Graduate-level public health programs should provide more advanced biostatistics instruction and training to keep up with the explosion of available health data and available software for analytics, according to a team of researchers led by an expert at the School of Public Health at Georgia State University.
“Education in modernized statistical methods, including advanced modeling and computationally intensive statistical techniques, is necessary for staying current and implementing new and advanced methods,” the researchers wrote.
Among the researchers’ suggestions for improving biostatistics higher education are:
- Including graduate-level instruction in classical statistics as well as logistic regression;
- Including training on modern methods for handling missing data at the graduate level and offering continuing education for current public health professionals on this topic;
- Offering training and exposure to statistical software, including SAS and/or STATA—the two most commonly used packages.
Dr. Matt Hayat
The results of the study are published in PLOS ONE in the article “Statistical methods used in the public health literature and implications for training of public health professionals.” The study’s lead author is Dr. Matthew J. Hayat, associate professor of epidemiology and biostatistics at Georgia State University.
The suggestions were based on a study examining the frequency at which advanced statistical methods were used in contemporary public health research. The study randomly sampled 216 articles published in 2013 by seven top-tier public health journals. (The selected journals American Journal of Public Health, American Journal of Preventive Medicine, International Journal of Epidemiology, European Journal of Epidemiology, Epidemiology, American Journal of Epidemiology, and Bulletin of the World Health Organization.) Reviewers read and quantified the statistical methods reported in those articles.
More than 95 percent of the articles in the study included descriptive statistics in table or graph form and more than 76 percent of the articles included conclusions based on statistical inference, according to the study. The widespread availability of high-powered computers and the rapid growth of user-friendly software packages has led to greater use of advanced statistical methods and analyses in health literature, the study noted.
“However, statistical training in public health may not have kept up with the modern data explosion and statistical complexities increasingly being applied in health studies and reported in scientific publications,” researchers said. “In order to properly and adequately train public health professionals to access scientific publications, it is essential to, at a minimum, be teaching statistical methods actually used and reported in top-tier public health journals.”
The study’s authors also included Dr. Amanda Powell, a sociology and public health graduate of Georgia State; Tessa Johnson, a public health graduate student at Georgia State; and Betsy Caldwell, a statistician with the Centers for Disease Control and Prevention.