Electroconvulsive Therapy in Veterans Health Administration Hospitals: Prevalence, Patterns of Use, and Patient Characteristics

Factors associated with electroconvulsive therapy treatment for adults with serious psychiatric conditions in Australia

OBJECTIVE

To identify factors associated with receiving electroconvulsive therapy (ECT) for serious psychiatric conditions.

METHODS

Retrospective observational study using hospital administrative data linked with death registrations and outpatient mental health data in New South Wales (NSW), Australia. The cohort included patients admitted with a primary psychiatric diagnosis between 2013 and 2022. The outcome measure was receipt of ECT.

RESULTS

Of 94,950 patients, 3465 (3.6%) received ECT. The likelihood of receiving ECT was higher in older (hazard ratio [HR] = 1.03), female (HR = 1.24) patients. Compared to depression, patients with schizophrenia/schizoaffective disorder (HR = 0.79), schizophrenia-related disorders (HR = 0.37), mania (HR = 0.64) and other mood disorders (HR = 0.45) had lower odds of receiving ECT. Patients with depression and one other serious psychiatric condition had higher odds of receiving ECT than depression alone. Bipolar disorder likelihood of ECT did not differ from depression. A higher number of mental health outpatient visits in the prior year and an involuntary index admission with depression were also associated with receiving ECT. Likelihood of receiving ECT increased with year of admission (HR = 1.32), private patient status (HR = 2.06), higher socioeconomic status (HR = 1.09) and being married (HR = 1.25).

CONCLUSIONS

ECT use for depression and bipolar disorder in NSW aligns with clinical national guidelines. Patients with schizophrenia/schizoaffective, schizophrenia-related disorders, mania and other mood disorders had lower likelihood of ECT than depression, despite ECT being recommended by clinical guidelines for these diagnoses. Variations in ECT were strongly associated with healthcare access, with private patients twice as likely to receive ECT than their public counterparts, suggesting a need to explore ECT accessibility.

Leveraging Natural Language Processing to Improve Electronic Health Record Suicide Risk Prediction for Veterans Health Administration Users

Background: Suicide risk prediction models frequently rely on structured electronic health record (EHR) data, including patient demographics and health care usage variables. Unstructured EHR data, such as clinical notes, may improve predictive accuracy by allowing access to detailed information that does not exist in structured data fields. To assess comparative benefits of including unstructured data, we developed a large case-control dataset matched on a state-of-the-art structured EHR suicide risk algorithm, utilized natural language processing (NLP) to derive a clinical note predictive model, and evaluated to what extent this model provided predictive accuracy over and above existing predictive thresholds.

Methods: We developed a matched case-control sample of Veterans Health Administration (VHA) patients in 2017 and 2018. Each case (all patients that died by suicide in that interval, n = 4,584) was matched with 5 controls (patients who remained alive during treatment year) who shared the same suicide risk percentile. All sample EHR notes were selected and abstracted using NLP methods. We applied machine-learning classification algorithms to NLP output to develop predictive models. We calculated area under the curve (AUC) and suicide risk concentration to evaluate predictive accuracy overall and for high-risk patients.

Results: The best performing NLP-derived models provided 19% overall additional predictive accuracy (AUC = 0.69; 95% CI, 0.67, 0.72) and 6-fold additional risk concentration for patients at the highest risk tier (top 0.1%), relative to the structured EHR model.

Conclusions: The NLP-supplemented predictive models provided considerable benefit when compared to conventional structured EHR models. Results support future structured and unstructured EHR risk model integrations.

Characteristics and current standard of care among veterans with major depressive disorder in the United States: A real-world data analysis

BACKGROUND

This study examined MDD treatment regimens received during the first observed and treated major depressive episode (MDE) among US veterans.

METHODS

This retrospective study, conducted using the Veterans Health Administration (VHA) database, supplemented with Medicare Part A/B/D data, included adults with ≥1 MDD diagnosis (index date) between 10/1/2015-2/28/2017 and ≥1 line of therapy (LOT) within the first observed complete MDE. Patient baseline (6-month pre-index) characteristics and up to six LOTs received during the first observed and treated MDE were assessed.

RESULTS

Of 40,240 veterans with MDD identified (mean age: 50.9 years, 83.9% male, 63.4% White, 88.6% non-Hispanic), hypertension (27.5%), hyperlipidemia (20.8%), and post-traumatic stress disorder (17.5%) were the most common baseline comorbidities. During the first observed and treated MDE, patients received a mean of 1.6 ± 1.0 LOTs, with 14.6% of patients receiving ≥3 LOTs. SSRI-monotherapy was the most commonly observed regimen in the first six LOTs, followed by SNRI-monotherapy in LOT 1 and antidepressants augmented by anticonvulsants in the remaining five LOTs. The antidepressant class of the previous LOT was commonly used in the subsequent LOT. SSRI-SSRI-SSRI was the most common LOT1-to-LOT3 sequencing pattern among patients receiving ≥3 LOTs.

LIMITATIONS

The study findings are limited to data in the VHA database and may not be generalizable to the non-veteran US population.

CONCLUSIONS

During the first observed and treated MDE, SSRI-monotherapy was the most common therapy in the first six LOTs. Cycling within SSRI class was the leading sequencing pattern of the first three LOTs among veterans who received ≥3 LOTs.

Twenty-year trends in use of electroconvulsive therapy among homeless and domiciled veterans with mental illness

BACKGROUND

To examine socioeconomic disparities in use of electroconvulsive therapy (ECT) among homeless or unstably housed (HUH) veterans with mental illness.

METHODS

National data from medical records in years 2000 to 2019 on 4 to 6 million veterans with mental illness, including 140 000 to 370 000 homeless veterans served annually from the U.S. Department of Veterans Affairs (VA) healthcare system, were analyzed to examine ECT utilization and changes in utilization over time.

RESULTS

ECT utilization was higher among HUH veterans (58-104 per 1000) than domiciled veterans with mental illness (9-15 per 1000) across years with a trend toward increasing use of ECT use among HUH veterans over time. Among HUH and domiciled veterans who received ECT, veterans received an average of 5 to 9 sessions of ECT. There were great regional differences in rates of ECT utilization among HUH and domiciled veterans with the highest overall rates of ECT use at VA facilities in the Northeast and Northwest regions of the country.

DISCUSSION

ECT is commonly and safely used in HUH veterans in a comprehensive healthcare system, but geographic and local factors may impede access to ECT for veterans who may benefit from this treatment. Efforts should be made to reduce barriers to ECT in the HUH population.

Mortality after electroconvulsive therapy

BACKGROUND

There are limited studies examining mortality associated with electroconvulsive therapy (ECT), and many studies do not include a control group or method to identify all patient deaths.

AIMS

We aimed to evaluate the risk of death associated with ECT treatments over 30 days and 1 year.

METHOD

We conducted a study analysing electronic medical record data from the Department of Veterans Affairs healthcare system between 2000 and 2017. We compared mortality among patients who received ECT with a matched group of patients created through propensity score matching.

RESULTS

Our sample included 123 479 individual ECT treatments provided to 8720 patients (including 5157 initial index courses of ECT). Mortality associated with individual ECT treatments was 3.08 per 10 000 treatments over the first 7 days after treatment. When comparing patients who received ECT with a matched group of mental health patients, those receiving ECT had a relative odds of all-cause mortality in the year after their index course of 0.87 (95% CI 0.79-1.11; P = 0.10), and a relative risk of death from causes other than suicide of 0.79 (95% CI 0.66-0.95; P < 0.01). The similar relative odds of all-cause mortality in the first 30 days after ECT was 1.06 (95% CI 0.65-1.73) for all-cause mortality, and 1.02 (95% CI 0.58-1.8) for all-cause mortality excluding suicide deaths.

CONCLUSIONS

There was no evidence of elevated or excess mortality after ECT. There was some indication that mortality may be reduced in patients receiving ECT compared with similar patients who do not receive ECT.