Serious Mental Illness and Cardiovascular Health

Serious mental illness (SMI) is defined generally as a major mental or behavioral disorder, causing substantial impairment in multiple areas of daily functioning. SMI includes disorders such as schizophrenia and bipolar disorder, but not isolated substance abuse or developmental disorders. SMI affects about 4 to 8 percent of adults.^1–3 Individuals with SMI have shortened life expectancies relative to the general population to an extent that is not explained by suicide and accidents alone.^4,5 This population experiences higher rates of morbidity from multiple general medical conditions, including diabetes^6–8 and cardiovascular disease (CVD).^9–11 Among patients using the public mental health system, heart disease was the leading cause of death.¹² This excess of CVD-related mortality may be due to a number of factors including direct effects of the illness, medications used to treat SMI, modifiable behavioral risk factors, and disparities in access and quality of health care.

For CVD, mental illness may be an independent risk factor that acts both directly through physiological effects such as underlying genetic vulnerabilities, or indirectly through effects on an individual’s access to or interaction with the health care system.^13–15 Modifiable CVD risk factors, such as smoking,¹⁶ obesity,^17,18 and physical inactivity^19,20 are highly prevalent among adults with SMI. Adverse effects of psychotropic drugs (notably second-generation antipsychotics) also may contribute to the development of CVD by increasing the risk of conditions such as hyperglycemia, hyperlipidemia, and obesity.²¹ Lower socioeconomic status is more common in individuals with SMI^22,23 and may limit access to healthy food, opportunities for physical exercise (e.g., walkable neighborhoods and access to fitness facilities), and high-quality medical care. Numerous studies have demonstrated disparities in the quality of general medical care provided to individuals with SMI.^24–28 Given these issues, identifying intervention strategies that address CVD risk in individuals with SMI is a pressing priority to avoid early morbidity and mortality.

Current Treatment Approaches

Managing CVD risk in individuals with SMI includes standard pharmacological and behavioral interventions used in the general population (Table 1) as well as treatments specific to this population (e.g., antipsychotic medication–switching to manage adverse effects). Multicondition lifestyle interventions such as combinations of physical activity promotion and nutrition counseling with medical management of chronic medical conditions (e.g., hyperlipidemia) may be used to manage CVD risk factors in individuals with SMI. In addition, peer support interventions have been used to improve mental health outcomes and show promise in improving general medical outcomes;³³ family interventions may have this potential as well. However, interventions and treatments used to improve CVD risk may vary importantly in efficacy, adverse effects, complexity of regimen, need for monitoring, costs, and potential for drug-drug and drug-disease interactions.

Table 1

Selected pharmacological treatments and behavioral strategies to manage CVD risk factors.

The efficacy of most pharmacological agents used to reduce CVD risk is expected to be similar in patients with SMI when compared with general populations, but the potential for more severe or higher frequency adverse effects may be greater in individuals with SMI than in general populations due to drug-drug interactions (e.g., thiazides and lithium) or drug-disease interactions (e.g., varenicline and mood disorders). For behavioral interventions, direct effects of SMI and the limited social and economic support systems often available to these individuals may decrease effectiveness. To be optimally effective, health behavior interventions used in the general population to manage CVD risk may benefit from customization to the context and needs of individuals with SMI. Given the broad range of potential interventions and uncertainty about the effectiveness of competing strategies, an evidence synthesis was requested to inform guidelines and policy decisions.

Scope of the Review

This comparative effectiveness review was funded by the Agency for Healthcare Research and Quality (AHRQ). The review was designed to evaluate strategies to improve CVD risk factors in adults with SMI. SMI has been defined variously by different groups over time.³⁴ For the purposes of this evidence review, people with SMI are defined as individuals who have (1) schizophrenia or schizoaffective disorder (or other related primary psychotic disorder), (2) bipolar disorder, or (3) current major depression with psychotic features. We also included studies that enrolled adults with SMI or severe and persistent mental illness (SPMI) but did not specify diagnoses. Individuals with a primary diagnosis of substance abuse, dementia, personality disorder, or mental retardation are excluded from this definition.

To prioritize interventions for review, we examined published systematic reviews of strategies to improve CVD risk factors in individuals with SMI and consulted with our Key Informants. Because we identified recent high-quality reviews of general health advice, interventions for smoking cessation, and models to provide integrated mental health–general medical care, we elected not to cover these interventions again in our review.^35–39 We included randomized controlled trials (RCTs) of the pharmacological and patient-focused behavioral strategies listed in Table 1, along with peer and family support interventions. For patient-level intervention strategies, RCTs yield the highest quality evidence. We included both active and control comparators. Major outcomes of interest for this report are primary CVD risk factors (excluding tobacco use as explained above), physical functioning or health-related quality of life, adverse effects, and all-cause mortality.

Key Questions

With input from our Technical Expert Panel, we constructed Key Questions (KQs) using the general approach of specifying the population of interest, interventions, comparators, outcomes, timing of outcomes, and settings (PICOTS; see the section on “Inclusion and Exclusion Criteria” in the Methods section for details). The draft KQs developed during this process were available for public comment from 28 October 2011 to 28 November 2011. Comments received led to revisions including the addition of a separate KQ for dyslipidemia and the inclusion of peer and family support interventions in the strategies examined for each KQ. The final KQs considered in this comparative effectiveness review were:

KQ 1.

What is the effectiveness of weight-management behavioral interventions (e.g., behavioral counseling, health education), peer or family support interventions, pharmacological treatments (e.g., orlistat, topiramate), antipsychotic medication–switching to an antipsychotic with a low or neutral impact on weight, or their combination on weight control and related physical health outcomes (e.g., health-related quality of life, mortality) compared with each other or with usual care (or other control) among adults with serious mental illness (SMI) who are overweight, obese, or taking antipsychotics?

KQ 2.

What is the effectiveness of diabetes-management behavioral interventions (e.g., behavioral counseling, health education), peer or family support interventions, pharmacological treatments (e.g., rosiglitazone, metformin), antipsychotic medication–switching to an antipsychotic with a low or neutral impact on glucose level, or their combination on glucose-level control and related physical health outcomes (e.g., health-related quality of life, mortality) compared with each other or with usual care (or other control) among adults with SMI who have diabetes or are taking antipsychotics?

KQ 3.

What is the effectiveness of dyslipidemia-management behavioral interventions (e.g., behavioral counseling, health education), peer or family support interventions, pharmacological treatments (e.g., statins), antipsychotic medication–switching to an antipsychotic with a low or neutral impact on lipid levels, or their combination on lipid-level control and related physical health outcomes (e.g., health-related quality of life, mortality) compared with each other or with usual care (or other control) among adults with SMI who have dyslipidemia or are taking antipsychotics?

KQ 4.

What is the effectiveness of multicondition lifestyle interventions (e.g., combinations of smoking cessation, physical activity, and nutrition counseling with or without medication management) on cardiovascular risk factors and related physical health outcomes (e.g., health-related quality of life, mortality) among adults with SMI who have cardiovascular disease, elevated cardiovascular risk (e.g., hypertension), or are taking antipsychotics?

Analytic Framework

Figure 1. shows the analytic framework for this systematic review.

Figure 1

Analytic framework. CVD = cardiovascular disease; KQ = Key Question

The population evaluated in this comparative effectiveness review is adults with SMI who also have at least one of the following conditions: are overweight or obese; have diabetes, dyslipidemia, or CVD; are at elevated CVD risk, or are taking antipsychotic medication and so are at elevated risk for obesity, diabetes, dyslipidemia, or CVD. Intervention strategies considered by the four KQs are (1) behavioral strategies, (2) peer and family support interventions, (3) pharmacological treatments, (4) combinations of behavioral and pharmacological interventions, (5) antipsychotic medication switching, and (6) multicondition lifestyle interventions. The intermediate outcomes considered are weight control, glucose levels, lipid levels, and CVD risk. The final outcomes considered are mortality, physical function, and health-related quality of life. All four KQs consider the adverse effects of treatment interventions.