Who should drive: patients, clinicians, or science?

The model of the expert physician making decisions for a compliant patient and giving orders to obedient nurses and technicians is thankfully disappearing, being replaced by a patient-centered team approach. 

Atul Gawande talks about a transition from cowboy to pit crew. During this same time period, there has also been a transition from tradition/consensus based care to evidence based care. There has been too little discussion of the potential conflict between evidence-based medicine and patient-centered medicine.

For the purpose of this post:

  • Traditional expertise driven care: the clinical expert, traditionally a physician, uses his personal store of medical knowledge and accumulated experience to determine what to tell the patient, and what testing or treatments to offer. The patient is encouraged to ask questions primarily in the service of compliance.
  • Evidence based care: the decisions about what testing or treatment options are appropriate are determined by the evidence. When there is an absence of evidence, or if the evidence finds various options equally valid, then either expert driven care or patient driven care may hold sway.
  • Patient driven care: the clinician helps the patient understand all the options and then helps the patient manage the diagnostic and treatment plan selected by the patient.

To the extent that these issues are discussed, the answer is usually a recommendation that the clinician and the patient use the best available evidence to make a collaborative decision about what is best, often called shared decision making (SDM). (See here  andhere.)  The process envisioned looks roughly like this:



Sounds great, but both EBM and SDM authors fail to mention - let alone address - some important problems with this approach:

  • We don’t always have the evidence we need, and what evidence we have may be hard to apply or suspect. (A discussion of the implications of bad evidence on evidence based care will be a topic for a later post.) 
  • The data, statistics and physiology are often complex and may be very hard for the patient to understand without education and training.
  • It takes a willing clinician considerable time and sensitivity to have even a partial understanding of the complex and subjective patient context.
  • Not all patients really want to make medical decisions, and the level of interest in participating in the decision making process varies from patient to patient, day to day, and issue to issue.
  • How much time is available and how much is needed? A simple interaction about a sore throat can be a simple interaction (here’s your prescription, take it for the full 10 days, call me if you have rash or diarrhea) or quite complex (discussing whether or not there is actually any benefit in treating strep throat with antibiotics). Imagine the issues in discussing screening with PSA, risks of over-diagnosis with mammography, when to start and how often to do bone densitometry studies, whether or not to treat mild depression with medication.)
  • Complex issues rarely have simple solutions. Shared decision making about the value of a PSA in a low risk 55 year old requires an understanding of complex issues of tumor behavior, epidemiology, test performance, diagnostic processes and the treatments and their potentials for benefit and harm. This discussion cannot be completed in an hour during a presentation to residents or practicing clinicians who already know much of the data and understand the issues. Doing this properly within the constraints of a routine annual health maintenance exam is a fantasy, and the patient is not likely to want to take more time off work or pay a second co-pay to do a formal SDM process. 

What do these conflicts look like in practice? 

  • A patient with 3 days of runny nose, scratchy sore throat, aching, cough requests antibiotics for their viral upper respiratory illness (URI). The evidence clearly says that antibiotics are not helpful and can be harmful, the patient says”My last doctor always gave me antibiotics when I felt like this and I always got better in a few days.’ The physician expert is torn: follow the evidence and avoid the risk of harm from a useless treatment, or satisfy the patient by acquiescing? (Note: the same dilemma occurs with sinus symptoms and viral bronchitis. Combined, these three diagnoses account for the majority of inappropriate antibiotic prescribing in the US, at a cost of millions of dollars and with significant individual and societal harm.)
  • A 37 year old in the midst of a messy divorce reports an increase in her chronic (since age 23) migraine and requests a CT scan of the head ‘because my neighbor’s cousin had headaches and died of a brain tumor.’
  • A young couple declines immunizations for their newborn, saying that diseases like polio and diphtheria and whooping cough and measles are rare, giving lots of immunizations wears out the immune system, and that they know someone whose child had immunizations and developed autism. They have researched the subject and decided that the reason the published evidence overwhelming refutes their concerns is that doctors and the drug industry are engaged in a conspiracy to hide the truth in order to make money with immunizations.
  • I start to discuss the option of immunization against invasive pneumococcal pneumonia (Pneumovax) with a patient. If I frame it as a safe and covered vaccine to reduce the chances of a rare but potentially lethal illness, a fair number of patients agree. If I point out that it takes over 4,000 vaccinations to prevent one case, most decline. Both are true, so I can to large degree control the outcome by what version of truth I offer. 
  • After a discussion of the potential benefits and risks of immunization against HPV, the mother or father asks me: ‘What would you do if it were your son/duaghter?’

These days, it feels as if we need a moderator, to help sort out the conflicts among my training and biases, the patient’s expressed and unexpressed needs and desires, and the evidence (which cares not a whit for what either I or my patients desire). 


Footnote: Learn more about EBM here: http://onlinelibrary.wiley.com/doi/10.1002/9781444342673.ch1/summary A nice discussion about EBM is here: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2349778/pdf/bmj00524-0009.pdf



Links to more on this topic::