Factors predicting transferal after psychiatric emergency management in the elderly
*Ching-Hua Lin, MD, Kai-Suan Psychiatric Hospital, 130, Kai-Suan 2nd Road, Ling-Ya District, Kaohsiung 802, Taiwan. Email: email@example.com
Aims: The purpose of this study was to examine the demographic and clinical characteristics that differentiate between elderly and non-elderly visitors in the psychiatric emergency room (ER), and to identify factors predicting transferal after psychiatric emergency management in the elderly.
Methods: Data were collected over four years for patients who visited the psychiatric ER. The elderly were defined as patients older than 65 years old. Demographic and clinical characteristics were analyzed using the χ2-test for categorical data and t-tests for continuous data. Multivariate logistic regressions were carried out to find predictive factors associated with being transferred to a general hospital for elderly visitors in the psychiatric ER.
Results: Elderly patients made up 3.4% of all included visitors (n = 243) during the four-year period. The mean number of visits for elderly visitors was 1.63 ± 1.18, ranging from 1 to 7. The χ2-test and the t-test indicated that the elderly visitors were different from controls in many demographic and clinical variables. Multivariate logistic regression analysis showed that being transferred to a general hospital for elderly visitors in the psychiatric ER was associated with age (odds ratio = 1.32) and a greater number of previous psychiatric hospitalizations (odds ratio = 1.42). Patients without a thought-form problem also required transferal to a general hospital more often in our study.
Conclusions: The study suggested that elderly visitors in the psychiatric ER were a unique group, and specific considerations should be included in the intervention for these patients.
EPIDEMIOLOGICAL RESEARCH INDICATES that the geropsychiatric population differs significantly from younger age cohorts.1 A consistent pattern of under-utilization by the elderly has been found in existing mental health services, including community mental health centers, psychiatric emergency services, and office practices of psychiatrists and mental health specialists.2–4 Herst has suggested that the more common geropsychiatric emergencies include depression, confusional states, and paranoid disorders.5 Perez and Blouin found that with increasing age, more patients were diagnosed as having organic brain syndromes and major affective disorders. Among the geriatric patients who were admitted after psychiatric consultation in the emergency room (ER), the three most common symptoms were depressive features, side-effects of psychotropic medication, and hallucinations and delusions.6
Waxman studied ER records and found that elderly patients were more frequently referred by family or friends than self-referred. Also, the elderly received a diagnosis of organic brain syndrome more often than middle-aged patients, and were less often diagnosed with substance abuse or schizophrenic disorders.2
Psychiatric emergency facilities must begin planning to meet the needs of the increasing number of elderly persons who will undoubtedly be seeking their attention.7 In circumstances with limited medical resources, hospitalization is a relatively expensive modality of medical intervention. To prevent elderly persons from being hospitalized, factors predicting transferal after psychiatric emergency management should be surveyed first. However, information about elderly persons who use psychiatric emergency services is insufficient. Better knowledge of the characteristics and needs of elderly persons can help these facilities plan and deliver services.8
This paper describes research aimed at determining the characteristics of elderly patients who sought help at a psychiatric ER and also factors predicting transferal after psychiatric emergency management.
This study was conducted at Kai-Suan Psychiatric Hospital, a major psychiatric center in Taiwan, under naturalistic conditions. This 820-bed hospital is located in Kaohsiung City, the second largest city in Taiwan, and accounts for approximately 70% of the psychiatric beds in this city. Kai-Suan Psychiatric Hospital is the core hospital of the Psychiatric Emergency Network in the Kaohsiung area so patients brought by police or ambulance are usually sent to the emergency department of this hospital. The emergency service is staffed 24 h a day by attending psychiatrists, psychiatric residents and nurses. Compared to ordinary examination during hospitalization, the emergency service emphasizes quick assessment of the psychiatric symptoms, physical problems, and suicide and violence risk, and decides if the patient needs hospitalization, outpatient department (OPD) follow up, or other treatment facilities.
This retrospective study used data from the medical record system database at Kai-Suan Psychiatric Hospital. Permission to use this database for analysis and publication was granted by the facility's institutional review board to be exempt from the requirement for written informed consent and was conducted in accordance with the Declaration of Helsinki. All visits to the psychiatric ER were collected from 1 January 2001 to 31 December 2004. The elderly were defined as patients older than 65 years old.
An assessment form was completed by clinicians on each visit, providing demographic information, items measuring severity of psychopathology, and clinical psychiatric variables. The demographic variables included gender, age, marital status, employment status, residence status and years of education. Items measuring severity of psychopathology were quoted from an assessment battery used for assessing the clinical status of psychiatric inpatients,9 which was tested for inter-rater reliability and validity. We added items measuring severity of suicidality and aggression because patients who visited the psychiatric ER often had suicide or violence problems. Therefore, items measuring severity of psychopathology included appearance and behavior, thought form, thought content, mood and affect, insight, suicidality, hostility and aggression (shown in Table 1). Clinical psychiatric variables included previous psychiatric hospitalization history, age at onset, visiting ER for medications refill or adverse effect due to psychotropic medications, hoarding behavior, medication compliance, substance use, being physically restrained, the source of the patients' referral, diagnoses and result of management in the psychiatric ER. Age at onset was regarded as age at the first psychiatric symptoms. All data were collected through chart review by the research team, which was composed of six trained psychiatrists. In order to improve consistency of data collection, the research team held regular meetings to discuss problems in coding psychopathological symptoms or other data.
Table 1. Definitions of the psychopathological variables
|Appearance and behavior|| Delusions|| Elated mood|
| Withdrawal|| Hallucinations|| Irritable|
| Motor retardation|| Grandiosity||Hostility, aggression|
| Tension|| Somatic concern|| Verbal aggression|
| Motor hyperactivity|| Helplessness/hopelessness|| Physical aggression against self|
| Mannerisms||Suicidality|| Physical aggression against objects|
| Uncooperativeness|| Suicidal ideation|| Physical aggression against other people|
| Agitation|| Suicide attempt||Insight|
|Thought form||Mood and affect|| Yes|
| Disorganization|| Depressed mood|| No|
| Distractibility|| Blunted affect|| Denial of psychiatric illness|
| Disorientation|| Guilty feelings|| Denial of substance abuse|
|Thought content|| Anxious mood|| |
| Suspiciousness|| Excitement|| |
Bivariate analyses of elderly and non-elderly visitors were conducted using the χ2-test for categorical data and t-tests for continuous data. The multivariate logistic regressions were performed to find predictive factors associated with being transferred to a general hospital for elderly visitors in the psychiatric ER. All tests were two-tailed, and significance of tests was defined as an alpha of less than 0.05. Data were analyzed with spss version 10.0 (spss Inc., Chicago, IL, USA) for Windows.
Among all patients (n = 7143) in the psychiatric ER, 243 patients (3.4% of all included visitors) were defined as elderly visitors during the four-year period. The other 6900 visitors were defined as controls. A total of 50.6% were female (n = 123) and 49.4% were male (n = 120). The mean number of visits for elderly visitors was 1.63 ± 1.18, ranging from 1 to 7; and the mean number of visits for control visitors was 4.09 ± 3.04, ranging from 1 to 18.
Comparisons of demographic and clinical characteristics of elderly and control visitors are presented in Table 2. The χ2-tests indicated that the elderly visitors were more likely than the control visitors to have been identified by the clinician as married, as having appearance and behavior problems, as having thought-form and thought-content problems, as violent, as visiting the ER for medication refill, and as having hoarding behavior; and less likely to have been identified as having a previous psychiatric emergency visit, as male, as employed, as having insight and as having a substance-use or -dependence problem.
Table 2. Demographic and clinical characteristics of elderly and control visitors in the psychiatric emergency room
|Previous psychiatric emergency visit||81||33.3||3867||56.0||χ2 = 49.08||<0.001|
|History of psychiatric hospitalization||85||35.0||2390||34.6||χ2 = 0.429||0.512|
|Gender, male||120||49.4||4251||61.6||χ2 = 14.87||<0.001|
|Married||163||67.1||2015||29.2||χ2 = 157.88||<0.001|
|Employed||12||4.9||1410||20.4||χ2 = 35.87||<0.001|
|Housing status|| || || || ||χ2 = 6.93||0.074|
| Homeless||1||0.4||56||0.8|| || |
| Living with somebody||208||85.6||6140||89.0|| || |
| Living alone||23||9.5||397||5.8|| || |
| Living at institution||8||3.3||167||2.4|| || |
|Appearance and behavior||200||82.3||5261||76.2||χ2 = 5.17||0.023|
|Thought form||60||24.7||1321||19.1||χ2 = 4.49||0.034|
|Thought content||196||80.7||4388||63.6||χ2 = 29.85||<0.001|
|Suicidality||22||9.1||900||13.0||χ2 = 3.43||0.064|
|Mood and affect||220||90.5||6135||88.9||χ2 = .52||0.469|
|Violence||92||37.9||2135||30.9||χ2 = 5.08||0.024|
|Insight, positive||61||25.1||2470||35.8||χ2 = 12.12||<0.001|
|Visiting ER for medication refill||44||18.1||926||13.4||χ2 = 4.31||0.038|
|Visiting ER for adverse drug effect||4||1.6||282||4.1||χ2 = 3.67||0.055|
|Hoarding behavior||3||1.2||22||0.3||χ2 = 5.62||0.018|
|Poor medical adherence||214||88.1||5820||84.3||χ2 = 2.30||0.129|
|Substance abuse or dependence||20||8.2||2395||34.7||χ2 = 74.15||<0.001|
|Physical restraint used||24||9.9||663||9.6||χ2 = .01||0.937|
|Brought by police or ambulance||63||25.9||1818||26.3||χ2 = .02||0.883|
|Axis I diagnosis|| || || || ||χ2 = 570.79||<0.001|
| Schizophrenia||38||15.6||2477||35.9|| || |
| Other psychotic disorder||44||18.1||394||5.7|| || |
| Mood disorder||63||25.9||1641||23.8|| || |
| Substance-related disorder||9||3.7||1639||23.8|| || |
| Neurotic disorder||12||4.9||334||4.8|| || |
| Dementia, delirium, organic brain syndrome||77||31.7||231||3.3|| || |
| Adjustment disorder||0||0||79||1.1|| || |
| Mental retardation||0||0||14||0.2|| || |
| Personality disorder||0||0||15||0.2|| || |
|Physical illness||12||4.9||201||2.9||χ2 = 3.33||0.068|
|Result of management at ER|| || || || ||χ2 = 48.89||<0.001|
| Voluntary hospitalization||129||53.1||3413||49.5|| || |
| Involuntary hospitalization||12||4.9||413||6.0|| || |
| Follow up at outpatient department||81||33.3||2655||38.5|| || |
| Transferal to a general hospital||12||4.9||108||1.6|| || |
|Number of previous ER visits (mean ± SD)||0.6 ± 1.1||2.9 ± 5.8||t = 6.163||<0.001|
|Number of previous psychiatric hospitalizations (mean ± SD)||1.7 ± 3.0||1.6 ± 3.1||t = −0.538||0.591|
|Age at onset (mean ± SD years)||61.6 ± 16.7||27.2 ± 10.0||t = −50.786||<0.001|
|Years of education (mean ± SD years)||5.0 ± 4.0||10.7 ± 3.1||t = 25.898||<0.001|
Among classifications of axis I diagnosis, elderly visitors were more likely to have been diagnosed as having another psychotic disorder, a mood disorder and dementia/delirium/organic brain syndrome; and less likely to have been diagnosed as having schizophrenia, a substance-related disorder, an adjustment disorder, mental retardation and a personality disorder. Considering the result of management at the psychiatric ER, elderly visitors were more likely to receive voluntary hospitalization and transferal to a general hospital, and less likely to receive involuntary hospitalization and follow up at the OPD. The t-test indicated that the elderly visitors had a greater age at onset, a lesser number of previous psychiatric emergency service visits and a lesser number of years of education.
Table 3 shows the results of multivariate logistic regression analysis. Predictive factors of being transferred to a general hospital for elderly visitors in the psychiatric ER were age (odds ratio [OR] = 1.32) and a greater number of previous psychiatric hospitalizations (OR = 1.42); and a thought-form problem (OR = 0.42) was negatively associated with being transferred to a general hospital.
Table 3. Results of multivariate logistic regression analysis of independent predictive factors associated with being transferred to a general hospital for elderly visitors in the psychiatric emergency room (forward conditional)
|Number of previous psychiatric hospitalizations||1.42||1.08–1.87||0.012|
The study found that elderly visitors made up a small percentage of individuals who visited the psychiatric ER (3.4%), and that the elderly were relatively infrequent visitors. The percentage of elderly visitors in the psychiatric ER was similar to other studies (5–6%).10 The elderly visitors were different from control visitors in many demographic and clinical characteristics as presented in Table 2. According to our definitions of the psychopathological variables, appearance and behavior problem included uncooperativeness and agitation. A thought-form problem included disorientation; and a thought-content problem included somatic concern and helplessness/hopelessness. The above psychopathological symptoms and signs were compatible with those seen in dementia, delirium or depressive disorder patients in the elderly. There was inconsistency about violence in previous studies. Some studies revealed no difference of violence between different age groups;11,12 however, some studies revealed that younger patients were more likely to be violent.13 Our study found that elderly visitors were more likely to have been identified as violent than control visitors. The possible explanation was that impaired memory functioning frequently found in elderly also predicted object aggression.14 Hoarding behavior is often seen in obsessive–compulsive disorder. One study reported a prevalence of hoarding behavior in patients with dementia (22.6%).15 Hoarding behavior had also been reported in Diogenes syndrome in elderly patients.9 However, a comprehensive survey of hoarding behavior in the elderly is lacking. There were more women among the elderly visitors than men. This was possibly due to the longer life span of women.
Logistic regression indicated that the predictive factors of being transferred to a general hospital for elderly visitors in the psychiatric ER were age, a greater number of previous psychiatric hospitalizations and being without a thought-form problem. The elderly with greater age had more physical illnesses. A greater number of previous psychiatric hospitalizations often meant a longer duration of illness and treatment, which may result in more complications and adverse drug effects. According to our raw data, only two patients with thought-form problems (disorganization, distractibility, disorientation) had physical illnesses that were both hypertension. On the contrary, patients without thought-form problems had more moderate to severe physical illnesses including asthma and epilepsy, which may require transferal to a general hospital. This might explain why being without a thought-form problem was one of the factors predicting transferal to a general hospital.
This study has some methodological limitations that should be addressed. First, it was retrospective and the patient population was selected from a single catchment area, although covering a relatively large city. When interpreting the results of this study one should bear in mind that not all of the patients coded as ‘no previous psychiatric emergency visit or hospitalization’ were really new patients without previous psychiatric contacts. Second, the inter-rater reliability was not proved. In addition, we used information not verified by laboratory findings.
Despite these limitations, the novel aspect of our study is to compare demographic and clinical characteristics of elderly visitors with other visitors in the psychiatric ER, and to find predictive factors of being transferred to a general hospital for elderly visitors in the psychiatric ER. Furthermore, the large sample size and analysis of data collected over a four-year period were also rarely seen in previous studies. The findings mentioned above reinforce the necessity to address the needs of elderly visitors, particularly aiming at factors associated with being transferred to a general hospital.
Together these findings found that in a four-year period, elderly visitors made up about 3.4% of individuals seeking care in the psychiatric ER. There were significant differences among elderly and non-elderly visitors in many demographic and clinical characteristics. Multivariate logistic regression analysis showed that being transferred to a general hospital for elderly visitors in the psychiatric ER was associated with age and a greater number of previous psychiatric hospitalizations. In addition, patients without thought-form problems required more transferal to a general hospital possibly due to more physical illnesses in our study. These findings suggest that elderly visitors in the psychiatric ER are a unique group, and specific considerations should be included in the intervention for these patients.