Dataset Information

Predicting Psychiatric Hospitalizations among Elderly Veterans with a History of Mental Health Disease.

ABSTRACT:

Introduction

Patient Aligned Care Team (PACT) care managers are tasked with identifying aging Veterans with psychiatric disease in attempt to prevent psychiatric crises. However, few resources exist that use real-time information on patient risk to prioritize coordinating appropriate care amongst a complex aging population.

Objective

To develop and validate a model to predict psychiatric hospital admission, during a 90-day risk window, in Veterans ages 65 or older with a history of mental health disease.

Methods

This study applied a cohort design to historical data available in the Veterans Affairs (VA) Corporate Data Warehouse (CDW). The Least Absolute Shrinkage and Selection Operator (LASSO) regularization regression technique was used for model development and variable selection. Individual predicted probabilities were estimated using logistic regression. A split-sample approach was used in performing external validation of the fitted model. The concordance statistic (C-statistic) was calculated to assess model performance.

Results

Prior to modeling, 61 potential candidate predictors were identified and 27 variables remained after applying the LASSO method. The final model's predictive accuracy is represented by a C-statistic of 0.903. The model's predictive accuracy during external validation is represented by a C-statistic of 0.935. Having a previous psychiatric hospitalization, psychosis, bipolar disorder, and the number of mental-health related social work encounters were strong predictors of a geriatric psychiatric hospitalization.

Conclusion

This predictive model is capable of quantifying the risk of a geriatric psychiatric hospitalization with acceptable performance and allows for the development of interventions that could potentially reduce such risk.

SUBMITTER: Burningham Z

PROVIDER: S-EPMC5982950 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BACKGROUND:Applications of natural language processing to mental health notes are not common given the sensitive nature of the associated narratives. The CEGS N-GRID 2016 Shared Task in Clinical Natural Language Processing (NLP) changed this scenario by providing the first set of neuropsychiatric notes to participants. This study summarizes our efforts and results in proposing a novel data use case for this dataset as part of the third track in this shared task. OBJECTIVE:We explore the feasibility and effectiveness of predicting a set of common mental conditions a patient has based on the short textual description of patient's history of present illness typically occurring in the beginning of a psychiatric initial evaluation note. MATERIALS AND METHODS:We clean and process the 1000 records made available through the N-GRID clinical NLP task into a key-value dictionary and build a dataset of 986 examples for which there is a narrative for history of present illness as well as Yes/No responses with regards to presence of specific mental conditions. We propose two independent deep neural network models: one based on convolutional neural networks (CNN) and another based on recurrent neural networks with hierarchical attention (ReHAN), the latter of which allows for interpretation of model decisions. We conduct experiments to compare these methods to each other and to baselines based on linear models and named entity recognition (NER). RESULTS:Our CNN model with optimized thresholding of output probability estimates achieves best overall mean micro-F score of 63.144% for 11 common mental conditions with statistically significant gains (p<0.05) over all other models. The ReHAN model with interpretable attention mechanism scored 61.904% mean micro-F1 score. Both models' improvements over baseline models (support vector machines and NER) are statistically significant. The ReHAN model additionally aids in interpretation of the results by surfacing important words and sentences that lead to a particular prediction for each instance. CONCLUSIONS:Although the history of present illness is a short text segment averaging 300 words, it is a good predictor for a few conditions such as anxiety, depression, panic disorder, and attention deficit hyperactivity disorder. Proposed CNN and RNN models outperform baseline approaches and complement each other when evaluating on a per-label basis.

Project description:BACKGROUND:Women Veterans have high rates of mental health disorders and other psychosocial factors which may render this population particularly vulnerable to negative health outcomes associated with unintended pregnancy. OBJECTIVE:The objective of our study was to assess the relationship between self-reported mental illness and history of unintended pregnancy among women Veterans. DESIGN AND PARTICIPANTS:Data are from a nationally representative, cross-sectional telephone survey of women Veterans, ages 18-45, who used VA for primary care within 12 months prior to interview (survey completion rate 83%). MAIN MEASURES:Predictors were self-report of any and number of mental health disorders (depression, anxiety, post-traumatic stress disorder, bipolar disorder, or schizophrenia). Outcomes were any and number of unintended pregnancies. Multivariable logistic and negative binomial regression were used to assess relationships between mental illness and unintended pregnancy. To assess women's current risk of unintended pregnancy, we examined associations between any mental health disorder and contraceptive use at last sex among heterosexually active women not desiring pregnancy. KEY RESULTS:Among 2297 women Veterans, 1580 (68.8%) reported a history of at least one mental health disorder, with 20.1, 21.6, and 27.0% reporting one, two, or three or more conditions, respectively. Any history of unintended pregnancy was reported by 1315 women (57.3%); 28.3% reported one, 15.6% reported two, and 13.4% reported three or more. Compared to women with no mental illness, women with any mental health disorder were more likely to report any unintended pregnancy (60.3 vs. 50.5%; adjusted OR 1.40; 95% CI 1.15, 1.71) and to have experienced greater numbers of unintended pregnancies (adjusted incidence rate ratio 1.29; 95% CI 1.15, 1.44). Increasing numbers of mental health disorders were associated with greater numbers of unintended pregnancies. Contraceptive use and method efficacy at last sex did not differ by mental health status. CONCLUSIONS:Women Veterans with mental health disorders are more likely to have experienced any and greater numbers of unintended pregnancies than Veterans without mental health disorders.

Project description:BackgroundThis study examined outcomes for two groups of stroke survivors treated in Veteran Health Administration (VHA) hospitals, those with a severe mental illness (SMI) and those without prior psychiatric diagnoses, to examine risk of non-psychiatric medical hospitalizations over five years after initial stroke.MethodsThis retrospective cohort study included 523 veterans who survived an initial stroke hospitalization in a VHA medical center during fiscal year 2003. The survivors were followed using administrative data documenting inpatient stroke treatment, patient demographics, disease comorbidities, and VHA hospital admissions. Multivariate Poisson regression was used to examine the relationship between patients with and without SMI diagnosis preceding the stroke and their experience with non-psychiatric medical hospitalizations after the stroke.ResultsThe study included 100 patients with SMI and 423 without SMI. Unadjusted means for pre-stroke non-psychiatric hospitalizations were higher (p = 0.0004) among SMI patients (1.47 ± 0.51) compared to those without SMI (1.00 ± 1.33), a difference which persisted through the first year post-stroke (SMI: 2.33 ± 2.46; No SMI: 1.74 ± 1.86; p = 0.0004). Number of non-psychiatric hospitalizations were not significantly different between the two groups after adjustment for patient sociodemographic, comorbidity, length of stay and inpatient stroke treatment characteristics. Antithrombotic medications significantly lowered risk (OR = 0.61; 95% CI: 0.49-0.73) for stroke-related readmission within 30 days of discharge.ConclusionsNo significant differences in medical hospitalizations were present after adjusting for comorbid and sociodemographic characteristics between SMI and non-SMI stroke patients in the five-year follow-up. However, unadjusted results continue to draw attention to disparities, with SMI patients experiencing more non-psychiatric hospitalizations both prior to and up to one year after their initial stroke. Additionally, stroke survivors discharged on antithrombotic medications were at lower risk of re-admission within 30 days suggesting the VHA should continue to focus on effective stroke management irrespective of SMI.