Background
Bipolar disorder (BD) is a chronic disease with high risk of relapse. This disease also results in a high rate of suicidal mortality. Many treatments have been made available in recent years to manage BD and to enhance positive outcome. To improve the quality of care, several guidelines have been developed with the help of evidence-based medicine and expert consensus meetings. Given the rapid growth in the number of available medications, some of these guidelines have been updated. The first guidelines which were published in 1997 by the The Canadian Network for Mood and Anxiety Treatments (CANMAT) were updated in 2005. This update resulted in major modifications in its recommendations while projecting integrated elements of efficacy, effectiveness, and side effects. Further updates appeared in 2007 and 2009. CANMAT guidelines pointed out specific recommendations for bipolar II disorder (BD-II), which was neglected by most of the previous guidelines.
Two studies from the Texas Medication Algorithm Project underline the importance of enhanced adherence to clinical guidelines which can effectively improve the outcome of bipolar patients. Suppes et al. assessed 141 bipolar patients treated in clinics where treatment guidelines were actively implemented in meetings with clinicians and in psychoeducational interventions with patients and families. They compared their outcome with that of 126 patients treated in a standard "as usual" protocol. The first group of patients showed greater sustained improvement at the initial and later stages with respect to overall severity of psychiatric symptoms, as well as in manic symptoms. For the same group of 141 bipolar patients, stricter adherence to treatment guideline recommendations was associated with greater reductions in depressive symptoms and overall psychiatric symptoms over time.
Despite the recent emphasis on evidence-based practice, guideline recommendations are not always implemented. Perlis found that 34% of psychiatrists reported not having recourse to guidelines on a regular basis to treat BD, where only a very small percentage identified guidelines as their primary source of information. Similar results were obtained in two other studies conducted in France and Serbia, where 40% and 34% of psychiatrists, respectively, had claimed non-use of guidelines in the management of BD in daily practice.
The treatment of bipolar patients in concordance with guidelines varies widely throughout the studies, as shown in Table 1. Except for the low percentages found in three studies, concordance percentages ranged from 50% to 80%. Smith et al. found that concordance of treatment was 52% in a naturalistic setting, with a percentage increasing to 75% after a one-year psychoeducational intervention aimed at implementing the guidelines. Bauer et al. found a concordance of 50–60% at the index hospitalization with respect to 306 participants monitored through a collaborative care model; there was, however, a marked decrease after 2 years. Dennehy et al. found high percentages of concordance with treatment guidelines among psychiatrists who had participated in extensive training based on published clinical practice guidelines.
Some factors were linked to better concordance with guidelines. Manic patients were more likely to receive concordant treatment than patients suffering from depressive or mixed episodes. The presence of psychotic features was related to higher percentages of concordance with treatment guidelines. Treatment setting influenced concordance to treatment guidelines, with in-patients more frequently receiving adequate treatment than out-patients. Patients diagnosed with anxiety or depressive disorder before being diagnosed with BD were less likely to receive antimanic medications and more likely to receive antidepressants without any antimanic treatment. In the STEP-BD study, earlier age at onset and administration of adequate pharmacotherapy at entry predicted those more likely to receive guideline-concordant care during new-onset mood episodes. In the same STEP-BD study, bipolar I patients were more likely to receive adequate treatment as opposed to bipolar II patients. Arvilommi et al. found that with the lack of clinical diagnosis of BD at entry, rapid cycling, polyphasic index episode, as well as depressive index phase, were independently associated with inadequate treatment. In a study assessing the treatment of veterans suffering from BD, race was not an associative factor in the administration of mood stabilizers.
To our knowledge, there has never been a study conducted on concordant treatment of bipolar patients in keeping with the CANMAT guidelines in Canada. The first object of our study was to examine the administration of psychotropic drugs among subjects diagnosed with BD and who had been referred to tertiary care services. The second objective was to assess concordance of prescribed treatments with Canadian guidelines. The third objective was to identify the clinical factors linked to poorer concordance of treatment with CANMAT guidelines.