Do clinicians follow heuristics in prescribing antidepressants?

Research Letter: Application of GPT-4 to select next-step antidepressant treatment in major depression

Introduction

Large language models perform well on a range of academic tasks including medical examinations. The performance of this class of models in psychopharmacology has not been explored.

Method

Chat GPT-plus, implementing the GPT-4 large language model, was presented with each of 10 previously-studied antidepressant prescribing vignettes in randomized order, with results regenerated 5 times to evaluate stability of responses. Results were compared to expert consensus.

Results

At least one of the optimal medication choices was included among the best choices in 38/50 (76%) vignettes: 5/5 for 7 vignettes, 3/5 for 1, and 0/5 for 2. At least one of the poor choice or contraindicated medications was included among the choices considered optimal or good in 24/50 (48%) of vignettes. The model provided as rationale for treatment selection multiple heuristics including avoiding prior unsuccessful medications, avoiding adverse effects based on comorbidities, and generalizing within medication class.

Conclusion

The model appeared to identify and apply a number of heuristics commonly applied in psychopharmacologic clinical practice. However, the inclusion of less optimal recommendations indicates that large language models may pose a substantial risk if routinely applied to guide psychopharmacologic treatment without further monitoring.

Initial antidepressant choice by non-psychiatrists: Learning from large-scale electronic health records

OBJECTIVES

Pharmacological treatment of depression mostly occurs in non-psychiatric settings, but the determinants of initial choice of antidepressant treatment in these settings are unclear. We investigate how non-psychiatrists choose among four antidepressant classes at first prescription (selective serotonin reuptake inhibitors [SSRI], bupropion, mirtazapine, or serotonin-norepinephrine reuptake inhibitors [SNRI]).

METHOD

Using electronic health records (EHRs), we included adult patients at the time of first antidepressant prescription with a co-occurring diagnosis code for a depressive disorder. We selected 64 variables based on a literature search and expert consultation, constructed the variables from either structured codes or through applying natural language processing (NLP), and modeled antidepressant choice using multinomial logistic regression, using SSRI as the reference class.

RESULTS

With 47,528 patients, we observed significant associations for 36 of 64 variables. Many of these associations suggested antidepressants' known pharmacological properties/actions guided choice. For example, there was a decreased likelihood of bupropion prescription among patients with epilepsy (adjusted OR 0.49, 95%CI: 0.41-0.57, p < 0.001), and an increased likelihood of mirtazapine prescription among patients with insomnia (adjusted OR 1.59, 95%CI: 1.40-1.80, p < 0.001).

CONCLUSIONS

Broadly speaking, non-psychiatrists' selection of antidepressant class appears to be at least in part guided by clinically relevant pharmacological considerations.