by Craig Klugman, Ph.D.
Confidentiality is one of the sacrosanct principles of medicine. By keeping the secrets that patients share with health care providers, the patient trusts the provider and the provider has the information necessary to diagnose and treat. The Hippocratic Oath, American Medical Association commentaries, 1974 Federal Privacy Act and 1996 Health Insurance Portability and Accountability Act place confidentiality front and center in ethics and law.
Although we place confidentiality on a high pedestal, it does have many exceptions—some which are acceptable and some of which are required. For example, under the Tarasoff rule a provider in most states must report a specific and explicit threat to a third party. Reporting is required for concerns of public health such as abuse or infectious disease. Information must be shared if law enforcement shows a subpoena. A physician may consult colleagues about a case. Administrative assessment and quality improvement review can access patient information without specific patient consent. In some states, a physician may inform a spouse of certain infectious diseases even when the partner does not want him/her to know. Exceptions are not made lightly since the lack of protecting secrets can decrease patient trust and thus the ability for health care providers to help patients. When exceptions are carved out they are generally because maintaining secrecy would substantially harm the patient or a third party.
A new required exception may be added to this list if New Jersey Governor Chris Christie has his way. He is asking the federal government to carve out a HIPAA exception to allow reporting to a family if a loved one has an opioid overdose. Currently, unless the patient consents, a provider cannot tell family if a patient has been treated for an overdose. Christie claims his request for an exception is because an overdose is always a “cry for help.” Even if the overdose is an accident, Christie sees it as a call for assistance. Specifically, Christie states that every use of naloxone (used to reverse opioid overdoses) should be reported to family.
Christie’s statement is heavily loaded with moral judgements and values. His statement presumes that (a) drug use is a bad problem; (b) that the state has an interest in reducing drug use; (c) that the state can reduce drug use; (d) that drug use is simply a choice and not an addiction; (e) every family wants to know and can help, and (f) that everyone using drug wants help. It also displays a lack of understanding of the importance of confidentiality for building trust between a patient and a provider. When trust erodes, patients do not seek care. Consider that one effect of the crackdown on undocumented persons in the United States is a dramatic decrease in the number of undocumented patients seeking care out of fear that the hospital or physician is required to call immigration (only 1 such case has been reported). Therefore, one likely result of such a change to confidentiality would be that people with opioid addictions stay away from treatment and medical care. People would be less likely to call for medical assistance in the event of an overdose and that would mean many more deaths.
According to the Centers for Disease Control and Prevention, the number of opioid related drug deaths has quadrupled since 1999. Sixty percent of the overdose deaths since 2010 have been from heroin as well as natural and synthetic opioid (fentanyl, tramadol) use. What is missed in such broad statistics is that the increase is not universal. From 2014-2015, while 19 states have experienced a statistically significant increase, 8 states actually experienced a decrease. Carving out an exception to confidentiality in a federal law is not a fine tool to deal with this problem, it is an evisceration of a key patient right.
Earlier attempts to loosen HIPAA confidentiality policies have not been successful. In 2016, Congress rejected a change for certain medical conditions including substance abuse, addiction treatment, and mental health related to gun violence.
Most scholars in ethics and public health officials have expressed a concern that the proposal was a bad idea because it shifts focus away from support and away from funding of treatment programs. If a patient presents with other chronic health issues, the family is not contacted, so why would addiction be different? If a patient presents in a seizure or in diabetic ketoacidosis, the family is not automatically contacted against patient wishes. There is not a presumption that the family can make a difference. Although the proposal may have come out of cases where a middle-upper-class family had no idea that their child was in trouble, not all families would make the situation better. What if the family was in fact not supportive but rather abusive, were addicts themselves, or even enabling of the behavior? Then contact could put the person in a worse situation. Arthur Caplan recommends that if there has to be a “compromise” then it should be toward telling healthcare providers rather than families. Many of these overdose patients are unlikely to have a regular health care provider, so the person they would be told is a stranger? That does not seem like it would help.
Perhaps what is needed is not an erosion of a sacred core of medicine, but rather increased funding and opportunities for treatment and support. The panel is looking for an easy, inexpensive legislative response to a medical and social problem. The reality is addiction will not be treated or go away with a tell-all requirement. The answer takes money for facilities, treatment professionals, and training programs to develop such professionals. What is right is not always easy, cheap or politically expedient and it is why politicians should not be making medical decisions.