Predictors of prolonged hospital stay for the treatment of severe neuropsychiatric symptoms in patients with dementia: a cohort study in multiple hospitals

Clinical profiles of people with dementia exhibiting with neuropsychiatric symptoms admitted to mental hospitals: A multicenter prospective survey in Japan

AIMS

Patients with severe behavioral and psychological symptoms of dementia (BPSD) are often admitted to mental hospitals, while, inpatient care could also lead to prolonged hospital stay. The present study aims to survey clinical profiles of patients who required inpatient treatment for BPSD, and then establish the criteria for introducing inpatient treatment through assessment by certified psychiatrists.

METHODS

We performed a prospective survey about clinical characteristics of people with dementia who required treatment of BPSD at 12 mental medical institutions. All patients were assessed by certified psychiatrists to determine the optimal treatment settings: outpatient or inpatient. The multivariate logistic regression analysis was performed to specify factors contributed to the judgement of clinicians. Subsequently, the receiver operating characteristic curve analysis was conducted to explore a score derived from the Neuropsychiatric Inventory to divide patients into outpatient or inpatient groups.

RESULTS

The present study included 386 patients, of which 242 were admitted to mental hospitals. BPSD were classified into four domains, and aggressive BPSD was significantly associated with assessment for inpatient treatment; the adjusted odds ratio was approximately 2 regardless of dementia severity. Furthermore, the composite score of agitation, irritability and aberrant behavior showed the highest area under the curve value (=0.706), which differentiated inpatients from outpatients with a sensitivity of 76% and a specificity of 54%.

CONCLUSIONS

Aggressive BPSD was the risk factor for inpatient treatment. The composite score of the Neuropsychiatric Inventory subdomain-related aggressive BPSD could be a screening tool to introduce inpatient treatment for BPSD. Geriatr Gerontol Int 2021; 21: 825-829.

Impact of the carer on length of hospital stay for mental health: Results from two Australian surveys

Informal carers play a vital role in supporting Australians living with a mental illness, including during the acute phases of illness; however, little is known about their impact on length of hospital stay. We aimed to investigate the impact of having a carer and of carer burden on length of hospital stay for mental health. Two Australian datasets were used. Data from the 2010 National Survey of High Impact Psychosis (n = 1825) were used to investigate the impact of having versus not having a carer on length of hospital stay for mental health. Data from the UQ Carer Survey 2016 (n = 105), a convenience sample of mental health carers, were used to investigate the impact of weekly hours of care (a measure of objective carer burden) on length of stay. Multiple logistic regression and correlation analyses were performed to investigate the association between carer status/burden and length of stay. Having a carer was associated with a significantly longer length of hospital stay; however, this relationship was no longer significant after adjusting for diagnosis, global functioning, depressive symptoms, deliberate self-harm, mental health outpatient contacts and type of admission. Weekly hours of care did not significantly impact on length of stay. Patients with carers had poorer functioning which may be related to longer stays. Our analysis was not able to look at subgroups of carers with different needs. Future work is required to determine other components of the admission and discharge process where having a carer is influential.

Factors for inhibition of early discharge from the psychiatric emergency ward for elderly patients

BACKGROUND

As society is aging, the number of elderly patients with psychiatric disorder, such as dementia, is increasing. The hospitalization period of elderly patients in psychiatric wards tends to be prolonged. In this study, we have determined the factors that inhibit early discharge from the psychiatric emergency ward for elderly patients in Japan.

METHODS

The information was collected from patients admitted to our hospital's emergency ward for elderly patients with psychiatric disorders between May 2015 and April 2016. We compared various factors between the early discharge group and the non-early discharge group. In addition, we used a multiple logistic regression model to clarify the risk factors for non-early discharge.

RESULTS

Of the 208 elderly patients, body mass index (BMI) and serum albumin level were significantly lower in the non-early discharge group. In addition, we also showed that higher serum C-reactive protein (CRP) (> 0.5 mg/dL) and use of seclusion or physical restraint significantly inhibited the early discharge of patients. The results of multiple logistic analysis showed that the BMI ≤ 17.5 kg/m² (OR, 2.41 [95% confidence interval (CI) 1.06-5.46]), serum albumin level ≤ 30 g/L (OR, 3.78 [95% CI 1.28-11.16]), and use of seclusion or physical restraint (OR 3.78 [95% CI 1.53-9.37]) are particularly important explanatory factors.

CONCLUSIONS

Hypoalbuminemia, low BMI, and the use of seclusion or physical restraint were identified as the factors that inhibit early discharge from the psychiatric emergency ward for elderly patients. These factors reflect malnutrition and extremely serious psychiatric symptoms.

Admission of people with dementia to psychiatric hospitals in Japan: factors that can shorten their hospitalizations

AIM

People exhibiting serious behavioural and psychological symptoms of dementia are usually voluntarily or involuntarily committed to psychiatric hospitals for treatment. In Japan, the average hospital stay for individuals with dementia is about 2 years. Ideally, individuals should be discharged once their symptoms have subsided. However, we see cases in Japan where individuals remain institutionalized long after behavioural and psychological symptoms of dementia are no longer apparent. This study will attempt to identify factors contributing to shorter stays in psychiatric hospitals for dementia patients.

METHODS

Questionnaires consisting of 17 items were mailed to 121 psychiatric hospitals with dementia treatment wards in western Japan.

RESULTS

Out of 121 hospitals that received the questionnaires, 45 hospitals returned them. The total number of new patient admissions at all 45 hospitals during the month of August 2014 was 1428, including 384 dementia patients (26.9%). The average length of stay in the dementia wards in August 2014 was 482.7 days. Our findings revealed that the rate of discharge after 2 months was 35.4% for the dementia wards. In addition, we found that the average stay in hospitals charging or planning to charge the rehabilitation fee to dementia patients was significantly shorter than in hospitals not charging the rehabilitation fee.

CONCLUSION

In Japan, dementia patients account for over 25% of new admissions to psychiatric hospitals with dementia wards. The average length of stay in a psychiatric hospital dementia ward is more than 1 year. A discharge after fewer than 2 months is exceedingly rare for those in a dementia ward compared with dementia patients in other wards. If institutions focus on rehabilitation, it may be possible to shorten the stay of dementia patients in psychiatric hospitals.

Effectiveness of treatment in wards for patients with dementia and issues associated with this treatment: A prospective analysis

In order to optimize the treatment in wards for patients with dementia, we investigated predictors of the length of stay in the ward for patients with dementia. A prospective analysis of 72 patients with dementia was conducted in two wards for patients with dementia. Severity of dementia was assessed by the Clinical Dementia Rating, basic activities of daily living were evaluated using the Physical Self-Maintenance Scale, and severity of behavioral and psychological symptoms of dementia and caregiver distress were determined with the Nursing Home version of the Neuropsychiatric Inventory. Stepwise regression analysis was used to identify predictors of the length of stay. Fewer patients were discharged to home compared to those discharged to institutions or transferred to a different hospital or ward. The mean length of stay was 92.0 days. Clinical Dementia Rating and total distress scale score on the Nursing Home version of the Neuropsychiatric Inventory were significant independent predictors of the length of stay. The total Nursing Home version of the Neuropsychiatric Inventory score and total distress scale score on the Nursing Home version of the Neuropsychiatric Inventory were significantly improved through treatment. Our results suggest that treatment in the wards for patients with dementia is effective for improvement of behavioral and psychological symptoms of dementia. However, some patients could not be discharged to their home despite improvement of behavioral and psychological symptoms of dementia, and this may be related to caregiver burden at admission.

A study in transfer learning: leveraging data from multiple hospitals to enhance hospital-specific predictions

BACKGROUND

Data-driven risk stratification models built using data from a single hospital often have a paucity of training data. However, leveraging data from other hospitals can be challenging owing to institutional differences with patients and with data coding and capture.

OBJECTIVE

To investigate three approaches to learning hospital-specific predictions about the risk of hospital-associated infection with Clostridium difficile, and perform a comparative analysis of the value of different ways of using external data to enhance hospital-specific predictions.

MATERIALS AND METHODS

We evaluated each approach on 132 853 admissions from three hospitals, varying in size and location. The first approach was a single-task approach, in which only training data from the target hospital (ie, the hospital for which the model was intended) were used. The second used only data from the other two hospitals. The third approach jointly incorporated data from all hospitals while seeking a solution in the target space.

RESULTS

The relative performance of the three different approaches was found to be sensitive to the hospital selected as the target. However, incorporating data from all hospitals consistently had the highest performance.

DISCUSSION

The results characterize the challenges and opportunities that come with (1) using data or models from collections of hospitals without adapting them to the site at which the model will be used, and (2) using only local data to build models for small institutions or rare events.

CONCLUSIONS

We show how external data from other hospitals can be successfully and efficiently incorporated into hospital-specific models.