The governor of Vermont, Peter Shumlin, devoted all of his annual speech to the problem of drug addiction. On the national news, Shumlin points out the link between prescription painkillers and death, and he calls for treating opiate addiction as a medical problem no different than cancer. The White House praised the governor’s position.
Meanwhile in another part of Washington, I’m involved in the federal effort to link the law enforcement Prescription Drug Monitoring Program databases to the health records physicians use, and to link the databases across state lines.
The unintended consequences of criminalizing addiction and driving medical problems underground need to be considered here as well.
Physician-patient confidentiality is important to public health, and networked electronic health records have both individual privacy and public health consequences. Privacy is essential in infectious disease testing, domestic violence, mental health, adolescent, reproductive, and addiction medicine. Subjecting clinical encounters to law enforcement surveillance beyond the physician’s discretion is life-threatening.
Well-meaning people are now working to link PDMP databases to EHRs and across state lines. The evidence to justify the coerced crossing of the criminal – medical boundary is anecdotal findings in pilot studies that more physicians are in a position to uncover addiction and offer treatment.
The other goal is to reduce illegal diversion of prescription drugs by both physicians and patients. What could possibly go wrong?
Let’s say a physician writes a prescription for Colchicine and accidentally orders “10.0 mg,” when he should have ordered “1.0 mg.” That’s a tiny decimal error, a mistake even the best doctor could make. But it can be catastrophic for the patient. The higher dose could cause Colchicine poisoning, similar to arsenic poisoning: burning in the mouth and throat, excruciating abdominal pain. Internal organs would melt away and death would likely occur within 24 to 72 hours.
The ease with which even the best doctors can make gruesome errors is why hospitals set up elaborate systems to check and double check orders before drugs are given to patients. Some hospitals are better at this checking than others. Medication errors happen all the time, an estimated one million each year, contributing to 7,000 deaths. On average there is one medication error every day for every inpatient. Let’s take a closer look at what’s contributing to these preventable errors.
Hospitals Are In The Technological Dark Ages
According to recent research, the best known way for hospitals to protect patients from errors is by adopting technology called computerized physician order entry (CPOE). The physician (or other authorized prescriber) enters orders for a patient on a computer that contains patient information such as key lab values, clinical condition, allergies, etc. The computer checks the safety and appropriateness of the order and sends it electronically to the pharmacy. In the Colchicine example, a good CPOE system would alert the physician to the misplaced decimal in the order, and the best systems would prevent the order from being written in the first place. In my mind, one of the greatest advances of CPOE is that it eliminates the need for pharmacists to decipher physician handwriting. I’ve often wondered how they do that.
The research suggests errors decline by as much as 85 percent when hospitals implement CPOE, yet the pace of adoption in the hospital industry is agonizingly slow. To jump start progress, the federal government used economic stimulus funds starting back in 2009 to incentivize hospital investment in CPOE and electronic medical records (EMRs). That improved the pace of change, but still, most hospitals are in the Dark Ages when compared to other industries like airlines or retail.
My nonprofit, Leapfrog, finds that only about a third of the hospitals that voluntarily report to our survey meet our standard for full implementation of CPOE. Even for that minority of hospitals that adopt CPOE, the system doesn’t always work as advertised. Like all technology, CPOE must be continually tested and modified. That’s not always happening.
Here’s an interesting clinical dilemma brought to my attention by another physician.
She was asked to refill a prescription for a drug called domperidone to help a patient with lactation. Domperidone is not FDA approved in the United States for any indication. However, in Europe and in Canada it is approved as a promotility agent for patients with a condition called gastroparesis, which causes the stomach to empty very slowly and results in chronic nausea and vomiting. As a side effect the drug is also known to increase the production of prolactin, a hormone that stimulates milk production. In the case of this physician’s patient, she had adopted a child and found that the medication had effectively enabled her to produce milk and nurse, with seemingly no untoward effects. It’s unclear who had initially prescribed the drug, but various online lactation support forums discuss it as an option for women who have trouble with lactation.
The questions: Is it legal, ethical or good medical practice for a physician in the United States to write a prescription for domperidone for a patient who has been using it for lactation with good results? How about for gastroparesis? Where does one get the drug? Is it even legal to sell the drug in the United States?
I’ve cared for at least two patients who have used domperidone. In both instances it was ordered by prescription from an overseas source by a local gastroenterologist. In these two cases my patients had tried just about everything on the market in the United States for gastroparesis and were still struggling with debilitating symptoms. In one case, my patient had required hospitalizations and ultimately a feeding tube because of intractable vomiting. The drug was ineffective in both patients and it was eventually discontinued.
The following column appears today on THCB, in the op-ed pages of the Los Angeles Times and at ProPublica.
Your doctor hands you a prescription for a blood pressure drug. But is it the right one for you?
You’re searching for a new primary care physician or a specialist. Is there a way you can know whether the doctor is more partial to expensive, brand-name drugs than his peers?
Or say you’ve got to find a nursing home for a loved one. Wouldn’t you want to know if the staff doctor regularly prescribes drugs known to be risky for seniors or overuses psychiatric drugs to sedate residents?
For most of us, evaluating a doctor’s prescribing habits is just about impossible. Even doctors themselves have little way of knowing whether their drug choices fall in line with those of their peers.
Once they graduate from medical schools, physicians often have a tough time keeping up with the latest clinical trials and sorting through the hype on new drugs. Seldom are they monitored to see if they are prescribing appropriately — and there isn’t even universal agreement on what good prescribing is.
This dearth of knowledge and insight matters for both patients and doctors. Drugs are complicated. Most come with side effects and risk-benefit calculations. What may work for one person may be absolutely inappropriate, or even harmful, for someone else.
Antipsychotics, for example, are invaluable to treat severe psychiatric conditions. But they are too often used to sedate older patients suffering from dementia — despite a “black-box” warning accompanying the drugs that they increase the risk of death in such patients.