(The following is a brief discussion of the ethical considerations involved in decisions about whether to disclose a patient’s HIV status to a third party. If you’re not interested in that, skip to the bottom of this page for links to this month’s articles)
This past Saturday was World AIDS Day 2018. Last week we learned that, marrying bad science and bad science ethics, a Chinese researcher claims to have used CRISPR to perform germline editing on human embryos. Why? Reportedly, in order to knock out the CCR5 genes and give the kids resistance to HIV infection – which the researcher deemed indicated because the children’s father has HIV infection. Other folks have explained why this work is more than a little ethically problematic, but let’s also appreciate that there was no substantial risk of HIV transmission worth mitigating in the first place – and it certainly wasn’t worth rolling the dice on CRISPR mis-splicing somewhere and setting the kids up for leukemia. Good grief.
Anyway, medical ethics have been on my mind because one of my colleagues in our Ethics & Health Policy department invited me to facilitate a workshop with their faculty on the topic of HIV disclosure. Specifically, I’m going to try to answer one question: when is it ethically permissible to disclose a patient’s HIV status to a third party without that patient’s consent?
My gut reaction to this question was “never,” and while my considered opinion is a bit more nuanced, it still leans heavily toward nondisclosure in most cases. This is coming from a few places, which I’ve detailed below.
First, there are established legal and professional standards surrounding disclosure of a patient’s HIV status. The AMA Code of Ethics advises us to “attempt to persuade patients who are identified as HIV positive to cease endangering others” but also “to adhere to state and local guidelines regarding public health reporting and disclosure of HIV status when a patient who is identified as HIV positive poses significant risk of infecting an identifiable third party.”
While I don’t want to give the impression that dilemmas in medical ethics should be framed as “what can I legally get away with?” the latter point is important because HIV status is protected health information. Both insurance companies and individual practices have been successfully sued to the tune of millions of dollars for legally non-protected disclosures. Admittedly, these were cases of negligent behavior rather than a physician’s deliberate disclosure of protected health information to protect a third party – and in fact some US states have legislated a clinician’s duty to warn when a patient endangers others. However, both the presence of a legal duty to warn and the legal status of HIV nondisclosure between a positive person and their partner vary significantly throughout the US and world. The law may even differ based on the relationship between the patient and their partner (e.g. it may be legal for a physician to disclose a patient’s HIV status to their spouse, but illegal to disclose to their boyfriend). This is all to say that you should understand your local health privacy laws and have a chat with your institution’s Risk Management office before even thinking about violating patient-physician confidentiality.
Now for the purely ethical considerations. My key point is that the consequentialist argument for HIV disclosure, the strongest argument FOR breaching patient confidentiality, is not as clear cut as it may seem.
Consider the situation in which your patient newly diagnosed with HIV is having unprotected sex with their partner, who is unaware of the patient’s HIV status, and your patient does not want to disclose to them. The argument for disclosing your patient’s HIV status to their partner without consent is that in doing so you might (1) prevent the partner from acquiring HIV or (2) diagnose the partner’s already-acquired HIV and get them into care to prevent further transmissions. Makes sense, right?
Or does it?
I argue that modern ART has changed the calculus of HIV disclosure. First, understand that the per-act risk of HIV transmission is quite small. Second, recall that our first-line integrase inhibitor-based regimens reliably drop the viral load to undetectable within a matter of weeks. And while a person needs to have an undetectable HIV viral load for six months to fall under the definition of Undetectable = Untransmissible, data suggest that even just being on effective ART reduces the per-sex-act HIV transmission risk to as little as 1 in 25,000 – that’s odds of being hit by lightning territory. Finally, let’s consider that the primary demographic driving new HIV transmissions in the US today is people who know they have HIV but are not retained in care; in fact, these folks are responsible for more than twice the number of transmissions accounted for by people who are unaware of their infection.
Keep in mind also that HIV-related stigma is still a very real problem and that your patient may have well-founded concerns about informing their friends and family. Patients who disclose their HIV serostatus or have their status disclosed by others may face ostracization from family members or verbal, emotional, and/or physical abuse. They may be cast out of their homes, lose their jobs (the protections of the ADA not being worth much to folks who can’t hire a lawyer), find their HIV status being wielded against them in court, or even face physical harm. Will your patient be more or less likely to stay on ART if you endanger their social support network, housing, source of income, and/or personal safety?
So here’s my question for you. Even if we decide we’re going to be ruthless pragmatists who care only for consequences when making ethical decisions, what’s really the best course of action in this situation? Is it betraying patient-physician confidentiality for a small absolute reduction in risk of HIV transmission to your patient’s partner – knowing that said betrayal might compromise your patient’s social support network and will certainly alienate them from you and possibly from the medical community writ large? Knowing you’ll be adding one more person to that high-transmission-risk group of folks who know they have HIV but aren’t getting care? Or is the better decision to encourage your patient to disclose their serostatus and temporarily avoid activities that carry transmission risk while ultimately respecting their decisions, working on getting them on ART and undetectable, and building the sort of therapeutic alliance that’ll enable you to persuade them to disclose to their partners voluntarily in the future? I argue it’s the latter. If the best way to prevent HIV transmission in the community is to get and retain people living with HIV in care, then respecting the patient-physician relationship is in the best interests of both the individual patient and society.
Clinicians are an outcomes-focused lot, so I’ve devoted the most effort to the appeal to consequences above, but there are other ethical considerations. I think most of these either lean toward nondisclosure or are easily construed to support either position:
Rights-based arguments can go either way here. Arguments for disclosure would point toward the partner’s right not to be harmed, and those for nondisclosure would point toward respect for the patient’s decisional autonomy. Weighing these arguments against each other devolves into questions about quantitating the risk of HIV transmission from patient to partner, how “bad” it is to have HIV infection in 2018, and how those factors weigh against the value of a patient’s decisional autonomy. Given the degree to which we can mitigate HIV transmission risk with ART, and the long duration and high quality of life so many of my patients with HIV enjoy, I lean toward respect for patient autonomy being the more valuable right to uphold in most cases. Your mileage may vary.
Virtue-based ethical arguments (i.e.; “what course of action best reflects virtuous behavior?”) are annoying because they can always be argued for every position in an ethical dilemma. Both choosing to break patient-physician confidentiality to protect a third party and choosing to honor said confidentiality can be interpreted as demonstrating virtues such as respect, integrity, compassion, courage, and self-sacrifice, and I don’t find one interpretation more compelling than the other. Virtue ethics are a fine way to approach the world – but I find they rarely bring much decisional clarity to medical ethics cases.
Finally, our special professional obligation to our patients as clinicians – what’s termed the fiduciary responsibility – clearly argues for respecting patient-physician confidentiality. Infectious disease specialists, perhaps more than practitioners from any other specialty, understand that our patients’ diseases have origins in and implications for the wellbeing of the overall society. By the same token, however, we ought not lose sight of the fact that first and foremost we work for our patients. Our fiduciary responsibility compels us to act in our patients’ best interests before anyone else’s – to some degree, even before our own. While we ought to take the health of the larger community into consideration when making medical decisions, it’s not our place to sacrifice our patients’ wellbeing purely for the benefit of other parties. Even when we’re at our most draconian – quarantining non-adherent patients with active pulmonary tuberculosis to protect the general public – we’re still acting in those patients’ best interests by ensuring they receive the treatment they need.
What are the situations in which it might be acceptable to disclose a patient’s HIV status to someone else without their consent? Succinctly: only after you’ve exhausted all other options, and even then only when the benefit clearly outweighs the harms.
A realistic scenario in which I would advocate for disclosure is one in which the patient’s HIV status needs to be disclosed in order to facilitate their own care. For example: a young man who’d been reluctant to seek care is brought into the hospital by his mother (his sole family & now sole caregiver) after several weeks of neurologic deterioration. You discover that he has advanced HIV infection and progressive multifocal leukoencephalopathy. He tells you that he wants to keep his HIV status to himself, but is unable to care for himself on account of his illness and is unable to provide a reasonable alternate plan.
In this situation, the treatment is ART. His mother is going to be the person taking care of him at home, taking him to doctor’s visits for follow-up, taking him to the pharmacy to get his meds, and probably also paying for and administering those meds. It’s not realistic to expect the patient will be able to receive the care he needs while keeping his mother out of the loop. In this situation, I’d sit down with the patient to talk about disclosing his status to his mom (at length and on more than one occasion), would try to identify and address his specific fears about disclosure, and would give the patient as much time as possible to come around on his own to disclosing (either directly or in a clinician-facilitated meeting). But ultimately the patient needs to go home with a plan for treatment and follow-up of his PML and HIV, which means mom needs to know about his serostatus.
Last but not least - if you ever find yourself struggling with a question about HIV disclosure, I highly recommend consulting the AIDS Education Training Center’s module on HIV Disclosure, Confidentiality, and Decision-Making, which has a nice ethical framework distilled into flowchart form. Not only is it a thoughtful and structured approach, but this way you’ll be able to say you followed AETC’s ethical guidance on the matter rather than what some guy said on his blog.
And now, for the medical literature:
Antimicrobial agents research this month included substituting linezolid for ethambutol in the treatment of pulmonary TB, the efficacy of cefepime and pip-taz for bacteremia with organisms producing AmpC, the outcomes of MRSA bacteremia deescalated to oral therapy, sertraline for aspergillus, aminoglycosides for UTI, and phages for the treatment of cholera (in rabbits).
ID diagnostics research covered the development of a computer algorithm to detect TB lesions on chest x-ray, the reliability of antimicrobial susceptibility testing performed directly on urine samples, the association between rigors and blood culture positivity, the utility of the BioFire FilmArray blood culture PCR panel when used on respiratory samples, and the use of blood culture bottles to improve culture yields in prosthetic joint infections.
General ID topics for the month included the burden of MDROs in India, pasteurization of breast milk to prevent neonatal CMV, risk factors for inadeqaute empiric treatment in diabetic foot infection, and the association between adequacy of perioperative prophylaxis and subsequent risk of vascular graft infection.
HIV and STDs literature this month included zolifodacin (a new drug for gonorrhea), another study on when asymptomatic patients with HIV need an LP, the diagnostic accuracy of the HIV proviral DNA genotype, more data to support 2-drug ART for maintenance of suppression, and GMO microbiota as a novel strategy to prevent HIV transmission.
Onc and Transplant ID research covered the association between ribavirin use and outcomes of RSV infection, ESBL colonization and bacteremia risk in stem cell transplant recipients, CMV mismatch and renal transplantation outcomes, and why Nico is angry about measles.
Antimicrobial stewardship and infection control literature topics include the lack of efficacy of probiotics for acute gastroenteritis in kids, FMT for the eradication of MDRO colonization, the lack of efficacy of UV light disinfection for C.difficile and VRE in a transplant ward, and the influence of diet on gastrointestinal carriage of antibiotic resistant bacteria.