, without budging, the study showed.
Additionally, “one of the things we do in the study is look at things that don’t so much kill you but lead to loss of good health, things like back pain, neck pain and things like depression,” Murray said.
Low-back pain and major depressive disorders remained the first and second causes of years lived with a disability in terms of rates between 1990 and 2016, the study showed. During that time, diabetes moved from eighth to third, and musculoskeletal disorders were in fourth.
A loss of health tied to a condition was measured in terms of disability in the study. “What’s interesting to me was the rise in low-back pain,” said Dr. Georges Benjamin, executive director of the American Public Health Association, who was not involved in the study.
“I don’t know whether that correlates in any way with the opioid epidemic, but low-back pain increased to be the third leading cause in 2016 in their study,” he said. “You wonder whether or not there is some correlation there, particularly with the rise in opioid use.”
Opioid use disorders rose from the 52nd to the 15th cause of “years of life lost” between 1990 and 2016 in the study, in terms of rates. Opioid use disorders ranked seventh in 1990 and then eighth in 2016 when it came to leading causes of nonfatal health loss.
“It doesn’t take a whole lot to notice that opioids and drugs more generally have shot up in the rankings and increased hugely over the period of this study,” Murray said.
“We did these analyses in other countries as well, and it’s very unusual to have opioids, or drugs more generally, as such a dominant cause of ill health,” he said.
How to fix America’s health problems
The findings in the study mirror what other research has shown when it comes to state-by-state disparities in health, Benjamin said.
“What’s interesting to me is that we still see that the states that tend to straddle the bottom of the rankings in all those kind of other studies are pretty much at the bottom of this study as well,” Benjamin said.
“Mississippi, Alabama, Georgia tend to still be at the bottom of the rankings — Alaska as well. Yet some of the states that tend to rank pretty high, like Hawaii, rank fairly well on just straight mortality and disability rankings,” he said.
“This demonstrates again that we have enormous disparities between states and how long you live.”
Strategies to deal with these state-by-state inequalities across the country should be threefold, the researchers wrote.
They stressed addressing modifiable risk factors such as diet and exercise; tobacco, alcohol and drug use; improving access to quality health care; and addressing the social determinants of health, such as the conditions in which people work and live.
“What do we need to focus on as we go through the future? It’s still going to need to be tobacco. It’s going to need to be getting people to ideal body weight, and nutrition plays an enormous role,” Benjamin said.
Policy makers can use the new study and Global Burden of Disease results to reconsider the current national stance toward disease prevention, Koh and Parekh wrote in their editorial.
“Long overlooked and underinvested, public health programs currently receive only an estimated 2.5% of US health care dollars, and prevention programs (broadly defined) receive only 8.6%. Earlier promise for major improvements in disease prevention initially offered by the [Affordable Care Act] has lately stalled,” Koh and Parekh wrote.
“Clinicians and policy makers can use these analyses and rankings to reexamine why so many individuals still experience preventable injury, disease, and death. Doing so could move the entire nation closer toward a United States of health,” they wrote.
Correction: An earlier version of this story included a life expectancy graphic in which the colors on the map did not accurately reflect the colors on the key.