Skip to Main Content

The revelations about Dr. Carmen Puliafito are an academic hospital’s worst nightmare — methamphetamine use, partying in the dean’s office, the doctor’s presence at the scene of a 21-year-old’s overdose.

But the case of the University of Southern California eye surgeon and former medical school dean, as reported recently by the Los Angeles Times, raises a bigger question: Could it all have been prevented by random drug testing?

California voters handily rejected a ballot question in 2014 that would have required physicians to submit to testing. Meanwhile, with no state or federal regulations on the matter, hospitals nationwide have enacted an array of policies. Some only test anesthesiologists; still others do not require any drug testing; one major hospital, Cleveland Clinic, recently took the rare step of instituting random drug testing for all of its 55,000 employees.


Drug testing is common in other industries, and it is required for pilots and other transportation workers whose impairment would directly undermine public safety. But hospitals have been resistant to embrace the practice.

Their trepidation reflects various objections that have been raised in the industry. Doctors groups have expressed concerns about privacy, false positives, and a lack of evidence that such testing would actually improve the safety of patients. There is also no consensus on the best way to implement random testing of doctors or on how to discipline those with positive results.


But Dr. Michael Fitzsimons, who oversees the drug testing program at Massachusetts General Hospital, said health care organizations should consider expanding random testing, especially with many states liberalizing drug laws.

“With legalization of marijuana, I do worry that that could put our patients in more jeopardy,” Fitzsimons said. “Physicians should be more widely tested. As medicine starts to focus more and more on safety and the human factors, I think it’s valid for hospitals to consider, and I’m actually in favor of drug testing.”

Mass. General’s policy calls for testing doctors when there is a suspicion of drug use. The hospital also randomly drug tests anesthesiologists, but does not require random testing for other doctors or caregivers.

Fitzsimons, director of cardiac anesthesia at the hospital, said drug testing should be implemented in concert with other measures, including substance control in hospitals as well as education and treatment programs for doctors with substance abuse problems.

“Drug testing alone is not the answer,” he said. “All arms really need to be enhanced, addressed, and utilized.”

A University of Southern California guide on disciplinary practices indicates that the university can drug test employees to determine whether they are working under the influence. But there is no provision for random testing.  The university did not respond to a question about whether Puliafito was ever tested, or whether it will revisit its testing policy in light of his case.

In a statement, USC President C.L. Max Nikias said the university is “working to determine how we can best prevent these kinds of circumstances moving forward.” The statement added that the school launched a new office of campus wellness and crisis intervention and continues to operate a peer-to-peer program with Vanderbilt University to help physicians in distress.

The investigation by the Los Angeles Times reported that Puliafito, 66, a renowned eye surgeon and prolific university fundraiser, repeatedly used drugs with prostitutes and other criminals — sometimes in the dean’s office at USC.

The Times reported that it reviewed photos and videos that show the Harvard-educated Puliafito using methamphetamine and ecstasy in 2015 and 2016. It also reported he was present in a Pasadena hotel room when a 21-year-old woman overdosed. She was rushed to the hospital and later recovered.

Attempts to reach Puliafito were unsuccessful. Following the newspaper’s report, USC officials released a series of statements saying they were working to fire Puliafito and strip him of his faculty tenure. The university is also conducting an investigation into the extent of his drug use.

In November 2014, California voters soundly rejected Proposition 46, the ballot proposal that would have required drug testing of doctors, by a margin of 67 percent to 33 percent. While earlier polls showed strong report for drug testing of physicians, the proposal was paired with a measure that would have increased California’s cap on non-economic damages in medical negligence lawsuits.

Opponents, including the state’s medical association and other prominent groups, argued that ballot proposal was about increasing payouts to trial lawyers more than protecting patients.

Following its rejection, the proposal faded into the background. However, hospitals have continued to face pressure to address substance abuse within their own ranks and beyond, due to legalization of marijuana and a national opioid epidemic fueled by prescription drug abuse.

Cleveland Clinic, whose home state is in the throes of the opioid crisis, instituted random drug testing for all of its employees in January 2016.

Dr. Paul Terpeluk, medical director of employee health and the clinic, said the policy was put in place to send a clear message to employees. “We’re a health care organization and we consider basically every job at the Cleveland Clinic to be a safety-sensitive job,” he said. “We don’t believe people who work in those positions should be under the influence of drugs.”

Terpeluk said the random testing policy has led to the discovery of drug use by employees. But he declined to elaborate on the number of positive results or whether the implementation of the new policy has reduced the number of employees with substance abuse problems.

“We use this policy as a deterrent to drug use,” he said. “Because most people are rational, they say, ‘No, I’m not going to do that because I might get caught.’ But we do have people who turn up positive.”

If there is a positive result, Terpeluk said, the clinic’s policy is to provide treatment to the employee. “Rather than fire people, we try to help them and get them back to work as soon as possible,” he said.

If the employee with a positive result is a licensed clinician, the clinic is obligated to report positive results to the state medical authorities, who run an independent process to determine whether the caregiver should be fined or have their license suspended or revoked.

Terpeluk said the clinic abides by the state’s determinations. He added that drug testing is important both for the safety of both patients and caregivers. “Doctors and nurses, like everyone else, are not immune to this issue,” he said. “We feel good this program is in effect and helping to identify people who may have a problem.”

  • More hospitals should institute random drug testing for their employees. I personally know of several nursing staff and physicians who engage in regular drug use. When I reported them, nothing was done. They all denied drug use, continued patient care, had no drug testing. This is about patient care and safety. How many of us want to be in the care of impaired healthcare workers? Not me.

  • If i were an editor at STAT. I’d ask you to break it down/do a bit more homework Drug testing is not a test of impairment on the job. Alcohol blood levels may be fair markers for impairment-Alcohol in the blood now means Im drunk NOW. But other substances leave traces in the blood that will give a positive test and the person tested will HAVE NO impairment. MJ is the obvious drug. As states push for legalization of Mj we are going to need to work on sensible policies that distinguish use from impairment.
    The way the prevalent m0del works an employee can be fired for a pos MJ test for consuming it weeks before- but another employee who goes home and drinks a 6 pack a night would not be considered a problem. Thats not due to impairment byt to the way the tests look for signs of the substances in the body.
    This is why privacy advocates get up in arms– it makes no sense.Impairment is what matters , not who does what in their time away from the job.

  • Because it does not specify, I wonder if the random testing at Cleveland, etc. looks for legal opiates and/or alcohol levels along with illicit substances. We know there are ‘functional alcoholics’ who are mostly harming themselves and their families, but aren’t drinking before work or on the job, and are possibly viewed as high-performing employees. I suspect this is the case for marijuana users too, more so than for users of meth/cocaine/heroin, but maybe I’m wrong.

Comments are closed.