A new study  says that students die at rates much lower than same-aged people who are not in college, but the methodology is likely to have some health professionals questioning the results.
By directly asking 1,154 public and private four-year institutions how many students died (and how) from Aug. 1, 2009 to May 31, 2010, researchers found that suicide rates were 40 percent lower and alcohol-related deaths 60 to 76 percent lower (vehicular or otherwise, respectively) than the same-age general population. They also found that homicide rates were 97 percent lower.
Nobody collects nationwide data on the number and cause of student deaths, but college officials generally look to federal offices that predict mortality rates for people of all ages, including estimates for 18- to 24-year-olds, the same group examined in the new study. Those estimates come from literature and researchers at the Centers for Disease Control and Prevention and the National Institute on Alcohol Abuse and Alcoholism, which determines them based on the percentage of the population that falls into separate age groups. The results of the new study, its authors say, suggest that colleges provide especially protective environments and are doing a good job on health education.
"We have these very comprehensive and elaborate safety nets," said James C. Turner, the study's lead author and executive director of the University of Virginia student health center. "I think that our mortality study reaffirms that those kinds of things are working, and I think is a strong statement about the public health value afforded by college health services, counseling services, our university security, and the various other student affairs support services."
That may be so, says Anita Barkin, president of the American College Health Association and student health service director at Carnegie Mellon University. But, as Turner acknowledges in the study, participation was voluntary and colleges don't have a standardized methodology for collecting this data. Barkin said that can result in botched numbers or no numbers at all.
"Any time you have a difference in the findings you need to do more research to find out which data is closer to the reality of the situation. But it has implications, certainly, for allocations of resources and where we focus our efforts in terms of educational intervention and health and counseling support for college students," Barkin said. "If indeed there's lower mortality rates for suicide, homicide and alcohol-related incidents than the same-age general population, if his numbers hold up based on further study and examination, then what I would say is, this provides support for the type of public health model that college health is based on."
But Turner, who is also a past president of ACHA, disputed that the methodology was flawed.
"Our low results really call into question [the NIAAA's] methodology," Turner said. He also pointed to four other studies from sources including the Department of Education that found rates of suicide, cancer and homicide deaths similar to Turner's .
"The Commonwealth of Virginia was in a tizzy over five or six alcohol deaths in one year a few years ago, but [the NIAAA’s] data would suggest we should be having 35 to 40 every year," Turner said. "I appreciate the comment about methodology, but I think these results are so unexpected, that folks have cognitive dissonance and want to find ways of questioning the results. Remember, no one has done this type of study in 72 years. All the other studies by [the NIAAA], which estimated college alcohol deaths, were nothing more than multiplying the total number of alcohol-related accident deaths in 18- to 24-year-olds by the percent of the population that are college students."
Turner determined that the highest rate of student deaths, at 6.88 per 100,000, comes from vehicular injuries, with slightly more than half being unrelated to alcohol. 6.17 per 100,000 die from suicide, and 3.88 per 100,000 die from non-traffic injuries, about two-thirds of them non-alcohol related. Three per 100,000 die from unknown causes, 1.94 per 100,000 die from cancer, and 0.53 per 100,000 from homicide, Turner found.
But he agrees that more research is necessary, which is why he is using a CDC grant to establish a network of colleges that will report data, including deaths, into an aggregated system.
"It is quite clear that extrapolating college-student deaths by assuming that their mortality rates are the exact same as the same-age population is not accurate, and we need further research to go directly to the schools for information," Turner said. "I would hope that our findings would motivate researchers to work with schools directly to aggregate information of profound importance."