A clinical decision support (CDS) system that can be implemented in primary care reduces cardiovascular risk factors (CV) common in patients with serious mental illness (SMI), new research suggests.
Investigators assessed more than 70 primary care clinics that treated close to 9000 patients with SMI. Disorders included schizophrenia, bipolar disorder, and schizoaffective disorder. The clinics were randomly assigned to either provide or not provide a CDS tool aimed at improving CV health by assessing modifiable CV risk factors and providing personalized treatment recommendations.
Results showed that among patients in the clinics that used the intervention, there was a 4% reduction in modifiable CV risk over 12 months in comparison with those who did not use the intervention.
“Clinicians can make an impact on cardiovascular risk for patients with serious mental illness, which is an important finding, as cardiovascular disease is the leading cause of death in people with serious mental illness,” lead author Rebecca Rossom, MD, senior investigator, HealthPartners Institute, and associate professor, University of Minnesota Medical School, Minneapolis, told Medscape Medical News.
“Our intervention particularly highlights the value of using 30-year CV risk estimates to prompt earlier intervention for CV risk in our younger patients with serious mental illness,” said Rossom.
The findings were published online March 7 in JAMA Network Open.
“In theory, excess CV mortality among people with SMI could be reduced by early recognition and management of modifiable CV risk factors, among other strategies,” the investigators write.
CDS systems are “health information technology tools that give clinicians and patients patient-specific information at the point of care to improve care,” Rossom said. She added these systems “help prompt clinicians to address gaps in evidence-based care.”
Previous studies of CDS have had “null” results, partly because of poor CDS design and low use of CDS tools, the researchers note. Improved design and implementation have led to higher user rates and decreased CV risk in populations without SMI, according to more recent studies.
The current trial was “designed to test the effectiveness of the intervention in real-world clinical practice” by assessing whether an electronic health records–linked CDS system might slow increases in CV risk among adults with SMI, the investigators write.
The study was conducted at 76 primary care clinics in three healthcare systems. Researchers randomized the clinics to either receive or not receive the CDS for implementation for their patients with SMI.
To be included in the study, a clinic had to have treated at least 20 patients with SMI during the previous year. The researchers chose a cluster randomization approach “to minimize contamination.”
Patients with SMI were required to be 18 to 75 years old and have at least one modifiable risk factor not at the goal set by the American College of Cardiology/American Heart Association (ACC/AHA) guidelines.
“Our CDS system summarized and prioritized each patient’s modifiable CV risk factors, estimated 10- and/or 30-year CV risk (depending on patient age), and gave patient-specific suggestions regarding medications, diet, exercise, and smoking cessation.” Rossom said.
There were no alerts or printouts in the clinics that did not implement the CDS.
If study participants had a subsequent visit with a prescribingpsychiatric practitioner, the CDS alerted the prescriber of potentially obesogenic SMI medications for those with elevated body mass index (BMI) or recent weight gain.
Primary, Secondary Outcomes
The primary outcome was a patient-level rate of change in total modifiable CV risk during the 12 months after the index visit. Secondary outcomes included rate of change in individual modifiable CV risk factors during the 12 months after the index visit. Total modifiable CV risk and individual CV risk factors were calculated by the CDS at each clinic visit.
To calculate a risk component for each modifiable CV risk factor, the researchers used the difference between the patient’s values and the goal, which was based on the ACC/AHA guideline, the Framingham Heart Study, and the UK Prospective Diabetes Study equations.
The study included 42 intervention and 34 control clinics, encompassing 8937 patients with SMI (55.1% women; mean age, 48.4 years).
Of the participants, 66% had bipolar disorder, 19.5% had schizoaffective disorder, and 14.4% had schizophrenia. Most patients (83.7%) were White, followed by Black (10.1%), Native American (2%), and Hispanic (1.4%) patients.
The mean 10-year total CV risk was estimated to be 8.3%, while the mean total modifiable CV risk was estimated to be 3.7%.
The 30-year CV risk estimates for most patients were one (46.3%) or at least two (40.3%) major CV risk factors that “did not change for the better,” the investigators report.
The intervention group had an estimated 1% decrease in total modifiable CV risk rate ratio (RR), in comparison with an estimated 4% increase in the control group, which translated into a net 4% lower increase in total modifiable CV risk among the intervention group vs the control group (RR, .96; 95% CI, .94 – .98).
Although the overall treatment effect was “positive,” there were no significant treatment effects for individual modifiable CV risk factors except differences for BMI, for which small favored the control group.
There were some “small, nonsignificant” changes in low-density lipoprotein cholesterol, A1c level, and smoking that favored the intervention group.
The intervention also favored particular patient subgroups, as shown in the following table:
|Patient subgroup||RR (95% CI)|
|Age 18 – 29 years||.89 (.81 – .98)|
|Age 50 – 59 years||.93 (.90 – .96)|
|Men||.96 (.94 – .99)|
|Women||.95 (.92 – .97)|
|black patients||.93 (.88 – .98)|
|White patients||.96 (.94 – .98)|
|Bipolar disorder||.96 (.94 – .99)|
|Schizoaffective disorder||.94 (.90 – .98)|
|Schizophrenia||.92 (.85 – .99)|
Rossom noted that CDS tools “do take some upfront investment by healthcare systems, as well as dedicated clinician time to keep algorithms up to date as treatment recommendations evolve.”
However, “well-designed CDS tools that are integrated with clinician workflows can summarize relevant clinical information in one place, saving clinicians time and clicks and informing their clinical decision-making,” she added.
Commenting for Medscape Medical NewsDebabrata Mukherjee, MD, chair of the Department of Internal Medicine and chief of cardiovascular medicine at Texas Tech University Health Sciences Center El Paso, said there is “definitely more that needs to be done to improve cardiovascular health” in those with SMI.
“A tailored approach to address cardiovascular risk factors, embedded in a routine outpatient specialty mental health care for adults with serious mental illness, may pay dividends,” added Mukherjee, who is also a member of the ACC.org editorial board and was not involved with the research.
He said that significant differences shown in intervention effectiveness by race and ethnicity are “concerning and need additional study, as well as how the clinical decision support system interventions may impact current outpatient workflows.”
Mukherjee noted that the similar, previously conducted CHANGE trial “did not support superiority of individual lifestyle coaching or care coordination” compared with treatment as usual in reducing cardiovascular risk in patients with schizophrenia spectrum disorders.
Therefore, “the current results need validation in other settings,” he concluded.
The study was funded by the National Institute of Mental Health. Rossom reports receiving grants from the National Institute of Mental Health during the conduct of the study as well as receiving grants from Otsuka, Bioxcel, and Adelphi outside the submitted work. The other authors’ disclosures are listed in the original article. Mukherjee reports no relevant financial relationships.
JAMA Netw Open. Published online March 7, 2022. Full article
For more Medscape Psychiatry news, join us on Facebook and Twitter.