FACTORS ASSOCIATED WITH READMISSIONS OF PATIENTS WITH MENTAL AND SUBSTANCE USE DISORDERS

FACTORES ASOCIADOS A LAS READMISIONES DE PACIENTES CON TRASTORNOS MENTALES Y USO DE SUSTANCIAS

ABSTRACT

Objective:  To analyze the factors associated with the readmissions of patients with mental and substance use disorders.

Method:  This is a quantitative, cross-sectional, and analytical study. Participants were enrolled from one general hospital, one university hospital and an Emergency Care Unit in the metropolitan area of Porto Alegre, RS, Brazil. The instrument used was the interRAI Emergency Screener for Psychiatry (interRAI ESP). Descriptive analysis and logistic regression were performed to compare the samples and identify factors associated with the risk of multiple readmissions (four or more).

Results:  The total sample consisted of 324 patients (average age: 41.79 ± 14.27 years, 61.04 % male), with four main diagnoses: mood disorders (29.14 %), substance use disorders (27.70 %), schizophrenia (23.74 %), and anxiety disorders (19.42 %). Most individuals had previous admissions, primarily at university hospital (81.82 %, p = 0.006), but no significant difference was found regarding multiple previous admissions (four or more), which ranged from 34 % to 40 % across data collection sites. A diagnosis of substance-related disorders was the factor most strongly associated with multiple readmissions (OR = 2.75; p = 0.039), followed by behavioral problems (OR = 2.62; p = 0.001) and mania (OR = 2.28; p = 0.012). The item on intrusive thoughts or previous trauma showed an odds ratio of 2.08 (p = 0.016). The presence of family support and community support networks had a protective effect, possibly preventing readmissions (OR = 0.49; p = 0.038).

Conclusion:  Considering that the main risk factor for readmissions is substance use disorders, there is an urgent need for investments in the Psychosocial Care Centers for alcohol and other drug users.

DESCRIPTORS:
Mental Health; Mental Health Disorders; Substance-Related Disorders; Patient Readmission; Risk Factors

RESUMO

Objetivo:  Analisar os fatores associados às readmissões de pacientes com transtornos mentais e uso de substâncias.

Método:  Estudo quantitativo, transversal e analítico. Os participantes são um hospital geral, um hospital universitário e uma Unidade de Pronto Atendimento da região metropolitana de Porto Alegre/RS, Brasil. Utilizou-se o instrumento interRAI Emergency Screener for Psychiatry. Foram realizadas análises descritivas e regressão logística para comparar as amostras e identificar fatores associados ao risco de múltiplas readmissões (quatro ou mais).

Resultados:  A amostra total foi composta por 324 pacientes (idade média: 41,79 ±14,27 anos, 61,04 % do sexo masculino), com quatro diagnósticos principais: transtornos do humor (29,14 %), transtornos relacionados ao uso de substâncias (27,70 %), esquizofrenia (23,74 %) e transtornos de ansiedade (19,42 %). A maioria dos indivíduos possuía internações prévias, especialmente no hospital universitário (81,82 %, p=0,006), mas não houve diferença significativa para múltiplas internações anteriores (4 ou mais), que variaram entre 34 % e 40 % nos locais de coleta de dados. O diagnóstico de transtornos relacionados ao uso de substâncias foi o fator com maior associação com múltiplas readmissões (OR=2,75; p=0,039), seguido por problemas comportamentais (OR=2,62; p=0,001) e mania (OR=2,28; p=0,012). O item sobre pensamentos intrusivos ou traumas prévios apresentou um odds ratio de 2,08 (p=0,016). A presença de suporte familiar e redes de apoio comunitário teve um efeito protetor, possivelmente prevenindo readmissões (OR=0,49; p=0,038).

Conclusão:  Considerando que o principal fator de risco para as readmissões são transtornos relacionados ao uso de substâncias, há urgência de investimentos nos Centros de Atenção Psicossocial para usuários de álcool e outras drogas.

DESCRITORES:
Saúde mental; Transtornos mentais; Transtornos relacionados ao uso de substâncias; Readmissão de pacientes; Fatores de risco

RESUMEN

Objetivo:  Analizar los factores asociados a las readmisiones de pacientes con trastornos mentales y uso de sustancias.

Método:  Estudio cuantitativo, transversal y analítico. Los participantes proceden de un hospital general, un hospital universitario y una Unidad de Atención de Urgencias de la región metropolitana de Porto Alegre, RS, Brasil. El instrumento utilizado fue el interRAI Emergency Screener for Psychiatry. Se realizaron análisis descriptivos y regresión logística para comparar las muestras e identificar los factores asociados al riesgo de múltiples readmisiones (cuatro o más).

Resultados:  La muestra total estuvo compuesta por 324 pacientes (edad media: 41,79 ±14,27 años, 61,04 % hombres), con cuatro diagnósticos principales: trastornos del estado de ánimo (29,14 %), trastornos por uso de sustancias (27,70 %), esquizofrenia (23,74 %) y trastornos de ansiedad (19,42 %). La mayoría de los individuos tenía hospitalizaciones previas, especialmente en el hospital universitario (81,82 %, p=0,006), pero no se encontró una diferencia significativa en cuanto a múltiples hospitalizaciones previas (cuatro o más), que oscilaron entre el 34 % y el 40 % en los distintos lugares de recolección de datos. El diagnóstico de trastornos relacionados con sustancias fue el factor con mayor asociación con múltiples readmisiones (OR=2,75; p=0,039), seguido por problemas de comportamiento (OR=2,62; p=0,001) y manía (OR=2,28; p=0,012). El ítem sobre pensamientos intrusivos o traumas previos presentó un odds ratio de 2,08 (p=0,016). La presencia de apoyo familiar y redes de apoyo comunitario tuvo un efecto protector, posiblemente previniendo readmisiones (OR=0,49; p=0,038).

Conclusión:  Considerando que el principal factor de riesgo para las readmisiones son los trastornos relacionados con sustancias, existe una necesidad urgente de invertir en los Centros de Atención Psicosocial para usuarios de alcohol y otras drogas.

DESCRIPTORES:
Salud Mental; Trastornos Mentales; Trastornos Relacionados con Sustancias; Readmisión de Pacientes; Factores de Riesgo

INTRODUCTION

Psychiatric readmissions are considered significant indicators for care planning and quality of life¹. High rates of readmissions are associated with worsening prognoses and negatively affect patient’s quality of life². Psychiatric readmissions are often justified for medical and legal reasons, generating high costs for the public health system³. A mental disorder is a syndrome characterized by a clinically significant disturbance in an individual's cognition, emotional regulation, or behavior, reflecting a dysfunction in the psychological, biological, or developmental processes underlying mental functioning⁴.

International research shows that patients with severe mental disorders, such as bipolar disorder, major depressive disorder, and schizophrenia, have a high number of readmissions^1,5. Bipolar Disorder (BD) was the most common primary diagnosis in frequent and no frequent service users in an Italian University psychiatric emergency department⁶. Readmissions of patients with major depressive disorder are associated with early onset of the disease, illiteracy, family history of mental illness, emotional problems, and other comorbidities⁷. A national American study found that one in three patients admitted for severe major depressive disorder were readmitted within 11 months⁸.

Individuals with substance use disorders present even higher rates of hospitalizations and prolonged hospital stays compared to other mental disorders⁹. Substance use disorder in the DSM-5 is characterized by a problematic pattern of use leading to clinically significant impairment or distress⁴. In patients with alcohol use disorder, a previous psychiatric diagnosis increased the risk of future rehospitalization. Diagnoses such as mood disorders or neurotic, stress-related, and somatoform disorders were associated with a 54 % and 39 % increase, respectively, in the risk of these patients experiencing their first rehospitalization for alcohol use disorder¹⁰.

The rates of hospital readmission within 30 days of discharge ranged from 9.1 % to 16 % ^1,5. Factors such as male gender, being single, and the length of hospital stay had a substantial impact on the likelihood of readmission¹. Other factors that were considered predictive of early readmission included shorter hospital stays, living in socioeconomically disadvantaged areas, having multiple diagnoses of psychiatric disorders or chronic physical illnesses¹¹.

Considering that national data on recent readmissions are scarce in Brazil^4,12 and that Emergency Care Units (UPAs) and emergency services often do not have an on-call psychiatrist, it is necessary to develop an evaluation tool for these patients to ensure proper diagnosis and treatment. The interRAI Emergency Screener for Psychiatry (ESP)¹³ addresses the functional, medical, and social factors of patients, with most items usable for risk algorithms.

The ESP provides insights for identifying individual needs, assessing risk factors for harm to self or others, appropriate interventions, and the type of care or service required. In a broader context, it could contribute significantly to the planning of public health and mental health policies, considering the influx of people with mental disorders and substance use issues in Emergency Care Units and general hospital emergencies.

This research aims to analyze the factors associated with readmissions of patients with mental disorders and substance use disorders.

METHOD

Study design and setting

This is a quantitative, cross-sectional, and analytical study. Interviews were conducted with patients in psychiatric crises and emergencies at a General Hospital (GH), a University Hospital (UH), and an Emergency Care Unit (UPA) in the metropolitan area of Porto Alegre, Rio Grande do Sul State, Brazil.

Study population and sample

The tool was applied to 343 participants, with 161 interviews conducted in the emergency department of a General Hospital (GH), 80 interviews conducted in the Mental Health Unit of the University Hospital (UH) l, and 102 interviews conducted with individuals awaiting psychiatric evaluation at the Emergency Care Unit (ECU). All interviews were conducted within the first 24 hours after admission to the services.

Recruitment and data collection

Inclusion criteria were individuals aged 18 years or older, regardless of gender, who were experiencing psychiatric crises or had been admitted to the Mental Health Unit for less than 24 hours. Exclusion criteria were individuals who could not be interviewed due to clinical circumstances requiring isolation during the pandemic, patients with psychomotor agitation, presenting a risk of aggression, intoxicated by psychoactive substances, sedated, or with moderate or severe cognitive impairment during the interview.

Data were collected using dedicated software with the Portuguese version of the interRAI Emergency Screener for Psychiatry (ESP) version 9.1¹³ which complied with interRAI’s international coding, allowing for multicenter research use. The interviews were conducted by a trained master student and a trained scholarship student, with an average duration of 50 minutes.

The first data collection took place in a General Hospital before the pandemic in 2019. The second data collection occurred simultaneously at a University Hospital (UH) and an ECU, starting in March 2023 and ending in November 2023. The University Hospital's mental health unit has 20 beds for adults and 9 for children, with psychiatric admissions lasting an average of 21 to 30 days. The Emergency Care Unit has two isolation beds, and 12 observation beds distributed between the emergency and urgent care rooms.

Instrument

The tool used for the interviews was the interRAI Emergency Screener for Psychiatry (ESP), designed to assess patients in psychiatric crises and emergencies. The ESP was created by the interRAI research network, composed of researchers and professionals from over 35 countries, aiming to improve health services. The principles of the interRAI mental health assessment systems focus on evaluating individuals to enhance their functional capacity and quality of life. This process aims to identify physical and mental health issues while promoting a higher level of independent functioning. To achieve these goals, it is essential to clearly define the purpose of the assessment and recognize psychiatric, functional, medical, and social factors that are either current challenges or potential concerns for the person.

The interRAI ESP¹³ is intended for use by healthcare and mental health professional such as general physicians, psychiatrists, psychologists, nurses, social workers, and family physicians. The ESP has 13 sections with items that assess different psychiatric disorders, psychoactive substance use, withdrawal symptoms, causes of harm to self and others, self-harm ideation or attempts, behavior, cognition, and communication, functional status, medications, relationships, support, and unstable life events, and environmental assessment. The software calculates specific triggers for risk algorithms (risk to self, risk to others, and self-care index).

Research^14,15 conducted with the ESP in Canada, Finland, and Brazil showed that the tool has adequate psychometric properties. The ESP has proven to be a reliable tool for triage in emergencies and mental health services, with consistent results in statistical analyses¹⁴. The validation process for Brazil demonstrated adequate psychometric properties, with Cronbach's Alpha and McDonald's Omega index values ranging from 0.60 to 0.94¹⁵.

Data analysis

First, descriptive statistics were used to describe the study sample. Patients from the three mental health units were compared according to sociodemographic and clinical characteristics using various items from the interRAI ESP tool, the most important provisional diagnostic category (DSM), and various scales and indices generated from the tool.

Secondly, an adjusted logistic model was constructed to identify factors associated with the likelihood of multiple past readmissions. To construct this model, the past readmissions item was used as a dependent variable, with the following dichotomization: 0: 0 to 3 readmissions; 1: 4 or more hospitalizations (Range 0-6). Relevant scales and items from the interRAI mental health assessment set were used in the regression model with their validated cut-off points. Validated scales and relevant items from the interRAI mental health assessment system were incorporated into the logistic regression model, using their established cut-off points (Table 1). These scales and indices are automatically generated based on the items of the interRAI ESP instrument and are associated with domains such as aggressive behavior, depression severity, mania, positive psychotic symptoms, self-harm, risk of harm to others, self-care capacity, and social withdrawal. Data were collected through the structured administration of the ESP, which identified key symptoms, risk factors, self-care abilities, and psychiatric disorder diagnoses. Age and gender were included to control the results for covariates.

Table 1 -
interRAI Scales in the interRAI Mental Health instruments set, their components, ranges, and cut-off values. Porto Alegre, RS, Brazil, 2024 (N= 324)

interRAI Scales	Components Items	Range	Abbreviations of the interRAI Instruments*	Cut-off values
Aggressive Behavior Scale	Verbal abuse; Physical abuse; Socially inappropriate/disturbing behavior; Resists care	0-12	MH CMH ESP	>=7
Depression Severity Index	Sad and painful facial expressions; Negative statements; Self-deprecation; Guilt/shame; Hopelessness	0-15	MH CMH ESP	>=8
Mania Scale	Inflated self-esteem; Hyperactivity; Irritability; Increased sociability/hypersexuality; Pressured speech; Labile affect; Sleep problems - hypomania	0-20	MH CMH ESP	>=9
Positive Symptoms Scale	Hallucinations; Command hallucinations; Delusions; Abnormal thought process	0-12	MH CMH ESP BMHS	>=6
Risk of Harm to Others	Extreme violence/behavior; Violent ideation; Intimidation/threats; Aggressive behavior scale; Positive symptoms scale; Insight; Delusions; Sleep	0-6	MH CMH ESP	>=3
Self-Care Index	Cognition; Positive symptoms; Insight; Decreased energy; Expressive communication; Hygiene; Mania scale; Anhedonia	0-6	MH CMH ESP	>=3
Self-Harm Severity Scale	Self-mutilation ideation; Suicide attempts; Suicide plan; Others concerned; Depression severity index; Positive symptoms scale; Cognitive performance scale	0-6	MH CMH ESP	>=3
Social Withdrawal Scale/Negative Symptoms	Withdrawal, reduction in social interactions, anhedonia/lack of pleasure, lack of motivation	0-12	MH CMH ESP	>=9

• *Abbreviations of the interRAI Mental Health instruments: CMH: interRAI Community Mental Health Instrument MH: interRAI Mental Health Instrument.
• ESP: interRAI Emergency Screener for Psychiatry Instrument BMHS interRAI Brief Mental Health Screener

RESULTS

A total of 324 individuals were assessed using the interRAI ESP tool. Table 2 shows that the mean age of the population analyzed was 41.79 (±14.27), with no significant difference in age across settings. The University Hospital (UH) had a higher proportion of males, 61.04 %, with a significant difference between the data collection sites (p=0.015). Significant differences were also found for the inability to care for dependents (highest in ECU: 73.68 % p=0.000).

Most of the people in the sample had previous hospitalizations, mainly in the University Hospital (81.82 % p=0.006), but no significant difference was found for multiple prior readmissions (4 or more), ranging from 34 % to 40 %, between the data collection sites. Likewise, no significant difference was found for family support and family members or health professionals concerned about the risk of self-harm.

Table 2 -
Comparison of sample characteristics of hospitalized patients, according to age, gender and clinical characteristics. Porto Alegre, RS, Brazil, 2024 (N= 324)

Characteristics	GH* n= 152	UH† n=77	ECU‡ n=95	Total Sample N=324	P values
Age (±SD§)	42.10 (±13.80)	40.29 (±14.88)	42.51 (±14.58)	41.79 (±14.27)	0.56
Sex Male ( %)	41.72 %	61.04 %	53.68 %	49.85 %	0.015||
Have dependents	26.97 %	38.96 %	34.73 %	32.10 %	0.150
Impossibility to care for dependents	28.95 %	36.36 %	73.68 %	43.83 %	0.000¶
Previous readmissions	62.50 %	81.82 %	74.74 %	70.68 %	0.006**
Multiple previous readmissions (4 or more)	34.21 %	40.26 %	37.89 %	36.73 %	0.643
Support available for discharge or community living	78.95 %	66.23 %	72.63 %	74.07 %	0.108
Family or professionals concerned about the risk of self-harm	60.53 %	68.83 %	65.26 %	63.89 %	0.441

• *GH: General Hospital †UH: University Hospital ‡ECU: Emergency Care Unit
• §SD=Standard deviation ||p≤0,05 ¶p≤0,000 **p≤0,01

Table 3 shows the clinical characteristics of the samples. The University Hospital and the Emergency Care Unit (ECU) had the highest rates of severe depression symptoms, with 61.04 % and 51.58 %, respectively (p=0.000). No significant difference was found for cognitive decline levels, self-care index, self-harm severity scale, or insight level. The ECU had the highest rates of aggressive behavior, mania, harm to others scale, and positive symptoms scale, with all these factors showing significant differences between the samples.

The social withdrawal scale had higher rates in University Hospital and the ECU, with 68.83 % and 65.26 %, respectively, much higher than Hospital 1 (34.21 % p=0.000). There are four main diagnoses in the sample: mood disorders (29.14 %), substance use disorders (27.70 %), schizophrenia (23.74 %), and anxiety disorders (19.42 %). Significant differences were found between the data collection locations, with a high rate of depression in the ECU (46.43 %, p=0.000).

Table 3 -
Comparison of clinical characteristics of the samples. Porto Alegre, RS, Brazil, 2024 (N = 324)

Characteristics	GH* n= 152	UH† n=77	ECU‡ n=95	Total Sample N=324	P values
Mild to severe cognitive decline (value interRAI Scale e>=2) ( %)	38.82 %	38.96 %	35.79 %	37.96 %	0.874
Positive symptoms scale ( %)	23.03 %	31.17 %	41.05 %	30.25 %	0.011§
Depression severity index ( %)	23.03 %	61.04 %	51.58 %	40.43 %	0.000||
Self-care index ( %)	37.50 %	40.26 %	36.84 %	37.96 %	0.888
Aggressive behavior scale ( %)	8.67 %	13.70 %	29.76 %	15.64 %	0.000||
Mania Scale ( %)	22.37 %	44.16 %	53.68 %	36.73 %	0.000||
Self-harm severity scale ( %)	50.66 %	61.04 %	58.95 %	55.56 %	0.240
Harm to others scale ( %)	38.82 %	61.04 %	63.16 %	51.23 %	0.000||
Social withdrawal scale ( %)	34.21 %	68.83 %	65.26 %	51.54 %	0.000||
No insight	9.21 %	14.29 %	7.37 %	9.88 %	0.297
Principal diagnoses (DSM)
Substance use disorders ( %)	17.21 %	36.11 %	35.71 %	27.70 %	0.000||
Schizofrenia ( %)	27.87 %	27.78 %	14.29 %	23.74 %	0.068
Depression ( %)	28.69 %	9.72 %	46.43 %	29.14 %	0.000||
Anxiety ( %)	26.24 %	26.39 %	3.57 %	19.42 %	0.000||

• *GH: General Hospital †UH: University Hospital ‡ ECU: Emergency Care Unit
• §p≤0,05 ||p≤0,000

Table 4 shows an adjusted logistic, regression model for the significant factors associated with the occurrence of multiple hospitalizations (4 or more). The diagnosis of substance use disorders was the factor with the highest association with an odds ratio of 2.75 (p=0.039), followed by behavioral problems (OR=2.62; p=0.001) and mania (OR=2.28; p=0.012). The item of intrusive thoughts or past traumas showed an odds ratio of 2.08 (p=0.016). The presence of family support or living in the community proved to be a protective factor (OR=0.49; p=0.038).

Table 4 -
Factors associated with multiple readmissions of hospitalized patients (4 or more). Porto Alegre, RS, Brazil, 2024 (n = 324)

Factors	Odds ratio	Confidence Intervals		P values
		CI-	CI+
Age from 36 to 64 (Reference: 18-36)	1.23	0.69	2.19	0.483
Age over 65 (Reference: 18-36)	0.94	0.21	4.18	0.940
Sex Male	0.61	0.33	1.15	0.130
Diagnosis of depression (Reference: diagnosis of anxiety)	2.36	0.96.	5.81	0.061
Diagnosis of schizophrenia (Reference: diagnosis of anxiety)	1.84	0.68	5.00	0.229
Diagnosis of substance use disorders (Reference: diagnosis of anxiety)	2.75	1.05	7.22	0.039*
Stressor factor	0.73	0.41	1.29	0.284
Behavioral problems	2.62	1.45	4.74	0.001†
Mania Scale (9-20)	2.28	1.19	4.37	0.012*
Family support for discharge or community living	0.49	0.25	0.96	0.038*
Self-care index	1.69	0.87	3.27	0.123
Depression severity index	1.37	0.71	2.64	0.348
Self-harm index	0.79	0.43	1.48	0.468
Intrusive thoughts or past traumas	2.08	1.15	3.78	0.016*
Insight	0.52	0.19	1.44	0.210

• *= p≤0,05 †= p≤0,01

DISCUSSION

This study provides relevant contributions regarding factors related to readmissions of patients with mental disorders and substance use disorders as well as the identified risk factors for such readmissions. In our study, four main diagnoses identified were mood disorders (29.14 %), substance use disorders (27.70 %), schizophrenia (23.74 %), and anxiety disorders (19.42 %). These findings resonate with international research, which identified higher rates of hospitalizations among individuals with depressive disorders, bipolar disorder, schizophrenia, and substance use disorders^1,5. Schizophrenia, bipolar disorder, stimulant-related disorders, depressive disorders, trauma and stressor related conditions, alcohol related disorders were significantly associated with increased odds of mental health and substance use readmissions¹⁶.

A Brazilian study found that the mental disorders most associated with readmissions were substance use disorders, schizophrenia, and mood disorders. Of these, 15.5 % had at least one previous hospitalization, 10.5 % had three readmissions, and 6.7 % had four or more hospitalizations¹⁷. Our sample showed that substance use disorders is the main diagnostic associated with readmissions, followed by behavioral problems and mania. The results about substance use disorders are similar to other Brazilian study, which analyzed 1. 549, 298 cases of patients hospitalized for mental disorders in the public health system, showing that 39.4 % had disorders related to substance use¹⁷. People with substance use disorders had the highest risk of readmissions, with no difference in readmission risk between women-only and mixed-gender programs¹⁸.

A five-year follow-up study¹² found that patients with severe mental disorders, such as psychoses and severe mood disorders, had a higher risk of readmission when associated with factors including the use of substances like tobacco, alcohol, cannabis, or cocaine, poor treatment adherence, interruptions in treatment due to medication shortages in public pharmacies, and the use of typical antipsychotics. In our sample, rehospitalizations occurred more frequently at the University Hospital (81.82 %) and the ECU (74.74 %). In the University Hospital many patients were admitted treating substance use disorders and comorbid disorders such as bipolar disorders, schizophrenia and borderline personality disorder.

Our study showed that 70.68 % of the population had at least one previous hospitalization, and 36.73 % had more than four readmissions. In international literature, 13.94 % of patients with severe mental disorders were readmitted within 30 days after discharge, and 25.04 % were readmitted within a year¹⁹. Other research shows that readmission rates were 9.1 % for women and 9.3 % for men²⁰; 18 % of patients had been hospitalized before their current hospitalization¹¹; and the overall readmission rate within 30 days after hospital discharge was 10.5 % ⁵.

Our research evaluated patients who were admitted to an ECU. Of these, 74.74 % had at least one previous hospitalization, and 37.89 % had four or more psychiatric hospitalizations. These rates are consistent with Barker’s²⁰ research, which found that patients who used the emergency department in the year prior to their admission had higher chances of being readmitted. Another study¹¹ identified that patients readmitted early were from low-income backgrounds and had higher comorbidities.

The main risk factor for readmissions was short hospitalization duration and early discharge^11,21, substance use and personality disorders, abscondence or discharge by initiation of the clinic, and discharged to any place except the patient's home²². Studies^6,21,22 showed that people with behavioral problems had a significantly higher risk of readmission. In our sample, the odds for behavioral problems are high (OR= 2,26; p= 0,001), maybe it can be associated with the large number of patients with substance use disorders and mania in BD. Aggressive behavior, self-harm, and suicidal tendencies can contribute to readmission in a short time, within 30 days after discharge²².

Physical impairment and disabilities are associated with significantly higher chances of readmission within 30 days of hospital discharge²³. Our data suggest that 43.83 % of patients had difficulty caring for others, 37 % had mild to severe cognitive impairment, and deficits in self-care. Hospitalizations can also be linked to poor treatment adherence and a lack of medications in public pharmacies¹². A comparative study²⁴ evaluated three groups of patients with psychiatric hospitalizations and indicated that the group of patients with two or more hospitalizations had an earlier onset of psychiatric disorders, longer hospital stays, and a 35 % higher chance of readmission compared to the groups with one or no prior readmissions.

Bipolar Disorder (BD) is a serious mental illness that can lead to frequent hospital readmissions². Among individuals diagnosed with bipolar disorder, 11.4 % of patients were readmitted at least once between 31 and 180 days after discharge²⁵. Another study²¹ showed that 11 % of patients diagnosed with bipolar disorder were readmitted within 30 days after discharge. The duration of hospitalizations in patients with bipolar disorder over a 12-month retrospective period and a 9-month follow-up was significantly linked to suicide occurrences and the use of antipsychotic and antidepressant medications²⁶. Our sample identified higher odds in the mania scale (OR= 2,28; p= 0,012), which is consistent with bipolar disorder.

The clinical investigation of patients with BD experiencing the revolving-door (RD) phenomenon revealed that these patients have a higher prevalence of mixed episodes, medical comorbidities, and were discharged to psychiatric facilities. Additionally, treatment characteristics associated with RD included higher prescription rates of atypical antipsychotics, benzodiazepines and antidepressants²⁷. Similar results were found in another study²⁸. The RD phenomenon was most closely linked to the co-occurrence of mood disorders and substance use disorders. Additional factors included the presence of a medical comorbidity and an extended duration of hospitalization. Furthermore, placement in community residential facilities and the use of a first-generation long-acting antipsychotic were also found to be associated with RD²⁸.

Childhood emotional abuse, physical abuse, and emotional neglect were revealed to be risk factors for developing early-onset, chronic depressive episodes²⁹. Neglect is associated with neurocognitive alterations, impairing psychosocial functioning and increasing the risk of developing psychiatric disorders³⁰. Childhood maltreatment, including neglect, focused on patients with BD, reported a higher prevalence of emotional neglect, approximately 40 %, in comparison with healthy subjects³¹. This result is consistent with a systematic review and meta-analysis about childhood neglect, neglected trauma and prevalence in psychiatric disorders³².

Women diagnosed with schizophrenia or bipolar disorder show a strong history of adverse childhood experiences. In women with severe mental disorders, a greater exposure to adverse childhood experiences was linked to a higher severity of depressive and anxiety symptoms³³. Early trauma and impulsivity are directly related to substance use disorders such as alcohol, cocaine, and polysubstance use disorder³⁴. In our sample, the odds of intrusive thoughts or past traumas is 2.08, highlighting the role of trauma in mental disorders.

The results of our study indicated that family support, friends, and professionals act as protective factors. A study showed that family involvement during hospitalization and after discharge, communication with the care team, and a post-discharge care plan were associated with scheduled follow-up appointments within 7 days³⁵. Research conduced in six European countries found that support from peers and family, discharge planning, and psychoeducation during and after hospitalization, along with structured plans, coping strategies, self-monitoring techniques, and ongoing contact with local community services, played a crucial role in preventing hospital readmission³⁶.

Another protective factor for readmissions included being married, as these individuals had a 40 % lower chance of readmission compared to single individuals³⁷. Aguglia et al.⁶ suggested that the extent of family support, along with the strength of a patient's social and work-related networks, influences the rehospitalization rate.

A study conducted with family members of individuals using psychoactive substances at a CAPS Alcohol and Drugs unit revealed that care actions directed toward families are primarily focused on family support groups, attention, welcoming, problem-solving, psychological support, and guidance³⁸. Reducing hospitalizations due to psychoactive substance use requires investment in and strengthening of the healthcare network, so that professionals can provide adequate care for users and offer support and active listening to their families.

The study found that substance use disorders were the factor most strongly associated with multiple readmissions (OR= 2.75; p= 0.039). This finding resonates with other international studies^9,18,37,39 and a recent review of the literature⁴⁰, reporting that substance abuse was a crucial factor that increased the risk of readmission. The results indicate that hospitals can be strategic sites for substance use interventions. This requires strong coordination with outpatient services and more focused discharge planning, including both individual and family-based approaches³⁹. Based on this statistical association, the implementation of specific programs in Psychosocial Care Centers (CAPS) in Brazil is recommended, aiming to prevent hospital readmissions. In addition, prevention and treatment programs for substance use disorders should be developed in Primary Health Care.

This result also points to the need for investment in the Psychosocial Care Network (RAPS), particularly in implantation of CAPS ad III for users of alcohol and other drugs, as this population experiences a higher risk of recurrent hospitalizations. Additionally, the use of the ESP tool could support proper assessments in emergency services like UPAs and general hospitals, where psychiatrists are not available full-time, a reality seen in Brazil. This is particularly relevant given the increasing number of people with mental disorders and substance use disorders in Emergency Care Units (ECUs) and hospital admissions.

One of the limitations of the study was focusing solely on the primary diagnosis. Therefore, it is suggested that future research include secondary diagnoses due to the comorbidity between different conditions, such as bipolar disorder (BD) and substance use disorders and behavioral disorders, as well as the abuse of licit and illicit drugs. Another limitation of this study includes the absence of sociodemographic data, such as education, income, employment and housing. These factors are recognized as social determinants of health and directly influence access to care, health behaviors, and exposure to risks, affecting health outcomes throughout the lifespan. It is recommended that future research incorporate studies from additional emergency units across Brazil, including secondary diagnoses and sociodemographic data.

CONCLUSION

The findings of this study shed light on the factors associated with the readmission of patients with mental disorders and substance use disorders, particularly in light of the scarcity of national research on this subject. Among the factors most strongly linked to multiple readmissions, the diagnosis of substance use disorders demonstrated the highest association, followed by behavioral problems, manic episodes, and intrusive thoughts or past traumas. The presence of family and community support networks proved to be a significant protective factor, with the potential to reduce the likelihood of subsequent psychiatric hospitalizations.

The risk factors identified in this study may serve as important foundations for future investigations and the development of clinical protocols aimed at effective discharge planning. Strengthening the Psychosocial Care Network (RAPS), expanding the coverage of the Psychosocial Care Centers (CAPS), implementing preventive programs targeting substance use, and promoting active family involvement in care are promising strategies for reducing psychiatric readmissions. These findings carry substantial implications for public health administrators and policymakers committed to lowering readmission rates, optimizing healthcare resource utilization, and formulating integrated and effective public mental health policies.

REFERENCES

• 1. Muhammad N, Rasool S, Ullah R, Khan W, Saad M, Amjad H. Independent predictors of 30-day readmission to acute psychiatric wards in patients with mental disorders: A systematic review and meta-analysis. Cureus [Internet]. 2023 [cited 2025 Jun 02];15(7):e42490. Available from: https://doi.org/10.7759/cureus.42490
  » https://doi.org/10.7759/cureus.42490
• 2. Abaatyo J, Kaggwa MM, Favina A, Olagunju AT. Readmission and associated clinical factors among individuals admitted with bipolar affective disorder at a psychiatry facility in Uganda. BMC Psychiatry [Internet]. 2023 [cited 2025 Jan 02];23(1):474. Available from: https://doi.org/10.1186/s12888-023-04960-0
  » https://doi.org/10.1186/s12888-023-04960-0
• 3. Akram F, Waseem M, Ullah S, Fatima I, Rasool S. Predictors of civil and forensic inpatient psychiatric readmissions at a Public Mental Health Hospital. Psychiatry Res [Internet]. 2020 [cited 2025 Feb 05];293:113447. Available from: https://doi.org/10.1016/j.psychres.2020.113447
  » https://doi.org/10.1016/j.psychres.2020.113447
• 4. American Psychiatric Association (APA). Manual diagnóstico e estatístico de transtornos mentais: DSM-5 5th ed. Porto Alegre, RS(BR): Artmed; 2023.
• 5. Germack HD, Bizhanova Z, Roberts ET. Substantial hospital level variation in all-cause readmission rates among Medicare beneficiaries with serious mental illness. Healthc (Amst) [Internet]. 2020 [cited 2025 Jun 10];8(3):100453. Available from: https://doi.org/10.1016/j.hjdsi.2020.100453
  » https://doi.org/10.1016/j.hjdsi.2020.100453
• 6. Aguglia A, Corsini GP, Costanza A, Berti A, Bruno E, Escelsior A, et al. Sociodemographic and clinical correlates associated with frequent service users in an Italian psychiatric emergency department. Diagnostics (Basel) [Internet]. 2023 [cited 2025 Jun 13];13(3):430. Available from: https://doi.org/0.3390/diagnostics13030430
  » https://doi.org/0.3390/diagnostics13030430
• 7. Parami S, Smith AP, Rahman M, Joyner JP, Harris MJ. Identifying factors associated with the hospital readmission rate among patients with major depressive disorder. BMC Psychiatry [Internet]. 2021 [cited 2025 Jan 20];21(1):542. Available from: https://doi.org/10.1186/s12888-021-03559-7
• 8. Eseaton PO, Harris MD, Walton A, Chambers L, Davis CP. Severe major depressive disorder readmissions: A national United States population study. J Affect Disord [Internet]. 2023 [cited 2025 Jun 15];333:177-80. Available from: https://doi.org/ 10.1016/j.jad.2023.04.070
  » https://doi.org/ 10.1016/j.jad.2023.04.070
• 9. Habermeyer B, Zuellig B, Schuler J, Simons P, Schaub MP. Referral to aftercare following inpatient withdrawal treatment and readmission: Retrospective chart review of patients with substance use disorders. Swiss Med Wkly [Internet]. 2022 [cited 2025 Jan 17];152:w30213. Available from: https://doi.org/10.4414/SMW.2022.w30213
  » https://doi.org/10.4414/SMW.2022.w30213
• 10. Manca F, Lewsey J. Previous psychiatric hospitalizations as risk factors for single and multiple future alcohol-related hospitalizations in patients with alcohol use disorders. Addiction [Internet]. 2024 [cited 2025 Jan 15];119(2):291-300. Available from: https://doi.org/10.1111/add.16352
  » https://doi.org/10.1111/add.16352
• 11. Gentil L, Andrades M, Correia J, Martins H. Predictors of length of hospitalization and impact on early readmission for mental disorders. Int J Environ Res Public Health [Internet]. 2022 [cited 2025 Feb 02];19(22):15127. Available from: https://doi.org/10.3390/ijerph192215127
  » https://doi.org/10.3390/ijerph192215127
• 12. Portela R, Santos V, Leite E, Amaral C, Pinto M. Risk factors associated with readmissions of patients with severe mental disorders under treatment with antipsychotics. BMC Psychiatry [Internet]. 2022 [cited 2025 Jun 20];22(1):189. Available from: https://doi.org/10.1186/s12888-022-03794-6
  » https://doi.org/10.1186/s12888-022-03794-6
• 13. Rabinowitz T, Hirdes JP, Curtin-Telegdi N, Martin L, Smith TF, Morris JN, et al. InterRAI Emergency Screener for Psychiatry (ESP): Assessment Form and User’s Manual. Version 9.1. Washington D. C., WS(US): InterRAI; 2013.
• 14. Hirdes JP, Ljunggren G, Morris JN, Frijters D, Finne-Soveri H, Gray L, et al. The interRAI suite of mental health assessment instruments: An integrated system for the continuum of care. Front Psychiatry [Internet]. 2020 [cited 2025 Jan 18];10:926. Available from: https://doi.org/10.3389/fpsyt.2019.00926
  » https://doi.org/10.3389/fpsyt.2019.00926
• 15. Cabral LAB, Machado WL, Ferlin EL, Hirdes JP, Marrone LCP, Hirdes A. Evidence of validity of the instrument interRAI emergency screener for psychiatry for the Brazilian context. Res Soc Dev [Internet]. 2021 [cited 2025 Jan 18];10(13):e525101321185. Available from: https://doi.org/10.33448/rsd-v10i13.21185
  » https://doi.org/10.33448/rsd-v10i13.21185
• 16. Ouyang L, Cox S, Xu L, Robbins CL, Ko JY. Mental health and substance use disorders at delivery hospitalization and readmissions after delivery discharge. Drug Alcohol Depend [Internet]. 2023 [cited 2025 Jan 20];247:109864. Available from: https://doi.org/10.1016/j.drugalcdep.2023.109864
  » https://doi.org/10.1016/j.drugalcdep.2023.109864
• 17. Rocha HA, Reis IA, Santos MA, Melo APS, Cherchiglia ML. Psychiatric hospitalizations by the Unified Health System in Brazil between 2000 and 2014. Rev Saude Publica [Internet]. 2021 [cited 2025 Feb 02];55:14. Available from: https://doi.org/10.11606/s1518-8787.2021055002155
  » https://doi.org/10.11606/s1518-8787.2021055002155
• 18. Olivari CF, Carstens N, Bravo A, Silva AA. Treatment outcome and readmission risk among women in women-only versus mixed-gender drug treatment programs in Chile. J Subst Abuse Treat [Internet]. 2022 [cited 2025 Jun 02];134:108616. Available from: https://doi.org/10.1016/j.jsat.2021.108616
  » https://doi.org/10.1016/j.jsat.2021.108616
• 19. Han X, Zhang H, Liu L, Wang Y, Yang J. A sequence analysis of hospitalization patterns and service utilization in patients with major psychiatric disorders in China. BMC Psychiatry [Internet]. 2021 [cited 2025 Jun 06];21(1):245. Available from: https://doi.org/10.1186/s12888-021-03251-w
  » https://doi.org/10.1186/s12888-021-03251-w
• 20. Barker LC, Gruneir A, Fung K, Seitz DP. Predicting psychiatric readmission: Sex-specific models to predict 30-day readmission following acute psychiatric hospitalization. Soc Psychiatry Psychiatr Epidemiol [Internet]. 2018 [cited 2025 Feb 06];53(2):139-49. Available from: https://doi.org/10.1007/s00127-017-1450-5
  » https://doi.org/10.1007/s00127-017-1450-5
• 21. Ortiz G. Predictors of 30-day postdischarge readmission to a multistate national sample of state psychiatric hospitals. J Healthc Qual [Internet]. 2019 [cited 2025 Feb 12];41(4):228-36. Available from: https://doi.org/10.1097/JHQ.0000000000000162
  » https://doi.org/10.1097/JHQ.0000000000000162
• 22. Müller M, Helmreich I, Keller K, Heinz A, Gallinat J. Predictors for early and long-term readmission in involuntarily admitted patients. Compr Psychiatry [Internet]. 2024 [cited 2025 Jan 12];128:152439. Available from: https://doi.org/10.1016/j.comppsych.2023.152439
  » https://doi.org/10.1016/j.comppsych.2023.152439
• 23. Reeves LE, Blais RK, Hale WJ, Cain B. The role of medical comorbidity in the rapid psychiatric readmission of patients with schizophrenia-spectrum disorders. J Psychiatr Pract [Internet]. 2021 [cited 2025 Jan 12];27(1):14-22. Available from: https://doi.org/10.1097/PRA.0000000000000517
  » https://doi.org/10.1097/PRA.0000000000000517
• 24. Berardelli I, Rossetti MC, Ambrosi E, Barlati S, Pompili M, Fiorillo A, et al. Exploring risk factors for re-hospitalization in a psychiatric inpatient setting: A retrospective naturalistic study. BMC Psychiatry [Internet]. 2022 [cited 2025 Jun 02];22(1):821. Available from: https://doi.org/10.1186/s12888-022-04472-3
  » https://doi.org/10.1186/s12888-022-04472-3
• 25. Hermer L, Nephew T, Southwell K. Follow-up psychiatric care and risk of readmission in patients with serious mental illness in state funded or operated facilities. Psychiatr Q [Internet]. 2022 [cited 2025 Jan 12];93(2):499-511. Available from: https://doi.org/10.1007/s11126-021-09957-0
  » https://doi.org/10.1007/s11126-021-09957-0
• 26. Shi X, Zhao Y, Yang H, Xu X, Fang Y, Yu X, et al. Factors associated with hospitalization times and length of stay in patients with bipolar disorder. Front Psychiatry [Internet]. 2023 [cited 2025 Jun 02];14:1140908. Available from: https://doi.org/10.3389/fpsyt.2023.1140908
  » https://doi.org/10.3389/fpsyt.2023.1140908
• 27. Menculini G, Gobbicchi C, Verdolini N, Cirimbilli F, Moretti P, Tortorella A. "Revolving door" and bipolar disorders: A retrospective study in an acute inpatient unit. Psychiatr Danub [Internet]. 2020 [cited 2025 Jan 20];32 Suppl 1:58-63. Available from: https://www.psychiatria-danubina.com/UserDocsImages/pdf
  » https://www.psychiatria-danubina.com/UserDocsImages/pdf
• 28. Gobbicchi C, Verdolini N, Menculini G, Cirimbilli F, Gallucci D, Vieta E, et al. Searching for factors associated with the "Revolving Door phenomenon" in the psychiatric inpatient unit: A 5-year retrospective cohort study. Psychiatry Res [Internet]. 2021 [cited 2025 Jan 20];303:114080. Available from: https://doi.org/10.1016/j.psychres.2021.114080
  » https://doi.org/10.1016/j.psychres.2021.114080
• 29. Wang M, Wei J, Dou Y, Wang Y, Fan H, Yan Y, et al. Differential association between childhood trauma subtypes and neurocognitive performance in adults with major depression. BMC Psychiatry [Internet]. 2024 [cited 2025 Jan 20];24(1):773. Available from: https://doi.org/10.1186/s12888-024-06226-6
  » https://doi.org/10.1186/s12888-024-06226-6
• 30. McCrory E, Foulkes L, Viding E. Social thinning and stress generation after childhood maltreatment: A neurocognitive social transactional model of psychiatric vulnerability. Lancet Psychiatry [Internet]. 2022 [cited 2025 Feb 02];9(8):710-21. Available from: https://doi.org/10.1016/S2215-0366(22)00202-4
  » https://doi.org/10.1016/S2215-0366(22)00202-4
• 31. Dualibe AL, Osório FL. Bipolar disorder and early emotional trauma: A critical literature review on indicators of prevalence rates and clinical outcomes. Harv Rev Psychiatry [Internet]. 2017 [cited 2025 Feb 14];25(5):198-208. Available from: https://doi.org/10.1097/HRP.0000000000000154
  » https://doi.org/10.1097/HRP.0000000000000154
• 31. Carvalho Silva R, Oliva F, Barlati S, Perusi G, Meattini M, Dashi E, et al. Childhood neglect, the neglected trauma. A systematic review and meta-analysis of its prevalence in psychiatric disorders. Psychiatry Res [Internet]. 2024 [cited 2025 Feb 12];335:115881. Available from: https://doi.org/10.1016/j.psychres.2024.115881
  » https://doi.org/10.1016/j.psychres.2024.115881
• 33. Köhler-Forsberg O, Ge F, Hauksdóttir A, Thordardottir EB, Ásbjörnsdóttir K, Rúnarsdóttir H, et al. Adverse childhood experiences and psychological functioning among women with schizophrenia or bipolar disorder: Population-based study. Br J Psychiatry [Internet]. 2024 [cited 2025 Jun 19];224(1):6-12. Available from: https://doi.org/10.1192/bjp.2023.128
  » https://doi.org/10.1192/bjp.2023.128
• 34. Volpatto VL, Schuch JB, Rabelo-da-Ponte FD, Borgonhi EM, de Freitas JC, Ornell F, et al. A network analysis of early trauma and impulsive behavior in individuals with alcohol, cocaine, and polysubstance use disorder. Braz J Psychiatry [Internet]. 2024 [cited 2025 Jun 18];46:e20233265. Available from: https://doi.org/10.47626/1516-4446-2023-3265
  » https://doi.org/10.47626/1516-4446-2023-3265
• 35. Haselden M, Corcoran C, Araujo A. Family involvement in psychiatric hospitalizations: Associations with discharge planning and prompt follow-up care. Psychiatr Serv [Internet]. 2019 [cited 2025 Jun 18];70(10):860-6. Available from: https://doi.org/10.1176/appi.ps.201900028
  » https://doi.org/10.1176/appi.ps.201900028
• 36. Ådnanes M, Cresswell-Smith J, Melby L, Westerlund H, Šprah L, Sfetcu R, et al. Discharge planning, self-management, and community support: Strategies to avoid psychiatric rehospitalisation from a service user perspective. Patient Educ Couns [Internet]. 2020 [cited 2025 Jun 20];103(5):1033-40. Available from: https://doi.org/10.1016/j.pec.2019.12.002
  » https://doi.org/10.1016/j.pec.2019.12.002
• 37. Silva M, Góis C, Santos A, Figueiredo E, Ferreira V, Rocha J. Factors associated with length of stay and readmission in acute psychiatric inpatient services in Portugal. Psychiatry Res [Internet]. 2020 [cited 2025 Jun 20];293:113420. Available from: https://doi.org/10.1016/j.psychres.2020.113420
  » https://doi.org/10.1016/j.psychres.2020.113420
• 38. Siqueira DF de, Terra MG, Vieira LB, Moreschi C, Mello A de L, Soccol KLS. Care actions for the relatives of users of psychoactive substances: the perspectives of professionals and families. Texto Contexto Enferm [Internet]. 2019 [cited 2025 Jun 22];28:e20180022. Available from: https://doi.org/10.1590/1980-265X-TCE-2018-0022
  » https://doi.org/10.1590/1980-265X-TCE-2018-0022
• 39. Rowell-Cunsolo TL, Liu J, Hu G, Larson E. Length of hospitalization and hospital readmissions among patients with substance use disorders in New York City, NY USA. Drug Alcohol Depend [Internet]. 2020 [cited 2025 Jun 22];212:107987. Available from: https://doi.org/10.1016/j.drugalcdep.2020.107987
  » https://doi.org/10.1016/j.drugalcdep.2020.107987
• 40. Owusu E, Nkire N, Oluwasina F, Lawal MA, Agyapong VI. Readmission of Patients to Acute Psychiatric Hospitals: Determining Factors and Interventions to Reduce Inpatient Psychiatric Readmission Rates. European Psychiatry [Internet]. 2023 [cited 2025 Jun 22];66(S1):S899-S900. Available from: https://doi.org/10.1192/j.eurpsy.2023.1905
  » https://doi.org/10.1192/j.eurpsy.2023.1905

NOTES

• ORIGIN OF THE ARTICLE
  Article extracted from the thesis - Avaliação psiquiátrica de pacientes em um hospital universitário e pronto atendimento da região metropolitana de Porto Alegre por meio do interRAI Emergency Screener For Psychiatry, presented to Programa de Pós-Graduação em Promoção da Saúde, da Universidade Luterana do Brasil, em 2024.
• APPROVAL OF ETHICS COMMITTEE IN RESEARCH
  Approved by the Ethics Committee in Research of the Universidade Luterana do Brasil, N. 7.031.179/2024 and N. 3.291.055/2019 Certificate of Presentation for Ethical Appreciation (CAAE): 61491922.6.0000.5349 and 09164119.2.0000.5349. All participants signed an informed consent form.
• TRANSLATED BY
  Johanna Almeida de Mello.
• DATA AVAILABILITY
  The data are available upon reasonable request to the corresponding author.

Edited by

• EDITORS
  Associated Editors: Bruno Miguel Borges de Sousa Magalhães.
  Editor-in-chief: Elisiane Lorenzini.

Data availability

The data are available upon reasonable request to the corresponding author.

Publication Dates

• Publication in this collection
  13 Oct 2025
• Date of issue
  2025

History

• Received
  17 Feb 2025
• Accepted
  30 June 2025