Metabolic Monitoring
Given the increased risk of metabolic syndrome among people with SMI, and the association of metabolic syndrome with increased morbidity and all-cause mortality, there has been a growing awareness of the importance of screening for metabolic syndrome among people with SMI. Metabolic monitoring involves routine screening for metabolic parameters and assessment of metabolic risk factors among people with SMI who are prescribed antipsychotic medications. Various practice guidelines have been developed in the United States and internationally to assess for metabolic risk factors in people prescribed antipsychotic medications [26]. Current metabolic monitoring guidelines in the United States stem from 2004 consensus recommendations of the American Diabetes Association and American Psychiatric Association along with the American Association of Clinical Endocrinologists and the North American Association for the Study of Obesity for metabolic monitoring among people prescribed SGAs [23]. These recommendations include a time line for routine monitoring of weight/body mass index, waist circumference, blood pressure, fasting blood glucose or hemoglobin A1c, and fasting lipids (Table 2). Guidelines recommend screening at baseline, more frequently within the first 3 months, and then annually [23].
Though guidelines recommend measurement of waist circumference as a marker for metabolic health, body mass index is often used alone as a measure of obesity [27,28]. This may be due to the relative ease of obtaining weight over waist circumference. For example, weight is more likely to be part of clinic workflows and many providers may not be accustomed to measuring waist circumference. However, waist circumference does provide additional information regarding metabolic health [29], as central adiposity is a marker of cardiometabolic risk and related to insulin resistance [21]. Further modifications of the guidelines have included ethnicity-specific waist measurements [30].
There is evidence that non-fasting lipids may be substituted for fasting lipid panels, particularly for patients who may have difficulty adhering to fasting due to cognitive difficulties. Vanderlip and colleagues argue that fasting serum cholesterol panels are not necessary for screening for dyslipidemia given that non-HDL cholesterol is calculated based on total cholesterol and HDL, which do not substantially differ between fasting and non-fasting values [31]. Hemoglobin A1c is recommended as a screening test for blood glucose abnormalities given that it does not require a fasting state and can therefore be more easily obtained for many patients. The choice to obtain a fasting blood glucose versus hemoglobin A1c may depend on multiple factors, including that a person can adhere to fasting and the cost of the laboratory test.
Routine monitoring of metabolic parameters is an integral step in targeting interventions to treat metabolic syndrome. These interventions include lifestyle modifications and evidence-based treatment guidelines for management of associated dyslipidemia, hypertension, and type 2 diabetes mellitus.
Current Metabolic Screening Practices
Despite the presence of defined guidelines, estimates show persistently low rates of metabolic monitoring among adults prescribed SGAs [32]. One study of 3 state Medicaid programs showed little to no improvement in screening rates for glucose and lipids post dissemination of the 2004 APA/ADA guidelines [33]. They noted a nonsignificant change in rates of glucose testing from 27% to 30% and small change in lipid testing from 10% to 11% among patients prescribed SGAs between 2002–2005 [33]. Examining screening rates among Medicaid recipients in Missouri between 2010–2012, Morrato and colleagues found glucose testing rates of 80% with lipid testing remaining at 41% [34]. A retrospective study of adult Medicaid recipients prescribed first- and second-generation antipsychotics between 2008 and 2012 showed rates of screening for lipids and glucose to increase over time; glucose monitoring increased from 56.6% to 72.6% and lipids from 38.3% to 41.2% [35]. A review by Mangurian and colleagues suggested rates of glucose (fasting blood glucose or hemoglobin A1c) and lipid screening as low as 30% among people prescribed antipsychotic medications [14]. Furthermore, they underscore the impact of low screening rates, stating that if 20% of adults with SMI have diabetes and 70% remain unscreened, then approximately 2 million adults with SMI and diabetes in the United States would not be identified within our current system [14].
Higher rates of screening have been shown for Medicaid populations than commercially insured populations [36]. Haupt et al compared lipid and glucose testing pre- and post- ADA/APA guideline implementation among commercially insured patients. They found an increase from 8.4% to 10.5% post guideline implementation for baseline lipid testing and from 6.8% to 9.0% for lipid testing at 12 weeks post-antipsychotic initiation [36]. Baseline glucose testing increased from 17.3% to 21.8% and from 14.1% to 17.9 % at 12-week post antipsychotic initiation. In alignment with findings from other studies, testing rates were particularly low for children [36].
Low screening rates have been found among children and adolescents prescribed SGAs [37] despite evidence that youth may be at risk of developing more significant metabolic sequelae from SGAs [19]. Edelsohn and colleagues found an increase from 30% to 50% for glucose screening and from 19% to 28% for lipid screening among youth Medicaid recipients prescribed first- and second-generation antipsychotics between 2008 and 2012 [35]. Connolly and colleagues reported on metabolic screening rates for children and adolescents prescribed SGAs over the 8 years following announcement of the 2004 ADA/APA guidelines. Using insurance claims data, they found screening rates for fasting blood glucose and hemoglobin A1c temporarily increased following guideline dissemination, then dropped during the period 2004–2008, and again increased slightly [38].