Clinical Inertia and Resource Utilization: Results for Bipolar Disorder Treatment

Context:  Clinical inertia has been defined as lack of treatment adjustment in situations where change appears warranted.   This phenomenon has been studied for other diseases including hypertension and depression, but not for bipolar disorder. 

Clinical inertia could have implications for the cost and utilization of health care.  For example, failure to resolve the patient’s disorder may lead to additional visits and hospitalizations compared to what would have resulted from successful treatment.  This could include visits or prescriptions to address side-effects of a medication.   Alternatively, patients may stop complying with treatment, which saves some initial costs (e.g. for medications) but may result in higher costs later or for other types of care.

This study measures the utilization impacts of clinical inertia for bipolar disorder, in order to assess potential savings from reducing the extent of this phenomenon.

Methods: Data are for 27,997 visits made by 2,201 patients treated for Bipolar I Disorder through the STEP-BD practical clinical trial (1).  The trial collected detailed data on patients’ clinical status and medication regimen at each visit, as well as their health care utilization.  We first identify visits at which a medication adjustment appears to be indicated, based on various criteria including side effects and non-response to current medication.  Visits are then classified as exhibiting clinical inertia if an indication was present but no medication adjustment was made.  We compare utilization of various types of medical care between patients who did and did not have any visits which exhibited clinical inertia.  Multivariate regression analyses are used to adjust for other patient characteristics that are predictive of utilization.  In addition, standard unit costs from the Medical Expenditure Panel Survey and elsewhere will be used to estimate the spending differences that correspond to measured utilization differences.

Results: 37% of visits showed at least 1 indication for adjustment.  The most common indications were non-response to medication, side effects, and start of a new illness episode.  Preliminary results indicate that among visits with an indication for adjustment, no adjustment occurred 19% of the time, which is suggestive of clinical inertia.  Patients who experienced inertia had higher use of psychiatric hospitalization, but not of hospital emergency departments.

Conclusions: Many patients remain on the same medication regimen despite indications that the regimen is causing side effects or not improving the patient’s condition.  Given the implications for both patient well-being and for resource utilization, further research should investigate possible reasons for this treatment pattern.