A California jury convicted Hsiu-Ying “Lisa” Tseng, DO, of second-degree murder in connection with three overdose deaths in October, making it the first time a U.S. doctor was convicted of murder for overprescribing drugs.
The jury found her guilty of 23 counts, including 19 counts of unlawful controlled substance prescription and one count of obtaining a controlled substance by fraud, related to the deaths of three young men who died from overdoses of prescription medication prescribed by Dr. Tseng between March and December 2009. Prosecutors suggested that her reckless prescription writing may have been responsible for as many as a dozen overdose deaths.
Emergency physicians who have published research on pain prescriptions in the emergency department agreed with the jury's decision. “Her prescribing was so completely egregious that I concur with the jury,” said Scott Weiner, MD, an emergency physician and the assistant director for ED operations at Brigham and Women's Hospital in Boston.
“It appears that she earned over a million dollars a year by prescribing pain medications, and had received multiple warnings from the [Medical Board of California] and family members who asked her to stop prescribing to these patients. If an emergency physician ever went that far, I would expect that he would be punished,” he said.
Dr. Tseng's defense lawyer, Tracy Green, told the media that her client was naïve and overworked, and had committed malpractice rather than murder. “It's pretty hard to figure out who is an addict and who is not,” she told the Los Angeles Times. “I don't think it bodes well for doctors in America.” (Oct. 30, 2015; http://lat.ms/1WnJRUF.)
Research on opioid prescribing that Dr. Weiner and colleagues published in the September issue of Annals of Emergency Medicine found that the median number of pills per prescription in the emergency department is 15, and the vast majority — more than 99 percent — were immediate-release formulations. “We also discovered that less than two percent of prescriptions were for more than 30 tablets, and those were for diagnoses like long bone fractures and cancer. Emergency physicians are not big prescribers of opioids, even though we're constantly told that we are,” he said. (Ann Emerg Med 2015;66:253.)
The Annals study included just one sample from 19 hospitals across the country over a single week, but Dr. Weiner noted that it is consistent with data from the Food and Drug Administration showing that only about five percent of all immediate-release opioid prescriptions come from emergency physicians. “We just presented data at ACEP showing that probably less than two percent of the morphine equivalents come from emergency medicine. That's very small compared with specialties like family practice, dentistry, orthopedics, internal medicine, and pain management.”
A case like Dr. Tseng's is very unlikely to happen in the ED, said Gabe Wilson, MD, who most recently served as the medical director of the ED at Christus Santa Rosa Hospital at Westover Hills in San Antonio. “The ED operates in an environment of much more scrutinized oversight compared with a private clinic like this one. Outlier physicians are unlikely to escape notice. Since the goal in the ED is to treat acute illness and if there is no objective etiology of the pain, warning flags will go off for most of us. There is always the option of prescribing eight or so tabs of an opioid if someone does not appear in one of our databases, just to get them through a weekend,” said Dr. Wilson, who is also a regional director for EmCare.
That said, Dr. Weiner conceded that emergency departments still struggle with teasing out drug-seeking behavior. “We do see patients with substance use disorders, and they are very hard to detect sometimes because we want to do the right thing for patients and help manage their pain,” he said.
Forty-nine U.S. states plus the District of Columbia and Guam had established Prescription Drug Monitoring Programs (PDMPs) by the end of 2015; those electronic databases collect designated data on substances dispensed in the state and distribute that data to authorized individuals, including emergency physicians. (Missouri is the lone holdout.)
“These are fantastic. I use mine at least once per shift,” said Dr. Weiner. “Whenever I consider prescribing an opioid, I can see where the patient's previous prescriptions have come from and the prescriber. If it's suspicious, I can print out the list and bring it into the room with the patient to discuss the situation.”
The PDMPs are state-based, and vary in characteristics and quality. Some have data available virtually in real time, some take a day or two to update, and others can take as long as a couple of weeks. Some PDMPs share information across state lines, and some don't.
So far, research is mixed about how much emergency physicians are using PDMP records to inform their opioid prescribing. Another study published by Dr. Weiner and colleagues in Annals of Emergency Medicine found that PDMP data influenced prescriber behavior in only 9.5 percent of cases and resulted in more prescribing. (2013;62:281.) A 2010 study, also in Annals, found PDMP use changed emergency physicians' prescribing frequency in 41 percent of cases and resulted in less prescribing. (Ann Emerg Med 2010;56:24.)
California, where Dr. Tseng practiced, has taken strong steps to control the prescription of long-acting opiates from the emergency department, according to Casey Grover, MD, an emergency physician at Community Hospital of the Monterey Peninsula, whose research indicated that classic drug-seeking behaviors, like reporting lost or stolen prescriptions or reporting “more than 10/10” pain, are relatively uncommon in the ED, making it more difficult for emergency physicians to tease out addictive behavior. (West J Emerg Med 2012;13:416.)
The California chapter of the American College of Emergency Physicians has endorsed a “prescribe safely” initiative, which includes a list of 10 safe prescribing principles that includes not refilling lost or stolen prescriptions and not prescribing long-acting pain medications such as oxycodone (OxyContin), morphine sulfate (MSContin), fentanyl (Duragesic), methadone, oxymorphone (Opana ER), hydromorphone (Exalgo), and others. (The full list is available online at http://bit.ly/1lPpW09.)
“Before we started taking action on this, San Diego County was seeing one death a day from prescription drug overdoses,” said Dr. Grover. “While drugs prescribed by emergency physicians are not causing the majority of the problem, the fact is that patients on chronic morphine, Xanax, oxycodone, and other similar powerful analgesics and sedatives are increasingly getting medications from multiple sources, and that may include the ED. That's really dangerous. We have to be cautious and wary when prescribing pain medications because of the risk of overdose.”
California is about to roll out the second iteration of its PDMP, which will include VA and military prescriptions and provide alerts to problematic usage rather than depending on the physician's initiative to access and review the record, Dr. Grover said.
Dr. Tseng's conviction probably has few implications for emergency physicians, Dr. Weiner said, adding that another legal development will: mandatory PDMP lookup. Twenty-two of the 49 states with PDMP systems now require providers to query those systems before prescribing controlled substances in most situations. A recent JAMA analysis reported substantial increases in PDMP queries and decreases in opioid prescribing after mandatory-use laws were adopted in Kentucky, New York, Tennessee, and Ohio. (2015;313:891.)
“If you're in a state with mandatory PDMP lookup and one of your patients overdoses, and they look up your records and find that you didn't access the PDMP before prescribing, you could be liable,” Dr. Weiner said.