Psychologist Prescribing Privileges: Bias and Risk Assessment

Oregon’s legislature has passed the bill: should the governor sign it? Most opinions on this issue are strong, and many have reached the point of invective. Even such a cool mind as Ronald Pies' has weighed in with an emotionally charged editorial.¹ To speak in favor when so many are opposed seems only to invite more affective discharge. On the other hand, editorial views thus far may be moving us toward extremes on an issue that is highly complex. Perhaps a dialectic approach -– what value can we find in an opposing view? -- would be wise at this point. In that spirit, here are 4 considerations that I hope will be useful.

1. Conflict of interest?
2. Oregon is not Massachussetts.
3. Risk assessment: compare the default prescribers, not psychiatrists.
4. Number needed to treat? Number needed to harm?

Conflict of interest?
When a speaker has strong financial ties to a pharmaceutical company, we know to maintain skepticism regarding her comments, and to be vigilant for evidence of undue influence.

So clearly we are familiar with the concept that having a financial stake in a process is a source of bias. And as psychiatrists, surely we are familiar with the subtlety of unconscious mechanisms. We know, for example, how a patient's resistance can blind him to evidence that might help him see more clearly the basis for his anger.

Should we not therefore be skeptical of our own opinions about psychologist prescribing privileges? How can we be certain that an argument against these privileges is appealing because of its logic, and not, instead, because it resonates with an underlying bias in favor of maintaining a beneficial status quo? 

We require speakers to reveal their financial connections to the pharmaceutical industry. We require judges to recuse themselves from cases in which they have vested interests. Why are we not more suspicious of our own thoughts about psychologist prescribing privileges? Some of our arguments might be right, but we should model the behavior we encourage in our patients: let’s look at our own responses to this contentious issue and see if we can find evidence of bias, in the form of the familiar “cognitive errors.”

A search for cognitive errors, such as we sometimes conduct with depressed patients, always warrants caution: the meta-message can too easily be heard as “you’re thinking wrong; that’s why you’re depressed; so it is your fault, actually.” That same caution is warranted here: I do not mean to accuse our profession of being wrong in its judgments about psychologist prescribing privileges. At the same time, we should at least wonder about our thoughts on this issue, looking at the standard list, which for purposes of neutrality I have copied from Wikipedia (see Box). 

BOX

 All-or-nothing thinking (splitting) - Thinking of things in absolute terms, like "always", "every", "never", and "there is no alternative." Few aspects of human behavior are so absolute.

Overgeneralization - Taking isolated cases and using them to make wide generalizations.

Mental filter - Focusing almost exclusively on certain, usually negative or upsetting, aspects of an event while ignoring other positive aspects. For example, focusing on a tiny imperfection in a piece of otherwise useful clothing. 

Disqualifying the positive - Continually reemphasizing or "shooting down" positive experiences for arbitrary, ad hoc reasons.

Jumping to conclusions - Drawing conclusions (usually negative) from little (if any) evidence. Two specific subtypes are also identified: 
    Mind reading - Assuming special knowledge of the intentions or thoughts of others.
    Fortune telling - Exaggerating how things will turn out before they happen. (See slippery slope.)

Magnification and minimization - Distorting aspects of a memory or situation through magnifying or minimizing them such that they no longer correspond to objective reality. In depressed clients, often the positive characteristics of other people are exaggerated and negative characteristics are understated. There is one subtype of magnification: 
     Catastrophizing - Focusing on the worst possible outcome, however unlikely, or thinking that a situation is unbearable or impossible when it is really just uncomfortable.

Emotional reasoning - Making decisions and arguments based on intuitions or personal feeling rather than an objective rationale and evidence.

Should statements - Patterns of thought which imply the way things "should" or "ought to be" rather than the actual situation the patient is faced with, or having rigid rules which the patient believes will "always apply" no matter what the circumstances are.

Labeling and mislabeling - Explaining behaviors or events, merely by naming them; related to overgeneralization. Rather than describing the specific behavior, a patient assigns a label to someone that implies absolute and unalterable terms. Mislabeling involves describing an event with language that is highly colored and emotionally loaded.

Personalization - Attribution of personal responsibility (or causal role) for events over which the patient has no control. This pattern is also applied to others in the attribution of blame.

See anything on the list that might be operative in some of the written criticisms of psychologist prescribing privileges? I do (all of them, in fact), but that could be the result of my own confirmatory bias, right? On the other hand: if you think that none of these cognitive errors are applicable, might that be, at least in part, the result of your own mental filter? How to proceed, given these potential blind spots which we ourselves might have difficulty recognizing? At a minimum, a brutally honest self-appraisal -- akin to examining one's own resistance in therapy -- is warranted. Better would be to operate with the presumption of bias, in which case we should largely recuse ourselves except for providing data, watching carefully lest we choose only those data which support our a priori assumptions.

Oregon is not Massachusetts
Relax, I’m not going to try to argue against the slippery slope by claiming The West is different from the rest of the US. Regardless of our governor's veto decision on this bill, other states will see new or continued efforts on similar legislation.
At the same time, you Easterners may not know how bad things are out here. (Want to move West? You’re welcome!) Nearly every psychiatrist I’ve talked to in Oregon has a waiting list, usually months long -- even in metropolitan Portland. Many have closed their practices to new patients.

What is organized psychiatry doing about the lack of access to care, here in Oregon? What actions are underway -- not just talk -- after years of this problem? We have one new psychiatric residency program as of this year, but that was initiated by my hospital without help from the Oregon Psychiatric Association (OPA). It may slowly address the shortage in our local area, starting in 2012.  Meanwhile, Rick Bingham, departing president of the Oregon Council on Child and Adolescent Psychiatry, continues his multi-year effort to create an analog to the successful phone-consultation programs in Washington and Massachusetts, but is still working on funding and administration issues (an adult component was once discussed with the Oregon Psychiatric Association, he reports). Other than that, I am not aware of any systematic attempts to address this problem, although the OPA may have launched some recently. To my knowledge, no state-wide efforts were underway prior to the current legislation.

Risk assessment: compare the default prescribers- - not psychiatrists
Meanwhile, in the face of this shortage in Oregon, who is taking care of patients with depression, bipolar disorder, anxiety disorders, ADHD, and sometimes even schizophrenia? Answer: the steadfast, well-intended primary care providers-- including family nurse practitioners and physician’s assistants (who, I assure you, do indeed prescribe plenty of psychotropics in our area). They-- not we psychiatrists- - are the relevant comparison group when assessing the risks of the psychologist prescribing program.

How many of these prescribers would understand complex drug interaction issues, such as in the cases posed by Dr Ronald Pies in his recent editorial?¹ What’s needed in such cases is an awareness of interactions as a potential explanation for complex symptoms; and a means of affirming or downgrading that possibility. Dr Pies acknowledged that psychologists might use online drug interaction databases in cases like these. Confirmatory case report: the psychologist I am currently working with, who is doing the post-PhD/post-Master's training required by the Oregon bill, is well aware of drug-drug interaction issues; and she routinely accesses online resources as needed (including showing me a better interaction database than I’d been using). Granted, as a pioneer, she is likely to be an exceptional student-of-the-art, not necessarily representative of those who will follow her. In the future, some will be as good, perhaps many, but we do not know.  We can predict that just like some of our psychiatrist colleagues, some of these psychologists will have poor prescribing habits as well. 

Perhaps data on error rates by prescribing psychologists (eg, from Louisiana, New Mexico, and the Armed Forces) could help us evaluate the wisdom of this legislation, if we had any data of significance. I have not seen any, even in editorials which have attempted to stay balanced and examine evidence. However, we can consider the error rates of my local well-intentioned primary care colleagues, who would be the first to admit that they are undertrained for what they've been asked to do. Looking at the patients they refer: I've seen such superb care as to make a psychiatrist unnecessary, even for a complex patient with Bipolar II. But I've also seen patients whose bipolar history is unmistakable diagnosed and treated as unipolar, with disastrous results; patients who developed lithium toxicity for lack of follow-up; patients with iatrogenic hypothyroidism, EPS, metabolic syndrome, and benzodiazepine-dependence; misdiagnosis, undertreatment, overtreatment; and (very significantly) non-adherence due to lack of patient education. Perhaps most important of all, I’ve seen many cases that should have been referred for psychotherapy before medication trials were initiated. (During this time, I've surely contributed a share of iatrogenic complications and treatment failures as well. My hat is still off to my primary care colleagues for their efforts to address what we Oregon psychiatrists have not been able to take off their shoulders). 

My point: the bar for safety in Oregon's psychotropic prescribing is not set very high right now. Nor is it likely to go up, given the efficacy of primary care CME for the problems noted above (including my own local CME efforts over the last 10 years, to which I have devoted deliberate effort, with only minimal evidence for significant change).

Number needed to treat/Number needed to harm
Ideally we should have data with which to compare efficacy and safety results for primary care prescribers and psychologists: how many patients will a prescribing psychologist need to see to produce benefit greater than that which could be expected from "treatment as usual" here in Oregon? (Number Needed to Treat; NNT). And how many patients will a prescribing psychologist see before causing harm greater than could be expected from usual care? (Number Needed to Harm; NNH).
 
Shall we wait for such data? Society seems to have decided to wait for more data confirming the slow emergency of climate change, so maybe I shouldn't be surprised at a willingness to wait for evidence that psychologists pose less risk than overwhelmed primary care doctors, NP's, and PA's.

In the meantime, I personally have very little doubt that the NNT is small, and that the NNH is negative. But that's an opinion, surely biased by many factors, not least of which is our profession’s lengthy waiting list in Oregon. Psychologist prescribing may not be the best solution (Danny Carlat suggests we create a joint training program, seeking an optimum balance between training for psychotherapy and training for prescribing.2) But it sure seems better than the status quo. So I'm going to help train the psychologists, when asked -- as long as the governor doesn't veto the bill, in response to lobbying by us, the vested interests with no current alternative. 

References:

References

1. Pies R. “Prescribing Psychologists:” Practicing Medicine without a License? www.psychiatrictimes.com/display/article/10168/1545667?verify=0. Accessed April 5, 2010.

2. Carlat D. Psychologist Prescribing : the best thing that can happen to psychiatry. The Carlat Psychiatry Blog. http://carlatpsychiatry.blogspot.com/2010/03/psychologists-prescribing-best-thing.html. Accessed April 5, 2010.