Factors affecting relapse in patients discharged from a public hospital: results from survival analysis

Factors Related to 30-day Readmission following Hospitalization for Any Medical Reason among Patients with Mental Disorders: Facteurs liés à la réhospitalisation à 30 jours suivant une hospitalisation pour une raison médicale chez des patients souffrant de troubles mentaux

OBJECTIVE

This study evaluated the contributions of clinical, sociodemographic, and service use variables to the risk of early readmission, defined as readmission within 30 days of discharge following hospitalization for any medical reason (mental or physical illnesses), among patients with mental disorders in Quebec (Canada).

METHODS

In this longitudinal study, 2,954 hospitalized patients who had visited 1 of 6 Quebec emergency departments (ED) in 2014 to 2015 (index year) were identified through clinical administrative databanks. The first hospitalization was considered that may have occurred at any Quebec hospital. Data collected between 2012 and 2013 and 2013 and 2014 on clinical, sociodemographic, and service use variables were assessed as related to readmission/no readmission within 30 days of discharge using hierarchical binary logistic regression.

RESULTS

Patients with co-occurring substance-related disorders/chronic physical illnesses, serious mental disorders, or adjustment disorders (clinical variables); 4+ outpatient psychiatric consultations with the same psychiatrist; and patients hospitalized for any medical reason within 12 months prior to index hospitalization (service use variables) were more likely to be readmitted within 30 days of discharge. Patients who made 1 to 3 ED visits within 1 year prior to the index hospitalization, had their index hospitalization stay of 16 to 29 days, or consulted a physician for any medical reason within 30 days after discharge or prior to the readmission (service use variables) were less likely to be rehospitalized.

CONCLUSIONS

Early hospital readmission was more strongly associated with clinical variables, followed by service use variables, both playing a key role in preventing early readmission. Results suggest the importance of developing specific interventions for patients at high risk of readmission such as better discharge planning, integrated and collaborative care, and case management. Overall, better access to services and continuity of care before and after hospital discharge should be provided to prevent early hospital readmission.

Factors associated with 30-days and 180-days psychiatric readmissions: A snapshot of a metropolitan area

Psychiatric re-hospitalization rate is a widely used quality indicator within mental health care. This study aims to investigate which variables are implied in determining readmissions over two intervals after the index event, 30 days and 6 months. The study sample included 798 inpatients, it was divided into two groups: not readmitted patients (NRP) and readmitted patients (RP), which has been further split into: Readmitted within 30 days (RP30dd) and Readmitted during the 150-day period (between 31 and 180 days) after the index discharge (RP150). A multivariate logistic regression with backward selection method was performed in order to find variables independently associated with readmission. The overall incidence of readmissions was 16.04%. Discharge to a Psychiatric Nursing Home was found to be a protective factor for all the groups. In adds, for the overall readmission, compulsory index admission and higher education (this lasts as in RP30dd group) were protective factors; whereas higher length of stay (as for readmission within 31-180 days) and a diagnosis of Personality Disorder were risk factors. The patient-specific factors significantly associated with likelihood of rehospitalization in the final model do identify some high-risk groups toward to whom possibly address prevention strategies.

Predictors of 30-day Postdischarge Readmission to a Multistate National Sample of State Psychiatric Hospitals

BACKGROUND

Early discharge from psychiatric inpatient care may pose challenges for the patient's recovery and may incite a rapid return to the hospital. This study identified demographic, clinical, and the continuing of care characteristics associated with rapid readmission into a sample of psychiatric inpatient hospitals.

METHODS

Cross-sectional analysis of 60,254 discharges from state psychiatric hospitals. Logistic regression explored the relationship between predictors of rapid readmission.

RESULTS

Eight percent of discharges were readmitted to the same hospital within 30 days after discharge. Factors significantly related with rapid readmission included white (odds ratio, 1.23; 95% confidence interval, 1.13-1.34), non-Hispanic (1.48, 1.26-1.73), not married (1.53, 1.32-1.76), voluntarily admitted (1.18, 1.05-1.33), with length of stay (LOS) ≤ 7 days (3.52, 3.04-4.08), or LOS 8-31 days (3.20, 2.79-3.66), or LOS 32-92 days (1.91, 1.65-2.22), with a schizophrenia or other psychotic disorders (1.69, 1.46-1.96) or personality disorder (1.76, 1.50-2.06), referred to a setting different from the outpatient (1.27, 1.16-1.40), or with a living arrangement different from private residence (1.54, 1.40-1.68).

CONCLUSIONS

Disparities in rapid readmission rates exist among state psychiatric hospitals. A national overview of the individuals with mental illness at risk of being prematurely discharged may suggests insights into quality initiatives aimed at reducing rapid readmissions into psychiatric inpatient care.

Predictors of psychiatric readmission among patients with bipolar disorder at an academic safety-net hospital

OBJECTIVE

Even with treatment, approximately one-third of patients with bipolar disorder relapse into depression or mania within 1 year. Unfavorable clinical outcomes for patients with bipolar disorder include increased rates of psychiatric hospitalization and functional impairment. However, only a few studies have examined predictors of psychiatric hospital readmission in a sample of patients with bipolar disorder. The purpose of this study was to examine predictors of psychiatric readmission within 30 days, 90 days and 1 year of discharge among patients with bipolar disorder using a conceptual model adapted from Andersen's Behavioral Model of Health Service Use.

METHODS

In this retrospective study, univariate and multivariate logistic regression analyses were conducted in a sample of 2443 adult patients with bipolar disorder who were consecutively admitted to a public psychiatric hospital in the United States from 1 January to 31 December 2013.

RESULTS

In the multivariate models, several enabling and need factors were significantly associated with an increased risk of readmission across all time periods examined, including being uninsured, having ⩾3 psychiatric hospitalizations and having a lower Global Assessment of Functioning score. Additional factors associated with psychiatric readmission within 30 and 90 days of discharge included patient homelessness. Patient race/ethnicity, bipolar disorder type or a current manic episode did not significantly predict readmission across all time periods examined; however, patients who were male were more likely to readmit within 1 year. The 30-day and 1-year multivariate models showed the best model fit.

CONCLUSION

Our study found enabling and need factors to be the strongest predictors of psychiatric readmission, suggesting that the prevention of psychiatric readmission for patients with bipolar disorder at safety-net hospitals may be best achieved by developing and implementing innovative transitional care initiatives that address the issues of multiple psychiatric hospitalizations, housing instability, insurance coverage and functional impairment.

Exploring the predictors of early readmission to psychiatric hospital

BACKGROUND

Aims of this study are to explore the associations of readmission to psychiatric hospital over time, to develop a statistical model for early readmission to psychiatric hospital and to assess the feasibility of predicting early readmission.

METHOD

The sample comprised 7891 general psychiatric discharges in South London, taken from a large anonymised repository of electronic patient records. We initially explored time to readmission using Cox regression - this included investigation of time-dependent effects. Subsequently, we used logistic regression to create a predictive model for 90-day readmission. We investigated the effect on readmission of a set of variables that included demographic variables, diagnosis and legal status during the index admission, previous service use, housing variables and individual item scores on the Health of the Nation Outcome Scales (HoNOS) at admission and at discharge.

RESULTS

Fifteen per cent of those discharged were readmitted within 90 days. Cox regression demonstrated that the estimated baseline hazard of readmission declined steeply after discharge and that the effects of several predictors, especially diagnosis, changed over time - most notably, personality disorder was associated with increased readmission relative to schizophrenia at the time of discharge, but did not significantly differ by 1-year postdischarge. In the logistic regression, increased readmission was associated with personality disorder diagnosis; shorter length of the index admission (excepting zero length admissions); number of discharges in the preceding 2 years; and having a high score at discharge on the HoNOS overactive and aggressive behaviour item, cognitive problems item or hallucinations and delusions items. Detention under Section 3 or a forensic section of the Mental Health Act during the index admission was associated with reduced readmission. The coefficient of discrimination for the logistic regression, which is equivalent to r 2, was 0.04 and the estimated area under the receiver operating curve was 0.65.

CONCLUSIONS

The association found between early readmission and personality disorder diagnosis merits further investigation, as does the possible trade-off between reduction in length of stay and increased readmission. Other novel findings such as the associations found with HoNOS item scores also merit replication. As with previous studies, we found that the rate of readmission declines steeply after hospital discharge, so that the period immediately subsequent to discharge is a period of comparatively high risk. However, prediction of early readmission within this high-risk group remains challenging - it seems most likely that many unmeasured influences operate subsequent to the time of discharge.

The relationship of illness management and recovery to state hospital readmission

The current study examined the association between number of hours attended of the Illness Management and Recovery (IMR) program and psychiatric readmission rates after discharge from a state psychiatric hospital. The study used archival data, N = 1186, from a large northeastern state psychiatric hospital in the United States. A Cox's regression survival analyses was conducted, adjusting for extreme outliers and controlling for sociodemographic covariates, to examine the association between different amounts of IMR and the risk for returning to the hospital. After controlling for the client characteristics of age, sex, marital status, psychiatric diagnosis, and Global Assessment of Functioning score at discharge, as well as controlling for mean daily dose of generic hospital programming and the number of days of hospitalization, it was found that, for each hour of IMR, there was an associated 1.1% reduction in the risk for returning to the hospital. This suggests that participation in IMR while in inpatient settings may assist individuals in reducing their risk for returning to the hospital.

The revolving door phenomenon revisited: time to readmission in 17’145 [corrected] patients with 37'697 hospitalisations at a German psychiatric hospital

Objective

Despite the recurring nature of the disease process in many psychiatric patients, individual careers and time to readmission rarely have been analysed by statistical models that incorporate sequence and velocity of recurrent hospitalisations. This study aims at comparing four statistical models specifically designed for recurrent event history analysis and evaluating the potential impact of predictor variables from different sources (patient, treatment process, social environment).

Method

The so called Andersen-Gil counting process model, two variants of the conditional models of Prentice, Williams, and Peterson (gap time model, conditional probability model), and the so called frailty model were applied to a dataset of 17’415 patients observed during a 12 years period starting from 1996 and leading to 37’697 psychiatric hospitalisations. Potential prognostic factors stem from a standardized patient documentation form.

Results

Estimated regression coefficients over different models were highly similar, but the frailty model best represented the sequentiality of individual treatment careers and differing velocities of disease progression. It also avoided otherwise likely misinterpretations of the impact of gender, partnership, historical time and length of stay. A widespread notion of psychiatric diseases as inevitably chronic and worsening could be rejected. Time in community was found to increase over historical time for all patients. Most important protective factors beyond diagnosis were employment, partnership, and sheltered living situation. Risky conditions were urban living and a concurrent substance use disorder.

Conclusion

Prognostic factors for course of diseases should be determined only by statistical models capable of adequately incorporating the recurrent nature of psychiatric illnesses.

Factors associated with hospitalization of adult psychiatric patients: cluster analysis

OBJECTIVE

Factors associated with acute admission to inpatient psychiatric wards have been difficult to replicate, possibly reflecting methodological limitations of analyzing individual variables. The objective of this analysis was to identify factors associated with hospitalization at an inpatient psychiatric unit using cluster analytic methods.

METHODS

Demographic, admission and treatment data for all admissions to a single inpatient unit in 2010 were collected retrospectively. Cluster analysis was performed using Ward's method.

RESULTS

The initial clustering identified a high suicidality/crisis group, which then gave two further subclusters, an internalizing one characterized by affective symptoms and an externalizing one characterized by intoxication at admission, and a population with poor medication compliance that included most cases of psychosis. These subclusters had different clinical and demographic characteristics, different rates of hospital readmission and different durations of stay.

CONCLUSIONS

Cluster analysis may be a useful exploratory technique to assist in planning and developing services for adult patients needing admission to a psychiatric hospital.

Predicting future hospital utilization for mental health conditions

To develop a model using administrative variables to predict number of days in the hospital for a mental health condition in the year after discharge from a mental health hospitalization. Background, index hospitalization and preindex inpatient, emergency room, and outpatient utilization information were collected for 766 adult members discharged from a mental health hospitalization during a 1-year period. A regression model was developed to predict hospitalized days for a mental health condition in the year after discharge. A regression model was created containing five statistically significant predictors: Medicare insurance coverage, preindex mental health inpatient days, index length of stay, depression diagnosis, and number of mental health outpatient visits with a professional provider. It is possible to predict future mental health inpatient utilization at the time of discharge from a mental health hospitalization using administrative data, thus allowing disease managers to better identify members in greatest need of additional services and interventions.

Relationship between diagnosis and disposition of patients admitted to a state psychiatric hospital

The lifetime outcome for individuals diagnosed with affective disorders is generally more favorable than for those diagnosed with a schizophrenic disorder. We determined if a similar differential outcome could be detected among 139 patients hospitalized on the admissions unit of a state psychiatric facility between 1998 and 2001, and diagnosed with a Schizophrenic, Schizoaffective or Affective Disorder. The placement of each patient on discharge was categorized as an outpatient environment, a minimum-security treatment unit, a locked ward, or a highly secure forensic facility. Patients with an affective disorder were significantly less likely than the other two groups to have a co-occurring diagnosis of substance abuse, and they performed better on the neuropsychological assessments. However, the groups did not differ in their discharge placements, or in their length of stay. These findings suggest that resolution of more acute symptomatology may be unrelated to factors associated with long-term outcome for individuals suffering from severe and persistent mental illness.

Operating conditions of psychiatric hospitals and early readmission--effects of high patient turnover

OBJECTIVE

To study the association between hospitals' operating conditions and the risk of early readmission. The hypothesis was that high patient turnover might lead to a rise in the risk of readmission soon after discharge (within 30 days).

METHOD

A multivariate model including hospital and patient variables was tested using Cox's regression analysis, adjusting for clustering effects. The material included data from 20 hospitals, with 5,520 patients in the final model.

RESULTS

High patient turnover (annual discharges per bed) was significantly associated with an increased risk of readmission (hazard ratio (HR)= 3.37 (95% CI = 2.39-4.75)). In addition, being discharged from a ward with relatively low access to therapists increased the hazard further.

CONCLUSION

High patient turnover at the discharging ward was found to increase the patients' hazard of early readmission. This observation supports the hypothesis of a link between the operation conditions of the hospitals and patient outcome on a short time-scale.

The role of psychometric data in predicting inpatient mental health service utilization

Inpatient mental health readmission rates have increased dramatically in recent years, with a subset of consumers referred to as revolving-door patients. In an effort to reduce the financial burden associated with these patients and increase treatment efficacy, researchers have begun to explore factors associated with increased service utilization. To date, predictors of increased service usage are remarkably discrepant across studies. Further exploration, therefore, is needed to better explicate the relevance of "traditional" predictors and also to identify alternate strategies that may assist in predicting rehospitalization. One method that may be helpful in identifying patients at high risk is the development of a psychometric screening procedure. As a means to this end, the present study was designed to assess the potential usefulness of psychometric data in predicting mental health service utilization. The sample consisted of 131 patients hospitalized during an index period of 8 months at an acute-care psychiatric hospital. Number of readmissions was recorded in a 9 month post-index period. Measures completed during the index admission included the Brief Psychiatric Rating Scale-Anchored (BPRS-A), Symptom Checklist-90-Revised (SCL-90-R), Kaufman Brief Intelligence Test (K-BIT), and the Beck Depression Inventory (BDI). Results indicated that psychometric data accounted for significant variance in predicting past, present and future mental health service utilization. The BPRS-A, SCL-90-R, and BDI show particular promise as time efficient psychometric screening instruments that may better enable practitioners to identify patients proactively who are at increased risk for rehospitalization. Implications are discussed with regard to patient-treatment matching and discharge planning.