Quality of care for major depression and its determinants: a multilevel analysis

Abstract

Background: Numerous studies highlight an important gap in the quality of care for depression in primary care. However, basic indicators were often used. Few of these studies examined factors associated with receiving adequate treatment, particularly with a simultaneous consideration of individual and organizational characteristics. The purpose of this study was to estimate the proportion of primary care patients with a major depressive episode (MDE) who receive adequate treatment and to examine the individual and organizational (i.e., clinic-level) characteristics associated with the receipt of at least one minimally adequate treatment for depression.

Methods: The sample used for this study included 915 adults consulting a general practitioner (GP), regardless of the motive of consultation, meeting DSM-IV criteria for MDE during the 12 months preceding the survey (T1), and nested within 65 primary care clinics. Data reported in this study were obtained from the "Dialogue" project. Adherence rates for 27 quality indicators selected to cover the most important components of depression treatment were estimated. Multilevel analyses were conducted.

Results: Adherence to guidelines was high (>75%) for one third of the quality indicators that were measured but was low (<60%) for nearly half of the measures. Just over half of the sample (52.2%) received at least one minimally adequate treatment for depression. At the individual level, determinants of receipt of minimally adequate care included age, having a family physician, a supplementary insurance coverage, a comorbid anxiety disorder and the severity of depression. At the clinic level, determinants included the availability of psychotherapy on-site, the use of treatment algorithms, and the mode of remuneration.

Conclusions: Our findings suggest that interventions are needed to increase the extent to which primary mental health care conforms to evidence-based recommendations. These interventions should target specific populations (i.e. the younger adults and the elderly), enhance accessibility to psychotherapy and to a regular family physician, and support primary care physicians in their clinical practice with patients suffering from depression in different ways such as developing knowledge to treat depression and adapting mode of remuneration.

Figure 1

Recruitment flow-chart, Dialogue Project.

Figure 2

Receipt of minimally adequate treatment for depression among a sample of 915 adults consulting in primary care and meeting criteria for past year MDE in the Dialogue Project in 2008.

Figure 3

Estimated probability of receipt of at least one minimally adequate treatment for 3 virtual patients meeting criteria for past year MDE across two virtual clinics in the Dialogue Project in 2008. Legend : Virtual patient 1 (“Standard ” patient ) : ·Female, Aged between 25 and 44, No family physician, No supplementary insurance coverage, No comorbid anxiety disorder, Grand mean score on the HADS scale. Virtual patient 2 (“Low probability” Patient): ·Male, Aged 65 or more, No family physician, No supplementary insurance coverage, No comorbid anxiety disorder, Grand mean minus 1 as score on the HADS scale. Virtual patient 3 (“High probability” Patient): ·Female, Aged between 25 and 44, Family physician, Supplementary insurance coverage, Comorbid anxiety disorder, Score on the HADS scale = 15. Virtual clinic A (“Worst” clinic): No psychotherapy on-site, None or some GP using treatment algorithms with individuals suffering from anxiety or depressive disorders, Mode of remuneration to offer an “optimal level” of care for patients suffering from anxiety or depressive disorders perceived as highly inadequate. Virtual clinic B (“Best” clinic): Psychotherapy on-site, All or most of GP using treatment algorithms with individuals suffering from anxiety or depressive disorders, Mode of remuneration to offer an “optimal level” of care for patients suffering from anxiety or depressive disorders perceived as not at all or slightly inadequate.