Collaborative care models for the management of mental disorders in primary care
Christina van der Feltz-Cornelis,
Harm van Marwijk,
and Leona Hakkaart-van Roijen
Mental health care in general practice
This chapter aims to provide a critical overview of collaborative care models and their clinical effectiveness, cost effectiveness, and applicability in the general practice setting with GPs as the primary target audience. For decades, it has been suggested that timely diagnosis and care might be improved for patients presenting with various forms of distress, anxiety, somatoform disorders, and depression in the general practice setting (Rollman et al., 2006). The delivery and uptake of antidepressant medication and evidence-based psychotherapies is often suboptimal (Simon, 2002; Bijl et al., 2003; National Institute for Health and Clinical Excellence, 2011; Piek et al., 2011, 2012). Improvement of care for patients with various forms of distress, anxiety, somatoform disorders and depression is more likely to come from changes in the way care is provided than from adding new treatment options at the interface between primary and secondary care (Katon & Unutzer, 2006).
Currently, the standard treatment approach for mental disorder in general practice is called matched care, i.e. the treatment does not follow a general approach but rather should match the patient’s needs at an individual level. In this way, the patient is usually coached in the general practice setting using some form of short-term psychotherapy or self-management in combination with antidepressant medication. However, if symptoms are severe, if the patient feels seriously invalidated, or if there is a high risk of suicide, the referral to specialty mental health care is recommended. The therapy choice is based on severity but also matched with the patient’s characteristics and preferences. As a result, the treatment varies. Available treatment options are watchful waiting, antidepressants, or different types of psychological interventions, and these options depend on patient preferences as well as on differences in setting and provider (e.g. GP, nurse, psychological wellbeing practitioner, psychologist, or consultant psychiatrist). GPs are known to vary in how they diagnose mental illness, either as more context-dependent symptoms, or as an ‘illness’ or disorder in its own right, and therefore treatment options can vary greatly between practices.
The problem with such matched care is that the current high variations of treatment in management are undesirable in terms of equity of access, safety, and quality, and that is if the treatment is provided at all. Variation between practices exists and some groups (e.g. the elderly, and people from different cultural backgrounds) have less access. This is the case in the Netherlands, a country deemed to have a well-developed health care system. Research shows that patients with depressive symptoms who present themselves in a general practice setting and ask for treatment receive guideline based treatment in less than 50% of the cases (Prins, 2010). It is likely that similar percentages exist in other countries as well.
A second problem with current matched care model is that it is less suitable for chronic care as it allows too many patients to ‘fall through the cracks’. Central coordination and monitoring are already difficult in current fragmented health care systems, but mental health issues tend to increase such fragmentation. Mental illnesses are increasingly conceptualized as long-term conditions. Those with the more socially disruptive forms of psychopathology or those who are more vocal in their need for care may receive more care, but others with more avoidant, more long-term symptoms, and equally large major invalidating mental illnesses may receive less appropriate or less immediate care.
Although tailoring matched care to the preferences of the patient corresponds naturally to the ‘regular’ care received in a primary care setting, a third problem is that it lacks clear prognostic determinants with which to match patients to available treatments. It has been argued that some patients who are perhaps suffering temporarily with problems are being ‘overdiagnosed’ as depressed and in need of antidepressants, and perhaps receive antidepressant treatment too early (Van Marwijk et al., 2003). Others who are truly in need of thorough diagnosis and perhaps also pharmacological treatment may receive too little, or even none.