This article is more than 5 years old. Information may no longer be current.

Cardiovascular disease risk may be underestimated among mentally ill

Current risk prediction algorithms for cardiovascular disease that do not include severe mental illness as a predictor could be underestimating CVD risk by as much as 60%, according to findings published in PLOS Medicine.

“The new QRISK3 score used in the U.K. includes [severe mental illness] and atypical antipsychotic prescription as predictors to rectify underestimation of risk,” Ruth Cunningham, PhD, MPH, from the department of public health, University of Otago Wellington, New Zealand, and colleagues wrote. “However, other CVD risk assessment algorithms currently being used (for example PREDICT in New Zealand) do not include [severe mental illness], and so empirical investigation is needed to understand the magnitude of underestimation.”

Researchers examined whether contemporary CVD risk prediction equations underestimate disease risk in people with severe mental illness from a prospective cohort study of 495,388 primary care patients aged 30 to 74 years without prior CVD (the PREDICT study).

Previously, the PREDICT study was used to derive new CVD risk prediction equations. In this study, Cunningham and colleagues calculated CVD risk in participants with and without severe mental illness using new equations, then compared the predicted CVD risk with observed risk using survival methods.

The investigators identified 28,734 individuals who had contact with specialist mental health services in the PREDICT cohort. These participants had a higher observed rate of CVD events than those without a history of mental health service use or severe mental illness, indicating that the PREDICT equations underestimated the risk of CVD events for this group.

The results showed that the predicted risk for a CVD event was about 11% over 5 years in the highest decile of risk; however, the observed risk was about 14%. Although Cunningham and colleagues detected this pattern in both men and women, it was more pronounced in women, with a mean ratio of observed to predicted risk of 1.64 for women vs. 1.29 for men.

“The observed risk of an event in 5 years was 60% higher than estimated by the algorithm for women and 30% higher than estimated for men,” the researchers explained.

In addition, the predicted and observed risks were about equal among participants without a history of mental health service use, according to the study.

“This study provides a clear rationale for the development of CVD risk prediction algorithms that include predictors for people with [severe mental illness],” the investigators concluded. “Demonstrating the magnitude of this underestimation of CVD risk is important for primary care practice, as mental illness is not specifically included most available risk prediction algorithms.” – by Savannah Demko

Disclosure: The authors report no relevant financial disclosures.