The Doctor’s Advocate | First Quarter 2017

San Diego Death Diaries

Roneet Lev, MD

Our thanks to Dr. Roneet Lev for her insights into the prescription opioid abuse epidemic. Dr. Lev is the chief of the Emergency Department at Scripps Mercy Emergency Department in San Diego. She chairs the Prescription Drug Abuse Medical Task Force and the Emergency Medicine Oversight Commission for the San Diego County Medical Society.

—David B. Troxel, MD, Medical Director, Board of Governors, and
Robin Diamond, JD, RN, Senior Vice President,
Department of Patient Safety and Risk Management

Does your name show up on a Prescription Drug Monitoring Program (PDMP) report of someone who died from a medication you prescribed? In San Diego, 713 well-meaning physicians wrote prescriptions for patients who died from accidental prescription drug overdose in 2013.

The San Diego Medical Examiner’s Office analyzed a 12-month PDMP report based on the date of death of 254 people who died accidentally from prescription overdose. These PDMP reports read as “death diaries,” chronicling stories of multiple prescribers, drug interactions, escalating dosages, and doctor shopping.

There were also patients who “double dipped,” seeking medications from two providers and two pharmacies on a regular basis. Other patients received their monthly clonazepam prescription only to die from oxycodone that was never prescribed—and with no clonazepam in their autopsy toxicology report.

Unfortunately, some medications we prescribe with good intentions end up causing harm. In the United States, 129 people die each day from an accidental drug overdose—78 of those deaths are from opioids. As physicians, we are often familiar with our patient satisfaction scores, but chances are we have no idea about our prescription death scores.

Physicians are also victims of the opioid epidemic. We have been pushed into prescribing by legislation, government regulation, and bad education. Twenty years ago, we were taught that legitimate pain patients do not get addicted, but now we know that 100 percent of chronic pain patients have dependence. We were told to use methadone as a long-acting agent for chronic pain, and now we are told not to use methadone as a first-line agent. It’s hard to keep up, and it’s difficult to reeducate a medical community of good physicians who may not realize the harm resulting from their prescriptions.

In our practice, we come face to face with patients who are angry with us about medications they demand and we won’t prescribe. We must also be aware of the thousands of Americans who are furious with us for the prescriptions we give, because they have lost loved ones from those prescriptions. We have a tough job. We get sued if we give prescriptions, and we get sued if we don’t give them. I call the key to prescribing the “Goldilocks effect”—not too much, not too little, but just right.

San Diego, which represents approximately 1 percent of the United States population, gives a good picture of prescribing patterns. The death diary data taught us several lessons.


  • The majority of prescriptions that resulted in death can be traced to use in the PDMP system. Of the 254 people who died, 80 percent received prescriptions from the PDMP system.

The Prescriptions

  • Prescription drug deaths cross all ethnicities, socioeconomic classes, genders, and ages. The average death was a male age 46.5.
  • Opioids alone are not the biggest problem. The 4,366 prescriptions given before death included 33 medications dispensed at 275 pharmacies. The majority, 80 percent, of people died of a “cocktail” of several medications.
  • Methadone was the prescription most likely associated with death from a single drug. Methadone was most likely to be obtained outside the PDMP system, with only seven of 46 deaths found on the PDMP. Methadone clinics are not allowed to enter data in the PDMP system; however, we urge clinics to check the PDMP system and to obtain regular EKGs for patients to monitor drug safety.
  • Opioids and benzodiazepines are known to have additive central nervous system depression. The death diaries showed that over 50 percent of patients who died were prescribed this combination, and over 20 percent of toxicology reports included this combination. These statistics explain why there is an FDA black box warning for this combination.
  • Chronic use was found to be a major risk factor for death. Of all the people who died, 68 percent were prescribed the same medication for three or more months. This accounted for 96 percent of all the prescriptions.

Doctor Shopping

  • Patients who met the definition of “doctor shoppers” (28 percent) accounted for the majority of the prescriptions (51 percent). But what is a doctor shopper? For research purposes, a doctor shopper is defined as a patient who sought four prescriptions from four pharmacies in 12 months. It does not include the “double dippers” (patients who sought the same prescription from two doctors every month). Clinically, doctor shopping can be considered anything outside the gold standard, that prescribing for controlled medications should be coordinated by one doctor and one pharmacy.

Following Directions

  • Do patients die from prescriptions if they follow a doctor’s directions? Very few. Of the 254 people who died, only 42 (16.5 percent) died of the same medication that was prescribed within two months of death. These patients were not using alcohol or illicit drugs and were not doctor shoppers; however, one can question whether the majority of the prescriptions resulted in benefit rather than harm.

The Providers

  • Of the 713 prescribers, 54 percent were in primary care and included 6.3 percent physician assistants, 22 percent internal medicine, and various internal medicine subspecialties. The remainder of the prescribers were emergency physicians (20 percent), psychiatrists (11 percent), surgeons (8 percent), dentists (4 percent), and pain specialists (3 percent). (See Figure 1.)
  • Primary care physicians provided the majority of the prescriptions, both in terms of the number of prescriptions written and the number of prescriptions that resulted in deaths. They were followed by psychiatrists, surgeons, and emergency physicians.
  • The average number of pills per prescription varied by medical specialty. Orthopedic surgeons averaged 189 pills per prescription, followed by all surgeons (123 pills), pain specialists (97 pills), primary care physicians (79 pills), psychiatrists (58 pills), emergency physicians (23 pills), and dentists (19 pills). (See Figure 2.)

Figure 1Figure 2

Helpful Tips

The Centers for Disease Control and Prevention (CDC) published 12 evidence-based prescribing tips for chronic pain. The tips are now the gold standard for prescribing. They include the famous line, “Start low and go slow,” and we can add, “Don't start without a plan to stop.”

Saying no is much harder than saying yes. But in reviewing your patient’s medication list, you may be saving a life. We have published guidelines on how to say no nicely.

Physicians who need assistance with a specific patient can use the Clinical Consultation Center at the University of California, San Francisco, and speak to a physician expert. The system is similar to the HIV prophylaxis hotline that many professionals used. (See the Additional Resources section for contact information and links to referenced guidelines.)

In addition to the CDC guidelines, the following tips are useful:

Rule number one is don’t prejudge patients as being pain medication seeking when there is a potential for missed diagnosis. Work up each chief complaint as you normally would without jumping to conclusions.

The gold standard is for patients to use one provider and one pharmacy for all chronic medications. We should not be practicing in silos, with primary care, psychiatry, and specialists not coordinating medications.

The emergency department should not be used as a referral for patients who run out of medications or extra shots. Many emergency departments have established guidelines that refer patients back to their physicians for all chronic medications.

The PDMP makes you a better doctor by telling you more than doctor shopping information. It gives you medication names and dosages and the names of the doctors prescribing them. Can you imagine giving an antibiotic without checking for allergies? Before writing a prescription that can result in death, take one to two minutes to check that your patient has no drug interactions or co-prescribing. If your patient is already receiving opioids from another provider for a different diagnosis, you do not need to prescribe more.

Medication agreements should be used for patients who need more than three months of a controlled medication. This can include opioids, benzodiazepines, and stimulants. Don’t misinterpret a medication agreement as a free ticket for prescription. You must still make sure that benefits outweigh risks each time you write a prescription. The death diaries show that this is the population at risk.

Discharge instructions should be clear. Warn patients not to drive when taking opioids, sleep aids, or anything that causes them to not be fully alert. Warn patients to keep medications secure.

Addiction should be treated with compassion, like other medical illnesses. About 1 percent of the population suffers from addiction, and only 10 percent get appropriate treatment. There is often a genetic association, and it is useful to ask about a family history. Learn your community resources for addiction. Addiction referral can be helpful for patients you are weaning from multiple medications. Don’t simply discharge noncompliant patients; give them referrals for addiction treatment.

Specific Medication Tips

Ibuprofen and acetaminophen are usually more effective than oral opioids. A 2013 study on patients who had third molar dental extractions reaffirmed this claim.

Opioids, if needed, should be “start low and go slow,” according to the CDC. In addition, don't start without a plan to stop. There should be a time-out if you are escalating dosages with no functional improvements. Avoid going over 90 morphine equivalents.

Opioids and benzodiazepines are a dangerous combination; the death statistics speak for themselves. Unfortunately, many patients have already been on this combination for many years. Providers can do two things:

  1. Do not start this combination for a new generation of patients with chronic pain and anxiety.
  2. Wean the medications to the lowest possible dosage, ideally choosing the importance of one or the other drug category. The Veterans Health Administration was able to take on this project, creating a computer alert that prevents this prescription combination.

Carisoprodol (Soma): Don’t prescribe carisoprodol. This drug is sold as a muscle relaxant and provides minimal muscle relaxant effect as it quickly metabolizes into meprobamate, a strong, addictive tranquilizer that is not sold in the U.S. and is illegal in Europe. Thirty patients who died were given this medication.

Alprazolam (Xanax): Alprazolam is the most prescribed and most addictive benzodiazepine. It is associated with many prescription deaths. The medication peaks at one to two hours but lasts only five hours. Tolerance and psychological and physical dependence may occur in as little as 10 days. Bipolar and mania are relatively contraindicated for Xanax because of the exacerbation of mania. The American Psychiatric Association guidelines recommend that Xanax be limited to short-term use only.

Methadone: Methadone should be used with only the most compliant patients with strong consideration for genetic testing to determine if they can metabolize the drug to avoid cardiac toxicity. Of the deaths from methadone prescriptions, 100 percent of the prescriptions were given by primary care practitioners.

Tramadol (Ultram): We are seeing increased prescriptions of tramadol and, therefore, increased deaths (19 in 2013). Ultram is falsely advertised as non-narcotic and nonaddictive. This is not true. Ultram, a synthetic opioid, is addicting. One tablet of 50 mg of tramadol has more morphine equivalents than one 5/235 tablet of Norco. Tramadol is addicting and must be used with caution in patients who have seizure disorders.

Zolpidem (Ambien): Sleep aids should be used on a temporary basis, not for daily infinite use. Ambien is the number one drug of abuse among addicted physicians under treatment at the Betty Ford Center. The death diaries included 43 patients who were given Ambien.

Cough medication with codeine: This is a medication of abuse. I tell patients who have a cough that I will be happy to prescribe an inhaler to get the cough out, rather than a syrup that will keep the cough in.

Marijuana: Marijuana, a central nervous system depressant, should be used with the same caution as other medications. In patients with chronic abdominal pain and cyclic vomiting syndrome caused by cannabinoid, avoid treatment with opioids that will result in an opioid addiction.

Naloxone: The CDC recommends prescribing naloxone, the reversal agent, for anyone who is on more than 50 morphine milligrams per day. Anyone with a history of accidental overdose should also receive the prescription. Naloxone is a carve-out medication for many Medicaid patients, and some states allow pharmacies to dispense it without a prescription.


Lev R, Petro S, Lee O, et al. A description of Medical Examiner prescription-related deaths and prescription drug monitoring program data. Am J Emerg Med. 2016 Mar;34(3):510-4.

Lev R, Lee O, Petro S, et al. Who is prescribing controlled medications to patients who die of prescription drug abuse? Am J Emerg Med. 2016 Jan;34(1):30-5.

Lev R, Petro S, Lee A, et al. Methadone related deaths compared to all prescription related deaths. Forensic Sci Int. 2015 Dec;257:347-52.

Moore PA, Hersh EV. Combining ibuprofen and acetaminophen for acute pain management after third-molar dental extractions: translating clinical research to dental practice. J Am Dent Assoc. 2013 Aug;144(8):898-908.

Donald Teater; National Safety Council. Evidence for the efficacy of pain medications. Published October 6, 2014.

Additional Resources

The Doctor’s Advocate is published by The Doctors Company to advise and inform its members about loss prevention and insurance issues.

The guidelines suggested in this newsletter are not rules, do not constitute legal advice, and do not ensure a successful outcome. They attempt to define principles of practice for providing appropriate care. The principles are not inclusive of all proper methods of care nor exclusive of other methods reasonably directed at obtaining the same results.

The ultimate decision regarding the appropriateness of any treatment must be made by each healthcare provider considering the circumstances of the individual situation and in accordance with the laws of the jurisdiction in which the care is rendered.

The Doctor’s Advocate is published quarterly by Corporate Communications, The Doctors Company. Letters and articles, to be edited and published at the editor’s discretion, are welcome. The views expressed are those of the letter writer and do not necessarily reflect the opinion or official policy of The Doctors Company. Please sign your letters, and address them to the editor.

Choosing a malpractice insurer?
Here’s what to ask.

Selecting a medical malpractice insurer is one of the most important decisions you’ll make. It's a prime opportunity to ask: Is this the type of insurer I want protecting my reputation?