Letter

In Reply: We very much appreciate the endorsement by Drs. Grace and Christensen of our findings that there are alarming rates of undetected metabolic disorders and infectious diseases among psychiatric inpatients and our conclusion that the lack of routine screening for these illnesses in inpatient settings presents a missed opportunity to connect patients with primary care and other health services.

We certainly agree with our colleagues' recommendation that new patients presenting with psychiatric problems in outpatient settings shouldbe screened for general medical conditions that might be related to their mental health symptoms. Clearly, general medical problems that might cause mental health problems should be ruled out. However, we believe that access to routine primary care is important for other reasons. Our concern in supporting the model that Drs. Grace and Christensen propose is that even if mental health professionals in outpatient settings refer patients to primary medical care, linkage is frequently poor. In fact, psychiatrists have been found to have less access to medical services for their patients than any other specialty care providers ( 1 ). Given the fragmented care system, exacerbated in many ways by the carve-out of behavioral health services from general medical services, persons with serious mental illness often have difficulty being seen by primary care providers, even when referred by psychiatrists. To us it seems unlikely that nonpsychiatrists would be any more effective than their psychiatrist colleagues in connecting their clients to primary care.

We believe that a more effective model for this population would be to place primary health providers, such as primary care physicians, nurses, and physician assistants, within the community mental health setting to screen and treat people and refer them to other providers when more intensive treatment is needed. A recent report from the Agency for Healthcare Research and Quality that addresses evidence for integration of primary health services into specialty outpatient settings for adults with serious behavioral health disorders presents results from three trials of this type of model ( 2 ). The collaborative care models that were studied had intermediate to high levels of involvement by primary care providers and regular contact between general medical and mental health staff, some of whom were co-located ( 3 ). These trials were consistent in reporting improvements in medical care, quality of care, and patient outcomes. Other examples of collaborative models have used advanced practice nurses effectively to provide health care services to people who were also receiving community mental health center services ( 4, 5 ).

Because community mental health centers are the most common organizational structure for treating patients with serious mental illness and because there is an established history of integrating other types of services into these centers, we think that this model is likely to be more successful than the model proposed by our colleagues, which is really an amplification of the existing model of fragmented general medical and mental health care. Of course, we will need to do more research on the clinical effectiveness and cost-effectiveness of these varying approaches. We are hopeful that the current crisis in health care presents a real opportunity to implement and test these alternative models of integrated care.