ProPublica Investigation Finds Major Flaws in Research on Pregnant Opioid Users

A ProPublica investigation has revealed significant flaws in widely publicized research by a University of Tennessee physician who studied detoxification treatments for pregnant women with opioid addiction, prompting an institutional audit and raising serious questions about research ethics and patient protections.

Dr. Craig Towers, a high-risk obstetrician who practiced at the University of Tennessee Medical Center, published a groundbreaking study in 2020 that challenged decades of medical doctrine regarding opioid use during pregnancy. The research suggested that detoxifying pregnant patients from opioids was far less risky to fetuses than previously believed, contradicting standard practice of maintaining patients on methadone or buprenorphine throughout pregnancy.

How the research gained national attention

When Towers arrived in Knoxville in 2010, he encountered what he described as an unprecedented crisis. The region's opioid epidemic had created a surge of pregnant patients addicted to painkillers and newborns suffering from withdrawal symptoms. "We were inundated," Towers later told the Knoxville News Sentinel.

Traditional medical practice held that pregnant patients with opioid addiction could not safely undergo detoxification because withdrawal could trigger miscarriage or fetal distress. Instead, doctors prescribed methadone or buprenorphine—legal opioid replacements that controlled cravings but often caused neonatal abstinence syndrome in newborns.

Towers questioned this approach. Over several years, he collected data on more than 300 patients, most of them his own, who stopped taking opioid maintenance medications. His study, published in the American Journal of Obstetrics and Gynecology, reported promising results using naltrexone—a non-opioid medication that blocks the euphoric effects of opioids and discourages use.

The study immediately gained national recognition. CNN's Dr. Sanjay Gupta featured Towers' work, describing babies born without withdrawal symptoms as "miracles." The News Sentinel honored him as a "Health Care Hero." Clinics across Tennessee and in Florida began adopting his approach. In 2019, the University of Tennessee medical school granted him additional research funding, praising "his many research efforts and leadership that has resulted in national recognition."

Experts raise concerns

Despite the study's positive reception, addiction and pregnancy specialists grew suspicious. Naltrexone had shown inconsistent results in previous research, with nonpregnant patients experiencing relapse rates averaging 72 percent in systematic reviews. Towers' study reported far better outcomes.

Eight specialists in addiction and pregnancy medicine shared concerns with one another and eventually drafted a letter to the University of Tennessee in July 2020. Two agreed to speak with ProPublica. One was Dr. Mishka Terplan, a practicing OB-GYN board-certified in addiction medicine who had served on federal committees developing treatment guidelines.

"Dr Towers' work has been impactful, especially in TN," Terplan wrote to the university. "I am concerned that each day the article(s) is out there, potential harm to a vulnerable population accumulates through both ethically suspect clinical care and consequently misguided public health."

The specialists questioned how Towers could have achieved such dramatically different results than previous research. They also noted that the study made no mention of patients refusing to participate or dropping out—both common occurrences in substance use research—raising concerns about whether informed consent had been properly obtained.

University audit uncovers serious problems

The concerns prompted a university audit that uncovered major problems with Towers' research methodology and documentation. The audit found that Towers failed to obtain university approval for the study, provided "inaccurate and/or unverifiable, contradictory source documentation" for study components, and did not provide research databases or data analysis as requested. Auditors also could not determine the extent to which patient health information was properly protected.

Towers retired shortly after the study's publication, citing health and family reasons. He defended his work to ProPublica, stating that auditors had reached erroneous conclusions and did not give him adequate opportunity to defend his research. He pointed to the journal's continued support of the study as evidence of its validity.

The American Journal of Obstetrics and Gynecology, published by Elsevier, initially took no public action for three years. In 2023, the journal published a 78-word correction stating that the study had been improperly classified as "prospective" research—a rigorous designation for studies that follow patients in real time rather than reviewing past records.

Five experts in research ethics and substance abuse during pregnancy who reviewed the audit documents and correction told ProPublica that the journal's response was inadequate given the severity of the problems identified.

"The misrepresentation is not minor," said Eugenie Reich, an attorney who has represented whistleblowers in research fraud cases. "It runs throughout the piece."

Patient experience raises additional questions

Katie Carringer, a former patient of Towers, shared her experience with ProPublica. Carringer, who was about six weeks pregnant when she first saw Towers in 2018, had been drug-free for most of the previous two years after completing an inpatient treatment program. She sought Towers' care because she wanted a high-risk pregnancy specialist who could help manage her anxiety and depression.

At her second appointment, Carringer said she expected to discuss her mental health. Instead, Towers recommended naltrexone—a medication typically prescribed for patients battling active cravings. Carringer, who worked at a drug rehab facility, knew the medication was designed for people in early recovery, not those who had been drug-free for extended periods.

"With me being clean so long, it didn't even make sense," Carringer told ProPublica.

According to her medical records, Carringer had not used drugs during either of her previous pregnancies in 2007 and 2009. Her records showed only two brief relapses since completing treatment in 2016. Towers noted that her medical record stated she was interested in naltrexone, but Carringer denied expressing interest, saying she tried the medication only because she felt pressured. She stopped after one dose due to severe nausea.

Towers told ProPublica that Carringer was not included in his study.

Ongoing impact on clinical practice

Despite the audit findings and subsequent correction, some clinics continue to use approaches based on Towers' research. The study influenced treatment protocols in Tennessee communities including Johnson City, Jellico, and Sevierville, as well as in Daytona Beach, Florida.

The case highlights ongoing tensions in addiction treatment for pregnant women. Standard care using methadone or buprenorphine, while reducing risks of overdose and relapse, frequently results in newborns requiring treatment for withdrawal symptoms. Towers' approach promised an alternative that could spare babies this experience—but only if the research supporting it was sound.

For pregnant women with opioid use disorder, the investigation underscores the importance of understanding treatment options and asking questions about the evidence behind recommended approaches. Standard protocols involving medication-assisted treatment remain the recommended approach by major medical organizations, based on decades of research demonstrating improved outcomes for both mothers and babies.

The University of Tennessee has not publicly disclosed what actions, if any, it took in response to the audit findings beyond the journal's correction. Elsevier, the journal's publisher, stated that it "upholds the highest standards of rigor and ethics in our publishing to protect the quality and integrity of research."