Wednesday, 18 Oct 2017

You are here

Forced, Rational or Glitch-Ridden Prescribing Practices

A recent analysis of 3 groups of treatment-naïve, early rheumatoid arthritis (ERA) patients looked at the factors that influenced the choice of therapy.

Those with commercial insurance were almost twice as likely to start a biologic DMARD compared to Medicaid patients. African-Americans had significantly lower odds (~30% lower) of initiating or switching to biologic DMARDs than non-Hispanic whites. Prior steroid and DMARDs use were more likely before biologic initiation and switching. 

While these “predictors” of treatment choice seem obvious, the driving influences behind treatment are numerous and not always as scientific or rational as we may hope. 

While you could’ve logically come to the same conclusions, do these qualify as best practices or even what you actually do? 

What if the patient is your wife, sister or mother with ERA? You’re smart, aggressive and expert. What would you do?

Would you give her the best therapy money or insurance can buy? Would you be guided by either ACR or EULAR treatment guidelines? Or has all your expertise and mastery of RA been undermined by managed-care algorithms designed for cost effectiveness?

The “Rationality Paradox” states that while we strive make rational decisions that lead to the best outcomes, we humans tend to make choices that are subconsciously driven. At worse, we are “glitch-ridden, superstitious creatures of habit”. At best we rationalize with logic that is part evidence and part mainstream thinking.

For instance, we prioritize the urgent, which tends to lead to reactionary decisions, rather than optimal or planful ones. 

While you may talk logically and evidence based about treatment choices, do your actions and prescriptions always adhere your philosophy?   

These few questions on prescribing (for an active RA patient, for instance) may identify between your rational, non-rational and forced decisions:

  1. Do you Risk Stratify?  There are numerous studies that identifying predictive factors for worse outcomes and that these can be used to guide therapy. These include many joints, nodules, erosions, extraarticular disease, low education, multiple DMARD failures, comorbidities, etc. But how many of these are needed for you to prefer and recommend the more aggressive vs less aggressive initial treatment regimen?
  2. Do you Use your Best Drug First?  A 2010 UK ERA Network study showed the first DMARD choice to be MTX (46%), SSZ (42%), other DMARD (9%) or in 3%, a combination DMARD therapy.  Most of these choices are based on habit or forced prescribing. Whether forced or not, you cannot guarantee whether a patient will preferentially respond to one DMARD over another.  Yet you believe you are “wise” because you may have your preferences or you may inappropriately quote comparative ACR20 results from non-head to head studies or you may ultimately rely on patient preference.  The good news is that rheumatologists are quite good given these limitations. The bad news is that that without a predictive variable or biomarker (that makes your choice better than other random choices), your still “guessing”.
  3. What is your Benchmark for Success? What’s your goal?  What’s your objective measure, if you use one (Editor’s note: my surveys show that 54% of US rheumatologists use some “objective” metric for disease assessment)?  More importantly, why do you wait 12 weeks or more to define response when for most drugs you can accurately make this determination in 6 weeks?  Waiting another 6 weeks for the “great” response not already evident by 6 weeks is indecisive optimism at the patient’s expense.  Why not accept that it’s best to cycle through your best therapies as quickly as possible to get to your desired response (hopefully remission) ASAP.
  4. Do you follow the 2015 ACR RA Treatment Guidelines?  I believe there is a strong role for such guidelines, that is until they make no sense with regard to the patient in front of me who should not be subjected to algorithmic choices when they came to me for a thinking man’s management of a highly complex disorder.
  5. What are your Switching Rules?  My surveys say that the average rheumatologist uses about 1.6 TNF inhibitors (TNFi) before switching to a non-TNFi agent. The actual, real world results say its two or more failures and this is supported by managed care guidelines that you must fail (for instance) Enbrel and Humira before you can use that other drug (a non-TNFi, tofacitinib, golimumab, certolizumab).  My review of the data suggests that all primary nonresponders to a TNFi should stop using TNFi (2nd or 3rd).  There is compelling data to suggest that regardless if the patient is a primary or secondary failure, the use of a non-TNFi when changing is better than sticking with the same TNFi class.  

On a good day your prescribed therapy was a good, educated guess that worked.  When you’re not so lucky, your best guess fails (for reasons unknown to you) or was forced upon the patient (by insurance protocol); leaving you and the patient to wonder about the next best solution.

What is the solution to the uncertainty of prescribing or the distasteful and financially motivated protocols force upon you and the patient?  There are a few: A) head to head trials; B) Big data (see “Moneyball”) or C) Doing the best with what you’ve got:

  • When you lay out your plan A, you better have a plan B and C in mind already;
  • Have a clear understanding between you and the patient as to what “success” looks like. Without a defined goal or success you will linger in indecision purgatory and the patient will needless suffer;
  • Treatment failure mandate reanalysis and understanding of why you or the therapy failed. If you don’t understand your history (of failure) you are destined to repeat it;
  • Choose rapid-acting DMARDs or biologics so as to make rapid “go or no-go” decisions at week 6 or 8;
  • Have a plan for “forced” prescribing that you don’t agree with.  This must include a well written “appeal letter” with supportive data or references; and
  • Ask the patient what they prefer.  Patients who have their say, are invested or inclined towards a certain therapy (no matter what the reason) are more likely to A) fill the prescription; b) take the drug as prescribed; c) respond better to their choice vs. the unappealing and apprehensive  drug that you preferred.  

 

Disclosures: 
The author has no conflicts of interest to disclose related to this subject
Dr. Cush is the Director of Clinical Rheumatology at the Baylor Research Institute and a Professor of Medicine and Rheumatology at Baylor University Medical Center in Dallas, TX. He is the Executive Editor of RheumNow.com and Co-Editor of the online textbook RheumaKnowledgy.com. Dr. Cush's research and interests include novel drug development, rheumatoid arthritis, spondyloarthritis, drug safety, pregnancy and Still's disease/autoinflammatory syndromes. He has published over 140 articles and 2 books in rheumatology.
He can be followed on twitter: @RheumNow.

Add new comment