Abstract
Objective:
To estimate 30-year CVD risk and modifiable risk factors in young adults with serious mental illness (SMI) versus those without, and assess variations in CVD risk by race, ethnicity, and sex.
Method:
In this cross-sectional study, we estimated and compared the Framingham 30-year CVD risk score and individual modifiable CVD risk factors in young adult (20-39 years) primary care patients with and without SMI at two US healthcare systems (January 2016-Septemeber 2018). Interaction terms assessed whether the SMI-risk association differed across demographic groups.
Results:
Covariate-adjusted 30-year CVD risk was significantly higher for those with (n=4228) versus those without (n=155,363) SMI (RR 1.28, 95% CI [1.26, 1.30]). Patients with SMI had higher rates of hypertension (OR 2.02 [1.7, 2.39]), diabetes (OR 3.14 [2.59, 3.82]), obesity (OR 1.93 [1.8, 2.07]), and smoking (OR 4.94 [4.6, 5.36]). The increased 30-year CVD risk associated with SMI varied significantly by race and sex: there was an 8% higher risk in Black compared to White patients (RR 1.08, [1.04, 1.12]) and a 9% lower risk in men compared to women (RR 0.91 [0.88, 0.94]).
Conclusions:
Young adults with SMI are at increased 30-year risk of CVD, and further disparities exist for Black individuals and women.
Keywords: schizophrenia, schizoaffective disorder, bipolar disorder, cardiometabolic disorders, health disparities
1.0. Introduction
Schizophrenia spectrum and bipolar disorders are serious mental illnesses (SMI) with a typical age of onset in late adolescence or early adulthood. Individuals with SMI face striking health disparities including premature mortality of up to 10-20 years [1]. Cardiovascular disease (CVD) is the leading driver of early mortality, with high rates of modifiable CVD risk factors, including obesity, smoking, diabetes, and hypercholesteremia [2-4]. Although increased CVD risk has been well-established for individuals with SMI, most studies have focused on risk in middle aged and older samples [5] who may have different risk profiles than young adults. Intervention trials aimed at reducing overall CVD risk or individual risk factors have shown promise but have also largely targeted middle aged patients [6-11], with few conducted in younger samples when early interventions could substantially reduce long term risk [12, 13]. Due to the early age of onset of SMI and the rapid accumulation of CVD risk factors after psychotropic treatment initiation [14-16], long-term CVD risk prediction is essential from the beginning of the illness to guide early intervention strategies for modifiable risk factors to reduce the burden of premature morbidity and mortality later in life.
Estimates of 30-year or lifetime CVD risk provide the most accurate prediction of long-term risk [17-19]. Only one prior study has characterized the Framingham 30-year CVD risk estimates in young adults with SMI, finding 30-year CVD risk to be twice as high in those recently diagnosed with bipolar disorder compared to healthy controls in a Danish sample. However absolute 30-year CVD risk scores were low in both groups [20]. Another previous report found that in adults with and without SMI, the largest differential lifetime CVD risk occurred in those under 40 years old [21]. The Framingham 30-year CVD risk scores and modifiable risk factors contributing to this increased risk in young adults with SMI remain to be elucidated.
As the burden of CVD risk factors is increasing among young adults in the US, especially for Black and Hispanic individuals [22], understanding risk factors for young adults with SMI is critical to inform prevention and intervention strategies for this at-risk population. Previous research was constrained by small sample sizes and lack of comparison groups that allow for adjustment of important patient characteristics. Additionally, CVD risk estimates for those with SMI reported by sociodemographic characteristics such as race, ethnicity, and sex have been scarce, largely limited to older patients with SMI, and results have been mixed [23, 24]. The combined effect of sociodemographic characteristics and SMI on 30 year-CVD risk has not been explored in young adults and could further inform intervention strategies.
We aimed to characterize continuous 30-year CVD risk and specific modifiable risk factors in a large sample of young adults aged 20-39 years with and without schizophrenia spectrum and bipolar disorders. Second, we explored differences in 30-year CVD risk by race, ethnicity, sex, and age to identify specific high-risk subgroups.
2.0. Methods
2.1. Study Design and Settings
Two large healthcare delivery organizations in Minnesota and Wisconsin, HealthPartners and Park Nicollet, participated in a larger trial of clinical decision support in primary care clinics aimed at reducing CVD risk for patients with SMI aged 18-75 years [11, 25]. The clinical decision support system is a web-based tool embedded within the electronic health record. The tool collects clinical data to generate evidenced based, patient specific treatment recommendations for addressing modifiable CVD risk factors, including blood pressure, lipids, glycated hemoglobin (HbA1c), smoking status, and body mass index (BMI). Study enrollment occurred between January 20, 2016 and September 19, 2018. The present study used data from index (baseline) visits for all patients who visited a randomized primary care clinic during the enrollment period of the randomized controlled trial. This study was approved by the HealthPartners Institutional Review Board, and a waiver of consent was granted.
2.2. Enrollment and Eligibility
Patients were included in the present study if they met the following criteria at the index visit: 1) aged 20 to 39 years, 2) not pregnant, 3) no active cancer, and 4) not residing in a nursing home. Patients younger than 20 years old were excluded because the 30-year CVD risk score used in this study has not been validated in those under 20 years old [17]. Patients were included in the SMI group if they had ≥2 outpatient or ≥1 inpatient diagnostic codes for SMI documented in the electronic health record (EHR; which included medical and mental health encounters) in the 2 years prior to index. Qualifying SMI diagnoses included bipolar disorder, schizophrenia, and schizoaffective disorder (see Supplemental Table S1 for International Classification of Disease (ICD) codes). Patients with ICD codes that crossed subcategories (i.e., bipolar disorder+schizophrenia) were assigned a diagnosis of schizoaffective disorder. Patients with schizophrenia or schizoaffective diagnoses were combined in the schizophrenia spectrum disorders group. Patients were included in the non-SMI group if they did not have any codes for SMI, depression, or anxiety.
2.3. Data Sources
Data were primarily collected via the clinical decision support tool, which recorded EHR data at each index visit, including age, race, sex, vital signs, diagnoses, and lab values. Ethnicity and insurance status were collected from the EHR data repository.
2.4. Measures
2.4.1. Estimated cardiovascular risk and cardiovascular risk factors.
Thirty-year CVD risk was estimated using a sex-specific multivariable algorithm, the Framingham 30-year risk score [17]. Recent evidence supports use of the Framingham 30-year risk models as the preferred method to identify at-risk adults under 40 years old [18, 19]. This score provides a continuous estimate of risk of a CVD event, which may be more clinically useful than categorical lifetime risk scores [18]. The full 30-year CVD risk score estimates the risk of both hard CVD events (coronary death, myocardial infarction, or stroke), and coronary insufficiency, angina pectoris, transient ischemic attack, intermittent claudication, or congestive heart failure occurring in the next 30 years. This score utilizes eight individual risk factors including age, sex, body mass index, systolic blood pressure, diabetes mellitus, antihypertensive prescription, and current smoking status. As antihypertensive medication orders were not available in the clinical decision system data for this secondary analysis, we used a hypertension diagnosis as a proxy for antihypertensive prescription. To assess the impact of this proxy measure, we also calculated the 30-year risk scores without the proxy (i.e., all patients coded “no” for antihypertensive treatment). Previous research suggests that most patients (approximately 85%) who are aware of their hypertension diagnosis are treated [26], with similarly high rates of concordance between hypertension diagnosis and treatment in the health system used in this study [27, 28]. We report results using the proxy measure, and in sensitivity analysis found no difference in the results when the proxy measure was excluded from the 30-year risk score calculations.
The clinical decision support system captured modifiable cardiovascular risk factors: blood pressure, lipids (total cholesterol, low-density lipoprotein (LDL), high-density lipoprotein (HDL), triglycerides, and statin use), HbA1c, weight (BMI), and smoking status. Blood pressure and weight were captured at the index encounter. Patients were considered to be treated with a statin if they had an active prescription for a statin at the index visit. For the remaining risk factors, the most recent value in the last 5 years was used for analyses.
2.4.2. Cardiometabolic Diagnoses.
Diagnoses for coronary heart disease, CVD, hypertension, and diabetes were defined as having ≥2 outpatient or ≥1 inpatient codes documented in the EHR in the 2 years before the index date (see Supplemental Table S2 for included ICD codes).
2.5. Statistical Analysis
Descriptive statistics were used to characterize unadjusted differences in demographics, CVD risk factors, and estimated 30-year CVD risk among patients with and without SMI and by SMI subtype. Chi-square analyses were used to assess the significance of unadjusted group differences in categorical variables, and general linear models for continuous variables with approximately normal distributions. Models were then adjusted for baseline demographics, including age, sex, race, ethnicity, and insurance status. For adjusted models with categorical dependent variables, binary, ordinal, or multinomial logistic regression models were used as appropriate. Generalized Poisson regression models estimated unadjusted and covariate-adjusted group differences in 30-year CVD risk. Interaction terms between demographic categories and SMI status were added to these models to estimate associations between SMI and 30-year CVD risk within demographic groups, and to assess whether the SMI-risk association was differential across demographic groups. As the data used in this study were collected as part of usual care, an absent value for a laboratory test was assumed to indicate that the test was not performed or collected, rather than performed but not entered in the EHR or missed by the clinical decision support system. Chi-square tests were used to assess differential patterns of absent data. Absent data were not imputed and listwise deletion was used in regression analyses.
3.0. Results
A total of 997 individuals with a schizophrenia spectrum disorder, 3231 with bipolar disorder and 155,363 without SMI were included in the analysis (see Supplementary Figure S1 for participant flow diagram). Patients with SMI were slightly older and were more likely to be female, to self-identify as Black, Native-American or with multiple races, and to be insured by Medicaid (Table 1). Compared to patients with bipolar disorder, patients with schizophrenia spectrum disorders were more likely to be male, to identify as Black, and to be insured by Medicaid. Patients without SMI were more likely to have absent data for lipids and BMI (Supplemental Table S3). As BMI is included in the 30-year CVD risk equation, patients without SMI were also more likely to have absent 30-year risk scores compared to those with SMI (18.4% vs. 12.3% absent).
Table 1.
Demographics, 30-year Cardiovascular Risk, and Individual Modifiable Cardiovascular Risk Factors between Patients with and Without SMI
| Patient characteristic |
Bipolar Disorder | Schizophrenia Spectrum | SMI | No SMI | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n = 3231 (2.0%) | n = 997 (0.6%) | n = 4228 (2.7%) | n = 155,363 (97.4%) | ||||||||||||||
| n | % | M | SD | n | % | M | SD | n | % | M | SD | n | % | M | SD |
p (SMI
vs. No- SMI) |
|
| Age | 30.6 | 5.4 | 30.7 | 5.4 | 30.6 | 5.4 | 29.7 | 5.6 | <.0001 | ||||||||
| Age, categorical | <.0001 | ||||||||||||||||
| 20-29 | 1335 | 41.3% | 400 | 40.1% | 1735 | 41.0% | 75094 | 48.3% | |||||||||
| 30-39 | 1896 | 58.7% | 597 | 59.9% | 2493 | 59.0% | 80269 | 51.7% | |||||||||
| Female | 2084 | 64.5% | 359 | 36.0% | 2443 | 57.8% | 79186 | 51.0% | <.0001 | ||||||||
| Race | <.0001 | ||||||||||||||||
| Asian | 69 | 2.1% | 75 | 7.5% | 144 | 3.4% | 12818 | 8.3% | |||||||||
| Black | 396 | 12.3% | 280 | 28.1% | 676 | 16.0% | 21387 | 13.8% | |||||||||
| Native American/Alaska Native | 21 | 0.7% | 11 | 1.1% | 32 | 0.8% | 462 | 0.3% | |||||||||
| Multiple | 39 | 1.2% | 16 | 1.6% | 55 | 1.3% | 971 | 0.6% | |||||||||
| Other | 27 | 0.8% | 10 | 1.0% | 37 | 0.9% | 3690 | 2.4% | |||||||||
| Unknown | 125 | 3.9% | 28 | 2.8% | 153 | 3.6% | 14174 | 9.1% | |||||||||
| White | 2554 | 79.1% | 577 | 57.9% | 3131 | 74.1% | 101861 | 65.6% | |||||||||
| Ethnicity | <.0001 | ||||||||||||||||
| Hispanic | 119 | 3.7% | 32 | 3.2% | 151 | 3.6% | 6757 | 4.4% | |||||||||
| Non-Hispanic | 2705 | 83.7% | 841 | 84.4% | 3546 | 83.9% | 106807 | 68.8% | |||||||||
| Unknown | 407 | 12.6% | 124 | 12.4% | 531 | 12.6% | 41799 | 26.9% | |||||||||
| Insurance Type | <.0001 | ||||||||||||||||
| Self-pay/Uninsured | 87 | 2.7% | 17 | 1.7% | 104 | 2.5% | 4944 | 3.1% | |||||||||
| Medicare Only | 54 | 1.7% | 46 | 4.6% | 100 | 2.4% | 167 | 62.6% | |||||||||
| Medicaid Only | 1154 | 35.7% | 405 | 40.6% | 1559 | 36.9% | 25770 | 16.6% | |||||||||
| Commercial Only | 1026 | 31.8% | 95 | 9.5% | 1121 | 26.5% | 104776 | 67.4% | |||||||||
| Other Only | 27 | 0.8% | 5 | 0.5% | 32 | 0.8% | 2242 | 1.4% | |||||||||
| Medicare + Medicaid | 241 | 7.5% | 250 | 25.1% | 491 | 11.6% | 703 | 0.5% | |||||||||
| Medicare + Commercial | 16 | 0.5% | 13 | 1.3% | 29 | 0.7% | 77 | 0.1% | |||||||||
| 2 or more insurances | 626 | 19.0% | 166 | 16.7% | 792 | 18.7% | 16684 | 10.7% | |||||||||
| 30-year riska (n=130,276) | 2818 | 14.3 | 11.5 | 868 | 18.6 | 13.9 | 3686 | 15.3 | 12.2 | 126590 | 10.8 | 9.0 | <.0001 | ||||
| CVD | 20 | 0.6% | 3 | 0.3% | 23 | 0.5% | 204 | 0.1% | <.0001 | ||||||||
| Blood Pressure | |||||||||||||||||
| HTN | 142 | 4.4% | 57 | 5.7% | 199 | 4.7% | 2773 | 1.8% | |||||||||
| SBP (n = 159546) | 118.3 | 14.1 | 118.4 | 13.5 | 118.3 | 13.9 | 119.3 | 14.4 | <.0001 | ||||||||
| DBP (n = 159548) | 76.4 | 10.9 | 76.6 | 10.8 | 76.4 | 10.9 | 75.3 | 10.9 | <.0001 | ||||||||
| Cholesterol | |||||||||||||||||
| Total cholesterol (n = 43061) | 178.2 | 37.1 | 175.9 | 38.2 | 177.6 | 37.4 | 177 | 35.2 | 0.47 | ||||||||
| LDL (statin only) (n = 1323) | 102.1 | 39.4 | 94.6 | 31.5 | 98.4 | 35.8 | 118.3 | 46.8 | <.0001 | ||||||||
| LDL (non-statin only) (n = 37768) | 103.7 | 30.1 | 104.3 | 33.9 | 103.9 | 31.1 | 105.3 | 30.2 | 0.04 | ||||||||
| HDL (n = 43107) | 48.4 | 15.7 | 42.9 | 12.7 | 46.8 | 15.2 | 50.4 | 14.6 | <.0001 | ||||||||
| Triglycerides (n = 37904) | 140 | 109 | 155.3 | 122.1 | 144.3 | 113.1 | 114 | 92.3 | <.0001 | ||||||||
| Statin use | 94 | 2.9% | 80 | 8.0% | 174 | 4.1% | 1462 | 0.9% | <.0001 | ||||||||
| Glucose | |||||||||||||||||
| DM | 101 | 3.1% | 67 | 6.7% | 168 | 4.0% | 1290 | 0.8% | <.0001 | ||||||||
| A1c (DM only)b (n=1330) | 7.9 | 2.2 | 7.5 | 2.1 | 7.72 | 2.2 | 8.1 | 2.3 | 0.06 | ||||||||
| Weight | |||||||||||||||||
| BMI (n=130532) | 30.7 | 8.3 | 31.5 | 7.9 | 30.89 | 8.2 | 27.6 | 6.5 | <.0001 | ||||||||
| BMI, categorical (n=130532) | <.0001 | ||||||||||||||||
| <18.5, underweight | 41 | 1.4% | 11 | 1.3% | 52 | 1.4% | 2282 | 1.8% | |||||||||
| 18.5-24.9, Normal | 748 | 26.4% | 181 | 20.8% | 929 | 25.0% | 48578 | 38.3% | |||||||||
| 25-29.9, Overweight | 766 | 27.0% | 218 | 25.0% | 984 | 26.5% | 40533 | 32.0% | |||||||||
| 30-34.9, Obese I | 545 | 19.2% | 214 | 24.6% | 759 | 20.5% | 19909 | 15.7% | |||||||||
| 35-39.9, Obese II | 350 | 12.3% | 129 | 14.8% | 479 | 12.9% | 8963 | 7.1% | |||||||||
| ≥40, Obese III | 389 | 13.7% | 118 | 13.6% | 507 | 13.7% | 6557 | 5.2% | |||||||||
| Smoking status (n = 159573)f | <.0001 | ||||||||||||||||
| Current smoker | 1288 | 39.9% | 498 | 49.7% | 1783 | 42.2% | 19721 | 12.7% | |||||||||
| Former smoker | 768 | 23.8% | 173 | 17.4% | 941 | 22.3% | 18498 | 11.9% | |||||||||
| Nonsmoker | 1175 | 36.4% | 329 | 33.0% | 1504 | 35.6% | 117126 | 75.4% | |||||||||
Note. 30-year risk = Framingham 30-year cardiovascular disease risk score; BMI = body mass index; CVD = cardiovascular disease; DBP = diastolic blood pressure; DM = diabetes mellitus; HDL = high density lipoprotein; HTN = hypertension; LDL = low density lipoprotein; LL = lower 95% confidence limit; M = predicted mean; P% = predicted percent; SBP = systolic blood pressure; UL = upper 95% confidence limit
30-year cardiovascular disease risk score only calculated for those without known CVD and with recorded age, sex, body mass index, systolic blood pressure, diabetes mellitus, hypertension status, and current smoking status (n = 130,276)
Calculated for patients with DM who have available A1c tests within the last 5 years (n = 1,330)
3.1. Unadjusted differences in estimated 30-year CVD risk and modifiable CVD risk factors
3.1.1. SMI compared to non-SMI.
Patients with SMI had significantly higher estimated 30-year CVD risk scores (mean 15.3%, SD 12.2%) compared to those without SMI (mean 10.8%, SD 9%, p < 0.001)) (Table 1). Regarding individual CVD risk factors, patients with SMI had higher rates of obesity (BMI ≥30) (47.1% vs. 28%), smoking (42.2% vs. 12.7%), hypertension (4.7 vs 1.8%) and diabetes (4% vs. 0.8%) than those without SMI. Patients with SMI were more likely to be treated with a statin (4.1% vs. 0.9%) and had higher triglycerides and HDL cholesterol, but not total cholesterol.
3.1.2. Specific SMI diagnoses.
CVD risk was generally higher in the schizophrenia spectrum group compared to the bipolar disorder group (Table 1). For example, mean estimated 30-year cardiovascular risk was 18.6% (SD 13.9%) for patients with schizophrenia spectrum disorder compared to 14.3% (SD 11.5%) for patients with bipolar disorder. Schizophrenia spectrum disorder patients had higher rates of hypertension (5.7% vs 4.4%), diabetes (6.7% vs. 3.1%), obesity (53.0% vs 45.3%), and smoking (49.7% vs 39.9%) compared to patients with bipolar disorder.
3.2. Adjusted differences in estimated 30-year CVD risk and modifiable risk factors
3.2.1. SMI compared to non-SMI.
After adjusting for age, race, ethnicity, sex, and insurance type, differences in CVD risk profiles between patients with and without SMI persisted (Table 2).
Table 2.
Adjusted Estimates of 30-year Cardiovascular Risk and Individual Modifiable Cardiovascular Risk Factors between Patients with and Without SMI
| Patient characteristic |
Bipolar Disorder | Schizophrenia Spectrum | SMI | No SMI | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n = 3231 (2.0%) | n = 997 (0.6%) | n = 4228 (2.7%) | n = 155,363 (97.4%) | |||||||||||||
| P% | M |
95%
LL |
95%
UL |
P% | M |
95%
LL |
95%
UL |
P% | M |
95%
LL |
95%
UL |
PP | M |
95%
LL |
95%
UL |
|
| 30-year riska (n=130,276) | 11.9 | 11.7 | 12.1 | 12.3 | 11.9 | 12.6 | 12.0 | 11.8 | 12.2 | 9.3 | 9.3 | 9.4 | ||||
| Blood Pressure | ||||||||||||||||
| HTN | 2.4% | 2.0% | 2.9% | 2.0% | 1.5% | 2.7% | 2.3% | 1.9% | 2.7% | 1.1% | 1.1% | 1.2% | ||||
| SBP (n = 159546) | 120.1 | 119.6 | 120.5 | 118.3 | 117.5 | 119.2 | 119.7 | 119.3 | 120.1 | 119.3 | 119.2 | 119.4 | ||||
| DBP (n = 159548) | 76.7 | 76.3 | 77.0 | 76.1 | 75.5 | 76.8 | 76.6 | 76.2 | 76.9 | 75.3 | 75.2 | 75.3 | ||||
| Cholesterol | ||||||||||||||||
| Total cholesterol (n = 43061) | 179.8 | 178.1 | 181.5 | 176.8 | 174.2 | 179.4 | 179.0 | 177.6 | 180.4 | 177.2 | 176.8 | 177.5 | ||||
| LDL (statin only) (n = 1323) | 103.9 | 93.7 | 114.1 | 93.5 | 82.3 | 103.8 | 98.7 | 91.5 | 106.0 | 118.2 | 115.6 | 120.8 | ||||
| LDL (non-statin only) (n = 37768) | 106.5 | 104.3 | 105.4 | 104.3 | 101.9 | 106.7 | 105.9 | 104.6 | 107.2 | 105.4 | 105.0 | 105.7 | ||||
| HDL (n = 43107) | 46.2 | 45.6 | 46.9 | 44.5 | 43.4 | 45.5 | 45.7 | 45.2 | 46.3 | 50.2 | 50.1 | 50.4 | ||||
| Triglycerides (n = 37904) | 146.8 | 142.4 | 151.2 | 155.6 | 148.6 | 162.6 | 149.3 | 145.6 | 153.1 | 114.2 | 113.2 | 115.2 | ||||
| Statin use | 1.2% | 0.9% | 1.5% | 1.6% | 1.2% | 2.2% | 1.3% | 1.1% | 1.6% | 0.5% | 0.4% | 0.5% | ||||
| Glucose | ||||||||||||||||
| DM | 1.3% | 1.1% | 1.7% | 1.6% | 1.2% | 2.1% | 1.4% | 1.2% | 1.7% | 0.5% | 0.4% | 0.5% | ||||
| A1c (DM only)b (n=1330) | 8.1 | 7.6 | 8.6 | 7.4 | 6.8 | 8.0 | 7.8 | 7.4 | 8.2 | 8.1 | 7.9 | 8.2 | ||||
| Weight | ||||||||||||||||
| BMI | 30.5 | 30.3 | 30.8 | 31.1 | 30.7 | 31.6 | 30.7 | 30.5 | 30.9 | 27.6 | 27.6 | 27.7 | ||||
| BMI, categorical | ||||||||||||||||
| <18.5, underweight | 0.9% | 0.8% | 1.0% | 0.8% | 0.7% | 0.9% | 0.9% | 0.8% | 0.9% | 1.7% | 1.6% | 1.8% | ||||
| 18.5-24.9, Normal | 24.9% | 23.6% | 26.2% | 23.5% | 23.4% | 23.6% | 24.5% | 23.3% | 25.6% | 38.3% | 38.2% | 38.3% | ||||
| 25-29.9, Overweight | 32.7% | 31.4% | 33.9% | 32.2% | 29.9% | 34.5% | 32.5% | 31.3% | 33.6% | 32.9% | 32.6% | 33.2% | ||||
| 30-34.9, Obese I | 21.4% | 19.8% | 23.1% | 22.0% | 19.0% | 25.1% | 21.6% | 20.1% | 23.1% | 15.5% | 15.2% | 15.7% | ||||
| 35-39.9, Obese II | 11.2% | 10.1% | 12.3% | 11.8% | 9.7% | 13.9% | 11.4% | 10.4% | 12.4% | 6.8% | 6.6% | 6.9% | ||||
| ≥40, Obese III | 9.0% | 8.4% | 9.6% | 9.6% | 8.6% | 10.8% | 9.2% | 8.7% | 9.8% | 4.8% | 4.8% | 5.0% | ||||
| BMI, binary | ||||||||||||||||
| >= 30, obese | 41.4% | 39.6% | 43.3% | 45.4% | 41.9% | 49.0% | 42.3% | 40.6% | 44.0% | 27.5% | 27.3% | 27.8% | ||||
| Smoking status | ||||||||||||||||
| Current smoker | 33.4% | 31.7% | 35.1% | 34.2% | 31.1% | 37.2% | 33.6% | 32.1% | 35.1% | 11.9% | 11.7% | 12.1% | ||||
| Former smoker | 23.2% | 21.7% | 24.8% | 19.7% | 16.9% | 22.5% | 22.6% | 21.2% | 24.0% | 11.0% | 10.8% | 11.1% | ||||
| Nonsmoker | 43.3% | 41.5% | 45.2% | 46.1% | 42.6% | 45.2% | 43.9% | 42.2% | 45.5% | 77.1% | 76.9% | 77.4% | ||||
Note. 30-year risk = Framingham 30-year cardiovascular disease risk score; BMI = body mass index; CVD = cardiovascular disease; DBP = diastolic blood pressure; DM = diabetes mellitus; HDL = high density lipoprotein; HTN = hypertension; LDL = low density lipoprotein; LL = lower 95% confidence limit; M = predicted mean; P% = predicted percent; SBP = systolic blood pressure; UL = upper 95% confidence limit
30-year cardiovascular disease risk score only calculated for those without known CVD and with recorded age, sex, body mass index, systolic blood pressure, diabetes mellitus, hypertension status, and current smoking status (n = 130,276)
Calculated for patients with DM who have available A1c tests within the last 5 years (n = 1,330)
Patients with SMI had an adjusted 30-year CVD risk score of 12.0% (95% CI = 11.8, 12.2%) compared to 9.3% (95% CI = 9.3, 9.4%) for patients without SMI, resulting in a 28% higher relative risk (95% CI = 1.26, 1.30) (Table 3).
Table 3.
Adjusted Models Predicting Estimated 30-year Cardiovascular Risk and Individual Modifiable Cardiovascular Risk Factors between Patients with and Without SMI
| Patient characteristic | SMI vs. No SMI | Bipolar Disorder vs. No SMI | Schizophrenia Spectrum vs. No SMI | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| OR | b |
95%
LL |
95%
UL |
p | OR | b |
95%
LL |
95%
UL |
p | OR | b |
95%
LL |
95%
UL |
p | |
| 30-year riska (Risk Ratio) | 1.28 | 1.26 | 1.30 | <.0001 | 1.27 | 1.25 | 1.29 | <.0001 | 1.31 | 1.28 | 1.35 | <.0001 | |||
| CVD | 1.78 | 1.08 | 2.95 | 0.025 | 2.24 | 1.34 | 3.72 | 0.002 | 0.65 | 0.20 | 2.17 | 0.48 | |||
| Blood Pressure | |||||||||||||||
| HTN | 2.02 | 1.70 | 2.39 | <.0001 | 2.11 | 1.75 | 2.54 | <.0001 | 1.77 | 1.30 | 2.40 | 0.0003 | |||
| SBP | 0.37 | −0.05 | 0.79 | 0.09 | 0.75 | 0.27 | 1.22 | 0.002 | −0.99 | −1.84 | −0.13 | 0.024 | |||
| DBP | 1.28 | 0.94 | 1.62 | <.0001 | 1.39 | 1.02 | 1.78 | <.0001 | 0.88 | 0.19 | 1.57 | 0.012 | |||
| Cholesterol | |||||||||||||||
| Total cholesterol | 1.82 | 0.40 | 3.24 | 0.012 | 2.66 | 1.01 | 4.33 | 0.002 | −0.35 | −2.97 | 2.27 | 0.79 | |||
| LDL (statin only) | −19.46 | −27.19 | −11.74 | <.0001 | −14.28 | −24.87 | −3.69 | 0.008 | −24.64 | −35.22 | −14.05 | <.0001 | |||
| LDL (non-statin only) | 0.54 | −0.74 | 1.82 | 0.41 | 1.13 | −0.36 | 2.61 | 0.14 | −1.04 | −3.45 | 1.37 | 0.40 | |||
| HDL | −4.50 | −5.06 | −3.93 | <.0001 | −4.01 | −4.67 | −3.35 | <.0001 | −5.76 | −6.81 | −4.72 | <.0001 | |||
| Triglycerides | 35.08 | 31.24 | 38.93 | <.0001 | 32.60 | 28.06 | 37.07 | <.0001 | 41.38 | 34.34 | 48.42 | <.0001 | |||
| Statin use | 2.90 | 2.38 | 3.53 | <.0001 | 2.58 | 2.05 | 3.26 | <.0001 | 3.64 | 2.72 | 4.89 | <.0001 | |||
| Glucose | |||||||||||||||
| DM | 3.14 | 2.59 | 3.82 | <.0001 | 2.99 | 2.39 | 3.74 | <.0001 | 3.50 | 2.59 | 4.73 | <.0001 | |||
| A1c (DM only)b | −0.29 | −0.67 | 0.10 | 0.15 | 0.00 | −0.49 | 0.50 | 0.99 | −0.69 | −1.25 | −0.11 | 0.02 | |||
| Weight | |||||||||||||||
| BMI | 3.03 | 2.82 | 3.24 | <.0001 | 2.89 | 2.65 | 3.13 | <.0001 | 3.49 | 3.06 | 3.92 | <.0001 | |||
| BMI, categorical | 1.96 | 1.82 | 2.09 | <.0001 | 1.91 | 1.79 | 2.05 | <.0001 | 2.07 | 1.82 | 2.34 | ||||
| BMI, binary | 1.93 | 1.80 | 2.07 | <.0001 | 1.86 | 1.72 | 2.01 | <.0001 | 2.19 | 1.90 | 2.53 | <.0001 | |||
| Smoking status | <.0001 | ||||||||||||||
| Current smoker | 4.97 | 4.60 | 5.36 | <.0001 | 5.00 | 4.59 | 5.45 | <.0001 | 4.80 | 4.13 | 5.59 | <.0001 | |||
| Former smoker | 3.62 | 3.31 | 3.96 | <.0001 | 3.77 | 3.43 | 4.16 | <.0001 | 3.00 | 2.47 | 3.67 | <.0001 | |||
| Nonsmoker | REF | REF | REF | ||||||||||||
Note. 30-year risk = Framingham 30-year cardiovascular disease risk score; BMI = body mass index; CVD = cardiovascular disease; DBP = diastolic blood pressure; DM = diabetes mellitus; HDL = high density lipoprotein; HTN = hypertension; LDL = low density lipoprotein; LL = lower 95% confidence limit; M = predicted mean; P% = predicted percent; SBP = systolic blood pressure; UL = upper 95% confidence limit
30-year cardiovascular disease risk score only calculated for those without known CVD and with recorded age, sex, body mass index, systolic blood pressure, diabetes mellitus, hypertension status, and current smoking status (n = 130,276)
Calculated for patients with DM who have available A1c tests within the last 5 years (n = 1,330)
Patients with SMI had greater odds of being diagnosed with hypertension (adjusted prevalence 2.3% vs. 1.1%; OR 2.02 [95% CI = 1.7, 2.39]) and diabetes (1.4% vs 0.5%; OR 3.14 [95% CI = 2.59, 3.82]) compared to patients without SMI. There were no significant differences in A1c values for those with diabetes.
There were no differences in total cholesterol between patients with and without SMI; however, patients with SMI had lower HDL and higher triglycerides than patients without SMI (HDL: b = −4.5 [95% CI = −5.06, −3.93]; triglycerides: b = 35.06 [95% CI = 31.24, 38.93]).
High rates of obesity and smoking persisted for those with SMI in the adjusted models. The adjusted mean BMI for those with SMI was 30.7 (95% CI = 30.5, 30.9) compared to 27.6 95% CI = 27.6, 27.7) in those without SMI. Patients with SMI were more likely to have obesity (BMI ≥30) (42.3% vs. 27.5%, OR 1.93 [95% CI = 1.8, 2.07]) and were nearly five times more likely to be current smokers (OR 4.94 [95% CI = 4.6, 5.36]).
3.2.2. Specific SMI diagnoses.
As with unadjusted comparisons, patients with schizophrenia spectrum diagnosis continued to have higher CVD risk profiles than those with bipolar disorder (Table 3). Schizophrenia spectrum disorder was associated with a 31% higher CVD risk (95% CI = 1.28, 1.35), and bipolar disorder with a 27% higher CVD risk (95% CI = 1.25, 1.29), compared to those without SMI.
3.3. Adjusted estimated 30-year CVD risk among patient subgroups
We next examined whether the increased risk associated with having SMI varied across demographic subgroups, including race, ethnicity, sex, and age (Table 4).
Table 4.
Adjusted Estimated 30-year Cardiovascular Risk by Patient Subgroups and Interaction Between SMI Status and Subgroup
| SMI | No-SMI | Adjusted Risk Ratio for 30-year CVD risk score (SMI vs. No-SMI) |
Adjusted Risk Ratio for 30-year CVD risk score (subgroup vs. reference) |
|||
|---|---|---|---|---|---|---|
| Adjusted Mean (95% CI) |
Adjusted Mean (95% CI) |
RR (95% CI) | p | RR (95% CI) | p | |
| Overall | 12.0 (11.8 – 12.2) | 9.3 (9.3 – 9.4) | 1.28 (1.26 – 1.30) | <0.001 | ||
| RACE * | ||||||
| Asian | 11.0 (10.2 – 11.9) | 8.6 (8.6 – 8.7) | 1.27 (1.18 – 1.38) | <0.001 | 1.01 (0.93 – 1.09) | 0.84 |
| Black | 12.7 (12.3 – 13.2) | 9.4 (9.3 – 9.4) | 1.36 (1.32 – 1.41) | <0.001 | 1.08 (1.04 – 1.12) | <0.001 |
| Multiple | 13.0 (11.4 – 14.9) | 9.3 (9.0 – 9.7) | 1.39 (1.21 – 1.60) | <0.001 | 1.10 (0.96 – 1.27) | 0.17 |
| White | 11.9 (11.7 – 12.1) | 9.4 (9.4 – 9.5) | 1.26 (1.24 – 1.29) | <0.001 | Reference | |
| ETHNICITY ** | ||||||
| Hispanic | 12.4 (11.5 – 13.4) | 9.8 (9.6 – 9.9) | 1.27 (1.18 – 1.38) | <0.001 | 0.99 (0.91 – 1.07) | 0.71 |
| non-Hispanic | 12.1 (11.9 – 12.3) | 9.3 (9.3 – 9.4) | 1.29 (1.27 – 1.32) | <0.001 | Reference | |
| Unknown | 11.2 (10.8 – 11.7) | 9.3 (9.2 – 9.4) | 1.21 (1.16 – 1.26) | <0.001 | 0.93 (0.89 - 0.98) | 0.002 |
| SEX *** | ||||||
| Women | 9.2 (9.0 – 9.4) | 6.8 (6.8 – 6.8) | 1.35 (1.32 – 1.38) | <0.001 | Reference | |
| Men | 16.0 (15.7 – 16.3) | 13.0 (13.0 – 13.1) | 1.23 (1.21 – 1.26) | <0.001 | 0.91 (0.88 – 0.94) | <0.001 |
| AGE GROUP | ||||||
| 20-29 | 7.7 (7.5 – 7.9) | 6.0 (6.0 – 6.0) | 1.28 (1.25 – 1.32) | <0.001 | 1.00 (0.97 – 1.03) | 0.98 |
| 30-39 | 18.1 (17.8 – 18.4) | 14.1 (14.0 – 14.1) | 1.28 (1.26 – 1.30) | <0.001 | Reference | |
Note. *Race x SMI interaction, p = 0.001
Hispanic x SMI interaction p < 0.01
Sex x SMI interaction, p < 0.001
3.3.1. Race.
The interaction between SMI diagnosis and race was assessed for groups with sufficient sample sizes, including Asian, Black, multiple races, and White. Within each race category, patients with SMI had greater estimated 30-year CVD risk than those without SMI. Specifically, 30-year CVD risk was increased for patients with SMI by 27% (95% CI = 1.18, 1.38) for Asian patients, 36% (95% CI = 1.32, 1.41) for Black patients, 39% (95% CI = 1.21, 1.60) for patients who identified with multiple races, and 26% (95% CI = 1.24, 1.29) for White patients.
The omnibus interaction between SMI status and race on 30-year CVD risk was significant (p=0.001). The association between SMI and 30-year CVD risk was 8% higher for Black relative to White patients (RR = 1.08 [95% CI = 1.04, 1.12]). No differential relationship between SMI and 30-year CVD risk was found among patients who self-identified as Asian or with multiple races.
3.3.2. Ethnicity.
Patients with SMI of both Hispanic and non-Hispanic ethnicity had higher 30-year CVD risk compared to patients of the same ethnicity without SMI (Hispanic: RR 1.27 [95% CI = 1.18, 1.38]; non-Hispanic RR 1.29 [95% CI = 1.27, 1.32]). The relationship between SMI and 30-year CVD risk was sufficiently smaller (RR 1.21 [95% CI = 1.16, 1.26]) among patients with unknown ethnicity that the comparison to non-Hispanic patients was statistically significant (RR 0.93, [95% CI = 0.89, 0.98]) as was the omnibus interaction (p<0.01).
3.3.3. Sex.
Both women and men with SMI had higher 30-year CVD risk compared to women and men without SMI, respectively (women: RR 1.35 [95% CI = 1.32, 1.38], men RR 1.23 [95% CI = 1.21, 1.26]). The increased risk associated with having SMI varied by sex, with a 9% lower association between SMI and 30-year CVD risk among men relative to women (RR 0.91 [95% CI = 0.88, 0.94], p<0.001).
3.3.4. Age group.
Patients with SMI in both the 20-29 and 30-39-year age groups had higher 30-year CVD risk compared to patients without SMI in each respective age group. Although 30-39-year-old patients with SMI had a 30-year CVD risk of 18.6% [95% CI = 18.2, 19.1] compared to 7.6% [95% CI = 7.4, 7.9] for 20-29-year-old patients, the increased risk associated with SMI did not vary by age group.
4.0. Discussion
In this cross-sectional study comparing 4,228 young adults with SMI to 155,363 without, the estimated 30-year risk of a cardiovascular event was 28% higher for patients with SMI, and patients with SMI had higher rates of modifiable CVD risk factors. The increased 30-year CVD risk was more pronounced for Black individuals and women with SMI, highlighting possible intersectional health disparities. This is the largest study of CVD risk specifically focused on young adults with SMI and one of only a few to characterize long-term CVD risk in this population. While 30-year risk estimates are more accurate in young adults than the more commonly used 10-year profiles [19], even these estimates may underestimate the true long-term risk associated with SMI because they were not developed in SMI populations [29, 30]. Our results underscore the need for more effective strategies to prevent and manage CVD starting at younger ages for these patients.
High rates of obesity and smoking drove the increased 30-year CVD risk in our sample of young adults with SMI. In fully adjusted models, obesity rates were 42.3% for patients with SMI compared to 27.5% for those without. These rates are similar to those in middle aged and older samples of patients with SMI, yet substantially higher than typically found in recently diagnosed patients [31]. Previous research points to a trajectory of substantial initial weight gain following SMI diagnosis and psychotropic medication initiation, followed by gradual gain over the next 10-20 years [15, 32]. Accordingly, this could be a driver of the large difference in estimated 30-year CVD risk that we observed between those with SMI aged 20-29 (7.7%) compared to 30-39 years (18.1%). Patients with SMI also had significantly higher rates of diabetes (4% vs. 0.8%) and hypertension (4.7% vs. 1.8%). Although elevated, these rates are lower than in samples of older patients with SMI, where rates of diabetes may exceed 15% [33, 34]. As early weight gain predicts longer-term risk for obesity for patients with SMI [32, 35], prevention of weight gain in the first years of SMI treatment could have significant impacts on long-term health trajectories, including diabetes prevention. These results emphasize the importance of obesity prevention strategies for patients with SMI, including behavioral interventions [13], use of lower metabolic-risk antipsychotic medications [36, 37], and early initiation of adjunctive medications that may attenuate antipsychotic induced weight gain [38, 39]. Additionally, smoking is likely the largest driver of early mortality for patients with SMI [40], yet safe and effective evidenced-based smoking cessation treatments are grossly underutilized in this population [41].
We observed significant variation in the 30-year CVD risk associated with having SMI across race and sex subgroups. The increased risk associated with SMI was 8% higher for Black compared to White patients. Although 30-year CVD risk was similar among Black and White patients without SMI in our sample, there is well-established evidence of the impact of structural racism on CVD risk for Black Americans [42, 43]. Additionally, individuals with SMI experience marginalization, social disadvantage, discrimination, and stigma, all of which are related to poorer health [44, 45]. Our results add to the small literature suggesting a possible “double jeopardy” for CVD risk for Black individuals living with SMI [44]. A review of US studies with small sample sizes suggested that Black individuals with SMI were at higher risk for obesity and diabetes [23]. One study found no difference in 10-year CVD risk between Black and White patients with SMI [46]. Our findings extend recent reports of increasing CVD risk factors for young adults in the US which disproportionately affect Black individuals [22], suggesting that Black individuals with SMI may be a particularly high-risk subgroup requiring targeted interventions.
Additionally, we found that SMI is associated with higher 30-year CVD risk to a greater degree for women than men. Sex and gender disparities in CVD risk have been established in the general population [47], with a growing literature suggesting that middle aged and older women with SMI may have higher relative risk of CVD incidence and risk factors compared to men [48-54]. Several potential drivers of sex and gender disparities in CVD risk for those with SMI could exist, including sex-hormone irregularities [55], higher rates of anxiety, depression, and early life traumatic experiences in women compared to men with SMI [56-58], and the additive impacts of socioeconomic disadvantage experienced by women [47].
Our results highlight the need to initiate CVD screening and intervention early for patients with SMI. Behavioral, pharmacological, and clinical decision support interventions have shown promising initial results in lowering total CVD risk and specific risk factors for those with SMI [6, 11, 13, 39], with some evidence suggesting these interventions may be most effective for younger adults [11, 38]. Yet few studies focused on CVD risk reduction have specifically targeted young adults with SMI [13], and evidence suggests young adults may have unique preferences, motivations, and barriers to engaging in health promotion interventions [59-62]. There is an urgent need for effective interventions tailored to young adults with SMI [63] to capitalize on a potential critical window for prevention and early intervention of cardiovascular disease [64].
4.1. Limitations
Our study has several potential limitations. Our sample consisted of patients receiving care in a large integrated healthcare system in the Midwest, with a large number insured with commercial insurance, which may limit generalizability to other clinical settings, geographic regions, and to patients with SMI with limited primary care access. Together, the effect of these limitations could be an underestimation of 30-year CVD risk in young adults with SMI. Absent lab values for lipids, especially for those without SMI, limited the sample size for comparisons on these measures. Similarly, those without SMI were more likely to have absent data for BMI, which is a variable in the 30-year risk equation. Patients with SMI may have routine lipid and weight monitoring if prescribed psychotropic medications, regardless of risk for dyslipidemia or obesity. Conversely, those without SMI undergoing lipid and weight monitoring may have poorer cardiometabolic health than patients without SMI who did not have these measures collected, resulting in a higher risk non-SMI sample in this comparison. Therefore, our results potentially underestimate the difference in these risk factors and 30-year risk between those with and without SMI. We used hypertension diagnosis as a proxy for antihypertensive medication orders in the risk models, however as previously noted, evidence suggests that most patients (approximately 85%) who are aware of their hypertension diagnosis are treated [26], and the clinics in this study have high rates of concordance between hypertension diagnosis and treatment [27, 28]. Our sensitivity analysis also indicated that using diagnosis as a proxy for antihypertensive medications did not change the results, likely because rates of hypertension were low in both groups. Finally, we were unable to perform subgroup analyses for some racial groups due to small sample size.
5.0. Conclusions
We found a clinically and statistically significant 28% higher 30-year risk of having a cardiovascular event for young adults with SMI compared to those without SMI. The disparity in 30-year CVD risk associated with having SMI was even higher for Black individuals and women. Early and effective strategies for prevention and management of major CVD risk factors – especially smoking and elevated body mass index – are critically needed for young adults with SMI to reduce health disparities.
Supplementary Material
Highlights.
Young adults with serious mental illness (SMI) are at elevated 30-year risk of cardiovascular disease (CVD)
Young adults with SMI are almost 2x as likely to have obesity and 5x as likely to be smokers compared to those without SMI
These disparities in 30-year CVD risk for young adults with SMI are even more pronounced in Black individuals and women
CVD risk factors must be addressed as early as possible for young adults with SMI to reduce later morbidity and mortality
Funding:
This work was supported by Cooperative Agreement U19MH092201 with the National Institute of Mental Health (NIMH). NIMH had no role in the design, analysis, interpretation, or publication of this study.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflicts of Interests Disclosures: The authors have no conflicts of interest to disclose.
References
- [1].Walker ER, McGee RE and Druss BG. Mortality in mental disorders and global disease burden implications: a systematic review and meta-analysis. JAMA Psychiatry 2015;72:334–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Osborn DP, Levy G, Nazareth I, Petersen I, Islam A and King MB. Relative risk of cardiovascular and cancer mortality in people with severe mental illness from the United Kingdom's General Practice Rsearch Database. Arch Gen Psychiatry 2007;64:242–9. [DOI] [PubMed] [Google Scholar]
- [3].Davidson S, Judd F, Jolley D, Hocking B, Thompson S and Hyland B. Cardiovascular risk factors for people with mental illness. Aust N Z J Psychiatry 2001;35:196–202. [DOI] [PubMed] [Google Scholar]
- [4].Nielsen RE, Banner J and Jensen SE. Cardiovascular disease in patients with severe mental illness. Nat Rev Cardiol 2021;18:136–145. [DOI] [PubMed] [Google Scholar]
- [5].Correll CU, Solmi M, Veronese N, Bortolato B, Rosson S, Santonastaso P, et al. Prevalence, incidence and mortality from cardiovascular disease in patients with pooled and specific severe mental illness: a large-scale meta-analysis of 3,211,768 patients and 113,383,368 controls. World Psychiatry 2017;16:163–180. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Daumit GL, Dalcin AT, Dickerson FB, Miller ER, Evins AE, Cather C, et al. Effect of a Comprehensive Cardiovascular Risk Reduction Intervention in Persons With Serious Mental Illness: A Randomized Clinical Trial. JAMA Netw Open 2020;3:e207247. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Daumit GL, Evins AE, Cather C, Dalcin AT, Dickerson FB, Miller ER 3rd, et al. Effect of a Tobacco Cessation Intervention Incorporating Weight Management for Adults With Serious Mental Illness: A Randomized Clinical Trial. JAMA Psychiatry 2023;80:895–904. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Anthenelli RM, Benowitz NL, West R, St Aubin L, McRae T, Lawrence D, et al. Neuropsychiatric safety and efficacy of varenicline, bupropion, and nicotine patch in smokers with and without psychiatric disorders (EAGLES): a double-blind, randomised, placebo-controlled clinical trial. Lancet 2016;387:2507–20. [DOI] [PubMed] [Google Scholar]
- [9].Evins AE, Cather C, Pratt SA, Pachas GN, Hoeppner SS, Goff DC, et al. Maintenance treatment with varenicline for smoking cessation in patients with schizophrenia and bipolar disorder: a randomized clinical trial. JAMA 2014;311:145–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Druss BG, von Esenwein SA, Compton MT, Rask KJ, Zhao L and Parker RM. A randomized trial of medical care management for community mental health settings: the Primary Care Access, Referral, and Evaluation (PCARE) study. Am J Psychiatry 2010;167:151–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Rossom RC, Crain AL, O'Connor PJ, Waring SC, Hooker SA, Ohnsorg K, et al. Effect of Clinical Decision Support on Cardiovascular Risk Among Adults With Bipolar Disorder, Schizoaffective Disorder, or Schizophrenia: A Cluster Randomized Clinical Trial. JAMA Netw Open 2022;5:e220202. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Corlin L, Short MI, Vasan RS and Xanthakis V. Association of the Duration of Ideal Cardiovascular Health Through Adulthood With Cardiometabolic Outcomes and Mortality in the Framingham Offspring Study. JAMA Cardiol 2020;5:549–556. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Naslund JA, Whiteman KL, McHugo GJ, Aschbrenner KA, Marsch LA and Bartels SJ. Lifestyle interventions for weight loss among overweight and obese adults with serious mental illness: A systematic review and meta-analysis. Gen Hosp Psychiatry 2017;47:83–102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Correll CU, Robinson DG, Schooler NR, Brunette MF, Mueser KT, Rosenheck RA, et al. Cardiometabolic risk in patients with first-episode schizophrenia spectrum disorders: baseline results from the RAISE-ETP study. JAMA Psychiatry 2014;71:1350–63. [DOI] [PubMed] [Google Scholar]
- [15].Vazquez-Bourgon J, Gomez-Revuelta M, Mayoral-van Son J, Labad J, Ortiz-Garcia de la Foz V, Setien-Suero E, et al. Pattern of long-term weight and metabolic changes after a first episode of psychosis: Results from a 10-year prospective follow-up of the PAFIP program for early intervention in psychosis cohort. Eur Psychiatry 2022;65:e48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [16].Garrido-Torres N, Rocha-Gonzalez I, Alameda L, Rodriguez-Gangoso A, Vilches A, Canal-Rivero M, et al. Metabolic syndrome in antipsychotic-naive patients with first-episode psychosis: a systematic review and meta-analysis. Psychol Med 2021;51:2307–2320. [DOI] [PubMed] [Google Scholar]
- [17].Pencina MJ, D'Agostino RB Sr., Larson MG, Massaro JM and Vasan RS. Predicting the 30-year risk of cardiovascular disease: the framingham heart study. Circulation 2009;119:3078–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Stone NJ, Smith SC Jr., Orringer CE, Rigotti NA, Navar AM, Khan SS, et al. Managing Atherosclerotic Cardiovascular Risk in Young Adults: JACC State-of-the-Art Review. J Am Coll Cardiol 2022;79:819–836. [DOI] [PubMed] [Google Scholar]
- [19].An J, Zhang Y, Zhou H, Zhou M, Safford MM, Muntner P, et al. Incidence of Atherosclerotic Cardiovascular Disease in Young Adults at Low Short-Term But High Long-Term Risk. J Am Coll Cardiol 2023;81:623–632. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [20].Coello K, Kjaerstad HL, Stanislaus S, Melbye S, Faurholt-Jepsen M, Miskowiak KW, et al. Thirty-year cardiovascular risk score in patients with newly diagnosed bipolar disorder and their unaffected first-degree relatives. Aust N Z J Psychiatry 2019;53:651–662. [DOI] [PubMed] [Google Scholar]
- [21].Rossom RC, Hooker SA, O'Connor PJ, Crain AL and Sperl-Hillen JM. Cardiovascular Risk for Patients With and Without Schizophrenia, Schizoaffective Disorder, or Bipolar Disorder. J Am Heart Assoc 2022;11:e021444. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Aggarwal R, Yeh RW, Joynt Maddox KE and Wadhera RK. Cardiovascular Risk Factor Prevalence, Treatment, and Control in US Adults Aged 20 to 44 Years, 2009 to March 2020. JAMA 2023;329:899–909. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].Carliner H, Collins PY, Cabassa LJ, McNallen A, Joestl SS and Lewis-Fernandez R. Prevalence of cardiovascular risk factors among racial and ethnic minorities with schizophrenia spectrum and bipolar disorders: a critical literature review. Compr Psychiatry 2014;55:233–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].Das-Munshi J, Chang CK, Dutta R, Morgan C, Nazroo J, Stewart R, et al. Ethnicity and excess mortality in severe mental illness: a cohort study. Lancet Psychiatry 2017;4:389–399. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [25].Rossom RC, O'Connor PJ, Crain AL, Waring S, Ohnsorg K, Taran A, et al. Pragmatic trial design of an intervention to reduce cardiovascular risk in people with serious mental illness. Contemp Clin Trials 2020;91:105964. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [26].Muntner P, Miles MA, Jaeger BC, Hannon Iii L, Hardy ST, Ostchega Y, et al. Blood Pressure Control Among US Adults, 2009 to 2012 Through 2017 to 2020. Hypertension 2022;79:1971–1980. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Margolis KL, Asche SE, Bergdall AR, Dehmer SP, Groen SE, Kadrmas HM, et al. Effect of home blood pressure telemonitoring and pharmacist management on blood pressure control: a cluster randomized clinical trial. JAMA 2013;310:46–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [28].Margolis KL, Bergdall AR, Crain AL, JaKa MM, Anderson JP, Solberg LI, et al. Comparing Pharmacist-Led Telehealth Care and Clinic-Based Care for Uncontrolled High Blood Pressure: The Hyperlink 3 Pragmatic Cluster-Randomized Trial. Hypertension 2022;79:2708–2720. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [29].Perry BI, Osimo EF, Upthegrove R, Mallikarjun PK, Yorke J, Stochl J, et al. Development and external validation of the Psychosis Metabolic Risk Calculator (PsyMetRiC): a cardiometabolic risk prediction algorithm for young people with psychosis. Lancet Psychiatry 2021;8:589–598. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [30].Osborn DP, Hardoon S, Omar RZ, Holt RI, King M, Larsen J, et al. Cardiovascular risk prediction models for people with severe mental illness: results from the prediction and management of cardiovascular risk in people with severe mental illnesses (PRIMROSE) research program. JAMA Psychiatry 2015;72:143–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [31].Mitchell AJ, Vancampfort D, De Herdt A, Yu W and De Hert M. Is the prevalence of metabolic syndrome and metabolic abnormalities increased in early schizophrenia? A comparative meta-analysis of first episode, untreated and treated patients. Schizophr Bull 2013;39:295–305. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [32].Strassnig M, Kotov R, Cornaccio D, Fochtmann L, Harvey PD and Bromet EJ. Twenty-year progression of body mass index in a county-wide cohort of people with schizophrenia and bipolar disorder identified at their first episode of psychosis. Bipolar Disord 2017;19:336–343. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [33].De Hert M, Dekker JM, Wood D, Kahl KG, Holt RI and Moller HJ. Cardiovascular disease and diabetes in people with severe mental illness position statement from the European Psychiatric Association (EPA), supported by the European Association for the Study of Diabetes (EASD) and the European Society of Cardiology (ESC). Eur Psychiatry 2009;24:412–24. [DOI] [PubMed] [Google Scholar]
- [34].Vancampfort D, Correll CU, Galling B, Probst M, De Hert M, Ward PB, et al. Diabetes mellitus in people with schizophrenia, bipolar disorder and major depressive disorder: a systematic review and large scale meta-analysis. World Psychiatry 2016;15:166–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [35].Vandenberghe F, Gholam-Rezaee M, Saigi-Morgui N, Delacretaz A, Choong E, Solida-Tozzi A, et al. Importance of early weight changes to predict long-term weight gain during psychotropic drug treatment. J Clin Psychiatry 2015;76:e1417–23. [DOI] [PubMed] [Google Scholar]
- [36].Yatham LN, Kennedy SH, Parikh SV, Schaffer A, Bond DJ, Frey BN, et al. Canadian Network for Mood and Anxiety Treatments (CANMAT) and International Society for Bipolar Disorders (ISBD) 2018 guidelines for the management of patients with bipolar disorder. Bipolar Disord 2018;20:97–170. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [37].Buchanan RW, Kreyenbuhl J, Kelly DL, Noel JM, Boggs DL, Fischer BA, et al. The 2009 schizophrenia PORT psychopharmacological treatment recommendations and summary statements. Schizophr Bull 2010;36:71–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [38].de Silva VA, Suraweera C, Ratnatunga SS, Dayabandara M, Wanniarachchi N and Hanwella R. Metformin in prevention and treatment of antipsychotic induced weight gain: a systematic review and meta-analysis. BMC Psychiatry 2016;16:341. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [39].Hiluy JC, Nazar BP, Goncalves WS, Coutinho W and Appolinario JC. Effectiveness of Pharmacologic Interventions in the Management of Weight Gain in Patients With Severe Mental Illness: A Systematic Review and Meta-Analysis. Prim Care Companion CNS Disord 2019;21. [DOI] [PubMed] [Google Scholar]
- [40].Chesney E, Robson D, Patel R, Shetty H, Richardson S, Chang CK, et al. The impact of cigarette smoking on life expectancy in schizophrenia, schizoaffective disorder and bipolar affective disorder: An electronic case register cohort study. Schizophr Res 2021;238:29–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [41].White SA, Stone E, Murphy KA, Daumit GL and McGinty EE. Smoking Cessation Medication Prescribing for Smokers With and Without Mental Illness. Psychiatr Serv 2022:appips202100690. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [42].Churchwell K, Elkind MSV, Benjamin RM, Carson AP, Chang EK, Lawrence W, et al. Call to Action: Structural Racism as a Fundamental Driver of Health Disparities: A Presidential Advisory From the American Heart Association. Circulation 2020;142:e454–e468. [DOI] [PubMed] [Google Scholar]
- [43].Mazimba S and Peterson PN. JAHA Spotlight on Racial and Ethnic Disparities in Cardiovascular Disease. J Am Heart Assoc 2021;10:e023650. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [44].Das-Munshi J, Stewart R, Morgan C, Nazroo J, Thornicroft G and Prince M. Reviving the 'double jeopardy' hypothesis: physical health inequalities, ethnicity and severe mental illness. Br J Psychiatry 2016;209:183–5. [DOI] [PubMed] [Google Scholar]
- [45].O'Gallagher K, Teo JT, Shah AM and Gaughran F. Interaction Between Race, Ethnicity, Severe Mental Illness, and Cardiovascular Disease. J Am Heart Assoc 2022;11:e025621. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [46].Keenan TE, Yu A, Cooper LA, Appel LJ, Guallar E, Gennusa JV 3rd, et al. Racial patterns of cardiovascular disease risk factors in serious mental illness and the overall U.S. population. Schizophr Res 2013;150:211–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [47].O'Neil A, Scovelle AJ, Milner AJ and Kavanagh A. Gender/Sex as a Social Determinant of Cardiovascular Risk. Circulation 2018;137:854–864. [DOI] [PubMed] [Google Scholar]
- [48].Ortiz A, Sanches M, Abdelhack M, Schwaiger TR, Wainberg M, Tripathy SJ, et al. Sex-specific associations between lifetime diagnosis of bipolar disorder and cardiovascular disease: A cross-sectional analysis of 257,673 participants from the UK Biobank. J Affect Disord 2022;319:663–669. [DOI] [PubMed] [Google Scholar]
- [49].Salvi V, Aguglia A, Barone-Adesi F, Bianchi D, Donfrancesco C, Dragogna F, et al. Cardiovascular risk in patients with severe mental illness in Italy. Eur Psychiatry 2020;63:e96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [50].Goff DC, Sullivan LM, McEvoy JP, Meyer JM, Nasrallah HA, Daumit GL, et al. A comparison of ten-year cardiac risk estimates in schizophrenia patients from the CATIE study and matched controls. Schizophr Res 2005;80:45–53. [DOI] [PubMed] [Google Scholar]
- [51].Fiedorowicz JG, He J and Merikangas KR. The association between mood and anxiety disorders with vascular diseases and risk factors in a nationally representative sample. J Psychosom Res 2011;70:145–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [52].Jackson CA, Kerssens J, Fleetwood K, Smith DJ, Mercer SW and Wild SH. Incidence of ischaemic heart disease and stroke among people with psychiatric disorders: retrospective cohort study. Br J Psychiatry 2020;217:442–449. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [53].Wu SI, Chen SC, Liu SI, Sun FJ, Juang JJ, Lee HC, et al. Relative Risk of Acute Myocardial Infarction in People with Schizophrenia and Bipolar Disorder: A Population-Based Cohort Study. PLoS One 2015;10:e0134763. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [54].Olfson M, Gerhard T, Huang C, Crystal S and Stroup TS. Premature Mortality Among Adults With Schizophrenia in the United States. JAMA Psychiatry 2015;72:1172–81. [DOI] [PubMed] [Google Scholar]
- [55].Gogos A, Ney LJ, Seymour N, Van Rheenen TE and Felmingham KL. Sex differences in schizophrenia, bipolar disorder, and post-traumatic stress disorder: Are gonadal hormones the link? Br J Pharmacol 2019;176:4119–4135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [56].Vega P, Barbeito S, Ruiz de Azua S, Martinez-Cengotitabengoa M, Gonzalez-Ortega I, Saenz M, et al. Bipolar disorder differences between genders: special considerations for women. Womens Health (Lond) 2011;7:663–74; quiz 675-6. [DOI] [PubMed] [Google Scholar]
- [57].Comacchio C, Lasalvia A and Ruggeri M. Current evidence of childhood traumatic experiences in psychosis - focus on gender differences. Psychiatry Res 2019;281:112507. [DOI] [PubMed] [Google Scholar]
- [58].Azorin JM, Belzeaux R, Kaladjian A, Adida M, Hantouche E, Lancrenon S, et al. Risks associated with gender differences in bipolar I disorder. J Affect Disord 2013;151:1033–40. [DOI] [PubMed] [Google Scholar]
- [59].Gates J, Killackey E, Phillips L and Alvarez-Jimenez M. Mental health starts with physical health: current status and future directions of non-pharmacological interventions to improve physical health in first-episode psychosis. Lancet Psychiatry 2015;2:726–742. [DOI] [PubMed] [Google Scholar]
- [60].LaRose JG, Leahey TM, Hill JO and Wing RR. Differences in motivations and weight loss behaviors in young adults and older adults in the National Weight Control Registry. Obesity (Silver Spring) 2013;21:449–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [61].Firth J, Rosenbaum S, Stubbs B, Vancampfort D, Carney R and Yung AR. Preferences and motivations for exercise in early psychosis. Acta Psychiatr Scand 2016;134:83–4. [DOI] [PubMed] [Google Scholar]
- [62].Browne J, Naslund JA, Salwen-Deremer JK, Sarcione C, Cabassa LJ and Aschbrenner KA. Factors influencing engagement in in-person and remotely delivered lifestyle interventions for young adults with serious mental illness: A qualitative study. Early Interv Psychiatry 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [63].Aschbrenner KA, Naslund JA, Gorin AA, Mueser KT, Scherer EA, Viron M, et al. Peer support and mobile health technology targeting obesity-related cardiovascular risk in young adults with serious mental illness: Protocol for a randomized controlled trial. Contemp Clin Trials 2018;74:97–106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [64].Phutane VH, Tek C, Chwastiak L, Ratliff JC, Ozyuksel B, Woods SW, et al. Cardiovascular risk in a first-episode psychosis sample: a 'critical period' for prevention? Schizophr Res 2011;127:257–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.