Bipolar Disorder Related Hospitalizations - a Descriptive Nationwide Study Using a Big Data Approach

Machine learning applied to the prediction of relapse, hospitalization, and suicide in bipolar disorder using neuroimaging and clinical data: A systematic review

BACKGROUND

Bipolar disorder (BD) is associated with increased morbidity/mortality. Adverse outcome prediction might help with the management of patients with BD.

METHODS

We systematically reviewed the performance of machine learning (ML) studies in predicting adverse outcomes (relapse or recurrence, hospital admission, and suicide-related events) in patients with BD. Demographic, clinical, and neuroimaging-related poor outcome predictors were also reviewed. Three databases (PubMed, Scopus, and Web of Science) were explored from inception to July 2023.

RESULTS

Eighteen studies, accounting for >30,000 patients, were included. Support vector machine, decision trees, random forest, and logistic regression were the most frequently used ML algorithms. ML models' area under the receiver operating characteristic (ROC) curve (AUC), sensitivity, and specificity ranged from 0.71 to 0.98, 72.7-92.8 %, and 59.0-95.2 % for relapse/recurrence prediction (4 studies (3 on relapses and 1 on recurrences). The corresponding values were 0.78-0.88, 21.4-100 %, and 77.0-99.7 % for hospital admissions (3 studies, 21,266 patients), and 0.71-0.99, 44.4-97.9 %, and 38.9-95.0 % for suicide-related events (10 studies, 5558 patients). Also, one study addressed a combination of the interest outcomes. Adverse outcome predictors included early onset BD, BD type I, comorbid psychiatric or substance use disorder, circadian rhythm disruption, hospitalization characteristics, and neuroimaging parameters, including increased dynamic amplitude of low-frequency fluctuation, decreased frontolimbic functional connectivity and aberrant dynamic functional connectivity in corticostriatal circuitry.

CONCLUSIONS

ML models can predict adverse outcomes of BD with relatively acceptable performance measures. Future studies with larger samples and nested cross-validation validation should be conducted to reach more reliable results.

Cause-specific mortality and comorbid neurodevelopmental disorder in 167,515 patients with bipolar disorder: An entire population longitudinal study

OBJECTIVE

Studies addressing premature mortality in bipolar disorder (BD) patients are limited by small sample sizes. Herein, we used almost 99 % of the population of Taiwan to address this issue, and its association with comorbid neurodevelopmental disorders and severe BD.

METHODS

Between 2003 and 2017, we enrolled 167,515 individuals with BD and controls matched 1:4 for sex and birth year from the National Health Insurance Database linked to the Database of National Death Registry in Taiwan. Time-dependent Cox regression models were used to examine cause-specific mortality (all-cause, natural, and unnatural causes [accidents or suicide]).

RESULTS

With adjustments of sex, age, income, urbanization, and physical conditions, suicide was associated with the highest risk of mortality (reported as hazard ratio with 95 % confidence interval: 9.15; 8.53-9.81) among BD patients, followed by unnatural (4.94; 4.72-5.17), accidental (2.15; 1.99-2.32), and natural causes (1.02; 1.00-1.05). Comorbid attention-deficiency hyperactivity disorder did not contribute to the increased risk of cause-specific mortality; however, comorbid autism spectrum disorder (ASD) increased such risks, particularly for natural (3.00; 1.85-4.88) and accidental causes (7.47; 1.80-31.1). Cause-specific mortality revealed a linear trend with the frequency of psychiatric hospitalization (all, p for trend <0.001), and BD patients hospitalized twice or more each year had 34.63-fold increased risk of suicide mortality (26.03-46.07).

CONCLUSIONS

BD patients with a higher frequency of psychiatric hospitalization have the highest risk of suicide mortality, and comorbid ASD was associated with an increased risk of natural and accidental causes of mortality.

Preliminary results from the Australian Genetics of Bipolar Disorder Study: A nation-wide cohort

OBJECTIVE

The Australian Genetics of Bipolar Disorder Study is a nation-wide cohort of adults living with bipolar disorder. The study aims to detect the relationships between genetic risk, symptom severity, and the lifetime prevalence of bipolar disorder, treatment response and medication side effects, and patterns and costs of health care usage.

METHODS

A total of 6682 participants (68.3% female; aged 44.8 ± 13.6 years [range = 18-90]) were recruited in three waves: a nation-wide media campaign, a mail-out based on prescriptions for lithium carbonate and through the Australian Genetics of Depression Study. Participants completed a self-report questionnaire. A total of 4706 (70%) participants provided a saliva sample and were genotyped and 5506 (82%) consented to record linkage of their Pharmaceutical and Medicare Benefits Schedule data.

RESULTS

Most participants were living with bipolar I disorder (n = 4068) while 1622 participants were living with bipolar II disorder and 992 with sub-threshold bipolar disorder. The mean age of bipolar disorder diagnosis was 32.7 ± 11.6 years but was younger in bipolar I (p = 2.0E-26) and females (p = 5.7E-23). Excluding depression with onset prior to bipolar disorder diagnosis, 64.5% of participants reported one or more co-occurring psychiatric disorders: most commonly generalised anxiety disorder (43.5%) and posttraumatic stress disorder (20.7%). Adverse drug reactions were common and resulted in discontinuation rates ranging from 33.4% for lithium to 63.0% for carbamazepine.

CONCLUSION

Our findings highlight the high rate of comorbidities and adverse drug reactions among adults living with bipolar disorder in the general Australian population. Future genomic analyses focus on identifying genetic variants influencing pharmacotherapy treatment response and side effects.

Self-rated health as a predictor of hospitalizations in patients with bipolar disorder or major depressive disorder: A prospective cohort study of the UK Biobank

BACKGROUND

To determine the association between self-rated health (SRH) and subsequent all-cause hospitalizations in patients with bipolar disorder (BD) or major depression (MDD).

METHODS

We conducted a prospective cohort study on people with BD or MDD in the UK from 2006 to 2010 using UK Biobank touchscreen questionnaire data and linked administrative health databases. The association between SRH and 2-year all-cause hospitalizations was assessed using proportional hazard regression after adjustment for sociodemographics, lifestyle behaviors, previous hospitalization use, the Elixhauser comorbidity index, and environmental factors.

RESULTS

A total of 29,966 participants were identified, experiencing 10,279 hospitalization events. Among the cohort, the average age was 55.88 (SD 8.01) years, 64.02 % were female, and 3029 (10.11 %), 15,972 (53.30 %), 8313 (27.74 %), and 2652 (8.85 %) reported excellent, good, fair, and poor SRH, respectively. Among patients reporting poor SRH, 54.19 % had a hospitalization event within 2 years compared with 22.65 % for those having excellent SRH. In the adjusted analysis, patients with good, fair, and poor SRH had 1.31 (95 % CI 1.21-1.42), 1.82 (95 % CI 1.68-1.98), and 2.45 (95 % CI 2.22, 2.70) higher hazards of hospitalization, respectively, than those with excellent SRH.

LIMITATIONS

Selection bias can exist as our cohort cannot fully represent all the BD and MDD cases in the UK. Moreover, the causality is questionable.

CONCLUSION

SRH was independently associated with subsequent all-cause hospitalizations in patients with BD or MDD. This large study underscores the need for proactive SRH screening in this population, which might inform resource allocation in clinical care and enhance high-risk population detection.

Long-term disability in common mental disorders in Chinese community: evidence from a five-year follow-up study

BACKGROUND

Common mental disorders are general term for mental disorders with high disability rates and significant social burden. The purpose of this study was to determine the degree of long-term disability associated with common mental disorders and to interpret the relationship between common mental disorders and long-term disability.

METHODS

Participants in the 2013 China Mental Health Survey were followed up by telephone between April and June 2018. This study evaluated long-term disability over a five-year period using the World Health Organization's Disability Assessment Schedule 2.0. Poisson regression was used to analyze the relationship between common mental disorders and long-term disability.

RESULTS

A total of 6269 patients were followed up by telephone. In patients with common mental disorders, the prevalence of disability ranged from 7.62% to 43.94%. The long-term disabilities were significantly associated with dysthymic disorder (DD, RR:2.40; 95% CI:1.87-3.03), major depressive disorder (MDD, RR:1.63; 95% CI:1.34-1.98), generalized anxiety disorder (GAD, RR:1.95; 95% CI:1.15-3.09), obsessive-compulsive disorder (OCD, RR:1.68; 95% CI:1.24-2.22) and alcohol use disorder (AUD, RR: 1.42; 95% CI:0.99-1.96).

CONCLUSIONS

In China, common mental disorders raise the risk of long-term disability, and there is a critical need for monitoring patients with DD, MDD, GAD, OCD, and AUD. For improved quality of life and reduced disability levels, more resources need to be dedicated to mental health in the future.

Depressive Disorder Related Hospitalizations in Portugal Between 2008-2015: a Nationwide Observational Study

Depression is a prevalent disease, being one of the most relevant contributors of disability in the overall global burden of diseases. Hospitalization episodes are important quality indicators in psychiatric care. The primary aim of this study is to analyse depressive disorder related hospitalizations in Portuguese public hospitals and to detail clinical and sociodemographic differences among various subtypes of depression. Admissions with a primary diagnosis of depression in adult patients(> = 18 years) were selected from a national mainland hospitalization database. ICD-9-CM codes were used to select the diagnoses of interest: 296.2 × to 296.3x (Major depressive disorder), 300.4 (Dysthymic disorder) and 311 (Depressive disorder, not elsewhere classified). Birth date, sex, residence address, primary and secondary diagnoses, admission date, discharge date, length of stay (LoS), discharge status, and hospital estimated charges were obtained. A total of 28,569 hospitalizations (22,387 patients) with a primary diagnosis of depression were analysed. In the 8-year period of the study, 19.1% of all hospitalizations with a primary diagnosis of psychiatric disorder were linked to Depression. Major Depressive episodes were the most common (n = 15,384; 53.8%), followed by Depression unspecified episodes (n = 6,793; 23.8%), and Dysthymia (n = 6,392; 22.4%). Most episodes occurred in female patients (70.2%; n = 20,052), with a mean age of 50.6 years, and 37.0% (n = 10,564) of the episodes were associated to other psychiatric comorbidities. Depressive disorders are one of the leading causes of hospitalization in Portuguese psychiatric departments, being responsible for approximately 1 in 5 hospitalizations with a psychiatric diagnosis.