A Vanderbilt study published in the American Journal of Epidemiology unveils new methods for determining safe and unsafe doses for opioid medications. The research, led by Andrew Spieker, PhD, Associate Professor in the Department of Biostatistics, and researchers in the Department of Health Policy at Vanderbilt, used simulation studies and real-world insurance claims data to develop and test new flexible modeling approaches for medication safety studies.
These new methods aimed to resolve existing inaccuracies and assumptions of current models that can lead to biased findings, which may mistakenly label low doses of certain drugs safe or unsafe.
“Existing approaches may erroneously assume a ‘safe’ dose, as well as fail to account for the behavioral impact of being prescribed or filling a drug which could decrease or increase an individual’s risk of certain health outcomes,” said Dr. Spieker.
Previous research from Vanderbilt researchers has documented the high risks of death and developing opioid-use disorder associated with opioid prescriptions of any amount, and even higher risks for certain populations like postpartum mothers who had never been prescribed opioids before birth.
The Vanderbilt team’s novel flexible modeling methods are designed to avoid assumptions band inaccuracies built into certain models used frequently by researchers that there is a “safe” dosage, and offer a potentially more reliable alternative. The research also demonstrated the use of these methods in a case study examining the initial opioid dose prescribed after vaginal delivery.
The case study found that among pregnant patients enrolled in Tennessee Medicaid (2004-2014), women filling an initial opioid prescription less than 100 MME did not receive higher opioid doses during the rest of the postpartum period compared to those that did not fill an initial opioid prescription. However, those filling an initial opioid prescription greater than 100 MME did fill higher subsequent opioid doses than those that did not fill an initial opioid prescription.
“Future studies should consider these novel approaches, especially when examining medications for which low doses are rare or unavailable,” said Andrew Wiese, PhD, Assistant Professor in the Department of Health Policy and co-author of the study.
“These approaches will improve the research we do that aims to inform health care professionals and patients regarding safe and effective prescription drug use,” added Dr. Wiese.
The Vanderbilt team of investigators also included Margaret Adgent, PhD, MSPH, Sarah Osmundson, MD, MS, Sharon Phillips, MSPH, Ed Mitchel, Jr., MS, Ashley Leech, PhD, and Carlos Grijalva, MD, MPH.
The study was funded by grants from the National Institutes of Health (R21HD104983, 5K12HD043483, K01DA051683, K23DA047476).