Socioeconomic and familial factors in the involuntary hospitalization of patients with schizophrenia

Authors


Dr Isao Hattori, Shizuoka Prefectural Mental Care and Rehabilitation Center, 4-1-1 Yoichi Shizuoka-shi 420-0949, Japan. Email: isaohatt@gw7.u-netsurf.ne.jp

Abstract

Abstract  In the catchment area of Matsumoto Public Health Center in Japan, 44 schizophrenic patients admitted between April 1992 and March 1997 under the national policy Involuntary Hospitalization Ordered by Prefectural Governor (IHOPG) were compared with 61 schizophrenic patients admitted under another policy, Hospitalization for Medical Care and Protection (HMCP), during the same period. The socioeconomic and familial factors that led patients to IHOPG were evaluated in detail. The results revealed the following characteristics of IHOPG patients as opposed to HMCP patients: (i) their morbidity was of longer duration; (ii) they were more likely to live in a densely populated area; (iii) they were less likely to be financially self-sufficient; (iv) prior to admission they were more likely than HMCP patients to have avoided psychiatric examination and to have refused to take medication, and most had received no treatment before their hospitalization under IHOPG; (v) their relationships with family members were more likely to be poor; and (vi) the family was less likely to have cooperated with treatment or to have solved the patient's problematic behaviors. This investigation and the ensuing discussion revealed that a patient's schizophrenia-based danger to hurt self or others, which is an essential impetus for admission to IHOPG, does not arise suddenly but rather stems from multiple factors developing over time.

INTRODUCTION

In Japan, since the enactment of the Mental Health Law (1988), Involuntary Hospitalization Ordered by Prefectural Governor (IHOPG) and Hospitalization for Medical Care and Protection (HMCP) have been mandated as the main forms of involuntary hospitalization for psychiatric treatment. After the Mental Health Law was amended in 1993 and 1995, the Mental Health and Welfare Law was enacted to finally fill the need for mental health services in the community and to promote the welfare of persons with mental disorders.

Although hospitalization for mental health may be either voluntary or involuntary, the Japanese Ministry of Health and Welfare recommends striving to obtain the consent of the patient. Nevertheless, the Ministry also recognizes that involuntary hospitalization may be necessary when persons with mental disorders have special characteristics that prevent them from having insight into their illnesses.1

We compared schizophrenic patients admitted under IHOPG with those admitted under HMCP in the catchment area of the same health center (Matsumoto Public Health Center), and conducted a detailed analysis to clarify the socioeconomic and familial factors that drive some of these patients to IHOPG as opposed to HMCP.

A patient is admitted under IHOPG when he or she is  mentally  disordered  and  is  likely  to  hurt  either  self or others due to the disorder. In contrast, HMCP is invoked when the patient's guardian consents to hospitalization for medical care and protection, provided that a designated physician requires such hospitalization. IHOPG is therefore more closely associated with socially severe problems, including criminal offenses, than are other forms of hospitalization, namely HMCP and voluntary hospitalization.

This leads to important viewpoints by which to clarify in detail the course of factors leading patients not to HMCP but to IHOPG, even though both programs are for involuntary hospitalization. First, by first considering how to deal with socioeconomic and familial factors that lead to admission under IHOPG, we will understand how to apply the law appropriately. Second, clarification of such factors will help medical staff, administrative staff and families prevent their patients from being in danger of hurting themselves or others as a result of their mental disorders. These viewpoints will lead indirectly to better prognoses, so they will be very useful in the promotion of public acceptance of and support for schizophrenic patients in the community.

MATERIALS AND METHODS

Subjects

General situation in the catchment area of the Public Health Center, Matsumoto City, Japan

The Public Health Center in Matsumoto City has jurisdiction over two cities, two towns, and eight villages, encompassing  a  total  population  of  about  320 000. The population density is 360 per km2, as opposed to 330 per km2 for Japan. Both cities are commercial centers. One city has a population of about 204 000 (63.8%) and the other about 59 000 (18.4%). The remaining area, including the two towns and eight villages, has about 57 000 residents (17.8%) and is primarily agricultural.

Characteristics of subjects

The subjects included 44 schizophrenic patients admitted under IHOPG and 61 schizophrenic patients admitted under HMCP into the catchment area of the Public Health Center in Matsumoto City between April 1992 and March 1997. Both IHOPG and HMCP were administered according to the Mental Health Law from April 1992 to June 1995, during the implementation period, and then according to the Mental Health and Welfare Law from July 1995 to March 1997.1 For each patient, three psychiatrists including authors reexamined the diagnosis of mental illness using ICD-10 after his or her admission and reached a consensus on the diagnosis of schizophrenia. Subjects admitted under IHOPG were hospitalized by invoking any of four channels (Articles 23–26) of the Mental Health (and Welfare) Law. Twelve cases (27.3%) were processed through Applications for Examination and Custody (Article 23), 23 cases (52.3%) through Notifications by Police (Article 24), eight cases (18%) through Notifications by Public Prosecutors (Article 25), and one case (2.3%) through a Report by an Administrator of a Psychiatric Hospital (Article 26, Clause 2).

Of the 44 IHOPG subjects, 42 fell within Asada's definition of ‘mentally disordered offenders’ as ‘those who violate the Penal Code regardless of their responsibilities for committed crimes’.2 Table 1 shows the types of crimes committed under Articles 23–26. Violent crimes were predominant.

Table 1.  Crimes committed by subjects under Involuntary Hospitalization Ordered by Prefectural Governor categorized by articles 23–26
 Article 23 (n = 10)Article 24 (n = 23)Article 25 (n = 8)Article 26 (n = 1)
  • Forty-two of 44 IHOPG subjects were mentally disordered offenders.

  • Because most subjects committed more than one crime, the total does not agree with the number of subjects.

Heinous crimes
 Homicide 0 0 20
 Arson/playing with fire 6 2 00
 Robbery 0 1 00
Violent crimes
 Injury 813 40
 Assault1016 61
 Intimidation 814 41
 Blackmail 7 8 20
Pecuniary crimes
 Theft 2 1 30
 Fraud 0 0 00
Total4155212

METHODS

The medical staff, composed of designated physicians, public health nurses, caseworkers and hospital nurses, investigated the items described below. The authors and two other psychiatrists participated as designated physicians in cooperation with their hospitals. A pair of designated physicians investigated a subject on the condition that they did not work at the hospital where the subject was admitted. They obtained the cooperation of public health nurses, caseworkers and hospital nurses, and referred to the opinions of the psychiatrists in charge of patients.

After the patients began to improve through their hospital treatment, and began to gain an insight into their mental illnesses to some extent, they were informed about the purpose and content of the present study. The patients who did not consent to this investigation were excluded from the study, while the patients who did consent permitted us to use their data for this study.

Investigation items (Table 2)

Table 2.  Clinical characteristics of Involuntary Hospitalization Ordered by Prefectural Governor (IHOPG) subjects and Hospitalization for Medical Care and Protection (HMCP) subjects
VariablesIHOPG
(n = 44)
HMCP
(n = 61)
Statistical analysisP
No.%No.%χ2 (d.f. = 1)
  • Chi-square test with Yates’ correlation was used when sufficient data were obtained, and Fisher's direct probability calculation was used when data were insufficient. NS, not significant.

  • Geographical situations of dwellings in the catchment area were divided as follows: the two commercial cities were defined as two ‘urban areas’ and the two towns and eight villages as two ‘agricultural areas’. According to the population density, each area was further categorized as ‘high-density’ or ‘low-density’ (see ‘investigation items No.2’ in the text).

  • §

    As reported by the family.

  • As judged by three medical staff members in the interview with the family.

  • ††

    The family did not encourage the patient to undergo a medical examination or take medication.

  • ‡‡

    The family considered the condition of the subject's mental illness as normal, not as symptomatic.

The dwellings and occupations
 Urban area with high population density2556.81829.5 6.80<0.01
 Urban area with low population density 818.22642.6 5.90<0.05
 Agricultural areas with high population density 818.2 1 1.6 <0.01
 Agricultural areas with low population density 3 6.81626.2 <0.05
 The patient takes part in his/her family business§ 4 9.1 813.1 NS
Livelihood
 Annual income of less than 200 000 yen4193.24980.3 NS
 Annual income of 200 000–1.5 million yen 3 6.8 3 4.9 NS
 Annual income of more than 1.5 million yen 0 0 813.1 <0.05
 Receiving a mental disability pension 4 9.11829.5 <0.05
Treatment situations
 Treatment continuing after first psychiatric examination 4 9.12541 <0.001
 Receiving treatment of own accord 715.9 813.1 NS
 Receiving treatment on encouragement of family 715.94370.528.4<0.001
 Refusal to receive treatment2761.4 711.526.8<0.001
 Taking medication of own accord 613.6 4 6.6 NS
 Taking medication on encouragement of family 4 9.13659 <0.001
 Refusal to take medication3272.72236.112.3<0.001
 No treatment just before admission3784.13354.1  9.04<0.01
 Receiving treatment just before admission 715.92744.3  8.13<0.01
Family situations
 Healthy2659.15183.6  6.65<0.01
 Suffering from somatic disease 613.6 5 8.2 NS
 Suffering from mental illness 920.5 5 8.2 NS
 Not cooperative with treatment¶,††1840.91219.7  4.66<0.05
 Poor relationship with the patient2965.92337.7  7.05<0.01
 Not understanding patient's mental illness¶,‡‡2761.43659 NS
 Unable to act to solve the problematic behaviors2045.51118  7.97<0.01
 Consulting with hospital1431.84167.211.5<0.001
Factors initiating exacerbation
 Attending school or starting work 613.63354.116.2<0.001
 Unknown3681.8142333.2<0.001
  • 1Basic characteristics of subjects: age, gender, educational attainment, marital status, age at onset of schizophrenia, age at first psychiatric examination, and duration of morbidity up to the time of the current hospitalization.
  • 2Dwellings and occupations: geographic situations of the dwellings, to what degree the patient took part in his or her family business, and the patient's school and/or work situations. For convenience, the geographic situations of the dwellings were divided into four zones encompassing the Center's catchment area: each city was its own zone, the two towns and one village formed the third zone, and the other seven villages formed the fourth zone. At the same time, the two commercial cities were also defined as ‘urban areas’, while the towns and villages were defined as ‘agricultural areas’. The population densities of the two urban areas were 770 per km2 (which we labeled a high-density urban area) and 340 per km2 (which we labeled a low-density urban area). The population densities of the agricultural areas ranged from 40 to 280 per km2. Furthermore, the two towns and one village that had population densities of more than 200 per km2 were defined as high-density agricultural areas. The other seven villages had population densities of less than 200 per km2 and so were defined as low-density agricultural areas.
  • 3Livelihood: A subject's annual income, and whether or not a subject receives a mental disability pension.
  • 4Treatment situations: Whether the patient received psychiatric treatment of his or her own accord or on recommendation of family, or instead refused treatment; and whether the subject took medication of own accord or on recommendation of family, or instead refused to do so.
  • 5Problematic behaviors, disarray in livelihood (Table 3): Whether or not behaviors included homicide, bodily injury to others, assault, intimidation, attempted suicide, self-injury, arson, destruction of personal property, theft, insult, robbery, blackmail, intrusion into a house, making noise, or nocturnal disturbances such as sleeplessness.
  • 6Family situation: the condition of family members’ physical health or mental health, the family attitude toward the patient's treatment, and the family members’ relationships  with  the  patient  were  noted. The item ‘not cooperative with treatment’ means that the family is not encouraging of the patient in undergoing medical examinations or in taking medicine. Likewise, ‘not understanding the patient's mental illness’ means that the family understands the condition of his or her mental illness as normal, not as symptomatic.
  • 7Factors initiating most recent exacerbation: attending school, starting work, or unknown.
  • 8Mental state at time current hospitalization under IHOPG or HMCP began: mental state on the day of admission was assessed using Positive and Negative Syndrome Scale scores.4
Table 3.  ‘Problematic behaviors and disarray in livelihood’ of subjects admitted under Involuntary Hospitalization Ordered by Prefectural Governor (IHOPG) versus those admitted under Hospitalization for Medical Care and Protection (HMCP)
VariablesIHOPG
(n = 44)
HMCP
(n = 61)
Statistical analysisP
No.%No.%χ2 (d.f. = 1)
  • Chi-square test with Yates’ correlation was used when sufficient data were obtained, and Fisher's direct probability calculation was used when data were insufficient. NS, not significant.

Homicide 2 4.5 0 0 NS
Injury to others2556.8 1 1.6 <0.001
Assault3477.32032.818.5<0.001
Intimidation2863.6 813.126.8<0.001
Attempted suicide1125 5 8.2 <0.05
Self-injury 818.2 4 6.6 NS
Arson/playing with fire 818.2 2 3.3 <0.05
Destruction of personal property2965.91016.424.8<0.001
Theft 613.6 2 3.3 NS
Insult1840.9 4 6.6 <0.001
Robbery 1 2.3 1 1.6 NS
Blackmail1534.1 0 0 <0.001
Wandering about1431.81524.6 NS
Intrusion into a house1329.5 0 0 <0.001
Abnormal sexual behavior 920.5 5 8.2 NS
Making noise 818.2 3 4.9 <0.05
Staying indoors 920.52337.7 NS
Awake at night1840.91321.3 NS

Statistical analysis

Data were analyzed using the following methods. To test independence between two variables composed of qualitative data, the χ2 test with Yates’ correlation was used when sufficient quantities of data were obtained; in the absence of sufficient data, Fisher's direct probability calculation was used. Student's t-test was used to assess the difference between two independent groups. A multiple logistics model was used to show the influence of each factor. For each test, statistical significance was set at 5%.

RESULTS

Basic characteristics of subjects

Among IHOPG patients, 36 (81.8%) of the 44 subjects were male, and their age at the beginning of the current hospitalization was 37.4 ± 12.9 years (mean ± SD, and so forth). The duration of morbidity up to the beginning of the current hospitalization was 14.9 ± 11.4 years. Among HMCP patients, 40 (65.6%) of the 61 subjects were male, and their age at the beginning of the current hospitalization was 34.2 ± 11.3 years. The duration of morbidity up to the beginning of the current hospitalization was 10.4 ± 9.99 years. Significantly high numerical values were found in the latter category.

Dwellings and occupations

Among the patients who lived in high-density urban areas, the IHOPG group comprised 25 subjects (56.8%) and the HMCP group 18 (29.5%), so there was a significant difference between the groups. As for living in high-density agricultural areas, the IHOPG group had eight subjects (18.2%) and HMCP one (1.6%); also a significant difference. Hence, the IHOPG group showed significantly higher rates of dwelling in high-density areas, whether urban or agricultural.

Livelihood

The Daily Life Protection System in Japan3 defines the annual cost of living in Japan as a minimum of 1.5 million yen per person. Therefore, we postulated that 1.5 million yen was the minimum annual income necessary for a patient to be financially self-sufficient. No subjects (0%) in the IHOPG group and eight (13.1%) in HMCP earned this base amount. Although the rate was significantly higher in the latter group than in the former, the great majority of subjects in both groups earned less than 200 000 yen prior to their hospitalization. In fact, 41 IHOPG subjects (93.2%) and 49 HMCP subjects (80.3%) earned less than 200 000 yen, so there was no significant difference between the groups.

Treatment situations

After their first psychiatric examinations, four subjects (9.1%) of the IHOPG group and 25 (41%) of HMCP continued to receive treatment; the rate was significantly higher in the latter group.

Seven IHOPG subjects (15.9%) and 43 HMCP subjects (70.5%) had taken psychiatric examinations because of family encouragement, and the HMCP group showed a significantly higher rate. Conversely, 27 subjects (61.4%) of the IHOPG group and seven (11.5%) HMCP refused examinations, indicating a significantly higher refusal rate among IHOPG subjects. Four subjects (9.1%) of the IHOPG group and 36 (59%) HMCP had taken medication because of family encouragement, revealing a significantly higher rate in the HMCP group. Thirty-two subjects (72.7%) of the IHOPG group and 22 (36.1%) HMCP refused medication, showing a significantly higher rate for IHOPG subjects.

As for treatment situations prior to the commencement of the current hospitalization, 37 IHOPG subjects (84.1%) and 33 HMCP (54.1%) did not receive psychiatric treatment, showing again a significantly higher rate for IHOPG.

Problematic behaviors, disarray in livelihood (Table 3)

The following problematic behaviors each showed a significant difference between the groups: injury to others, assault, intimidation, attempted suicide, arson or playing with fire, destruction of personal property, insult, blackmail, intrusion into a house and making noise. The IHOPG group showed a significantly higher rate than the HMCP group in every item.

Family situations

The families of 18 subjects (40.9%) of the IHOPG group and those of 12 (19.7%) HMCP did not cooperate  with  treatment,  so  the  IHOPG  group  showed a significantly higher rate. Twenty-nine subjects (65.9%) of the IHOPG group and 23 (37.7%) of the HMCP  had  poor  relationships  with  their  families, and again the IHOPG group showed a significantly higher rate.

The families of 20 subjects (45.5%) of the IHOPG group and those of 11 (18%) HMCP were unable to use action to resolve the patients’ problematic behaviors, and the IHOPG group showed a significantly higher rate.

Factors in prior exacerbations

Thirty-six subjects (81.8%) of the IHOPG group and 14 (23%) HMCP had been getting worse without initiating factors that we could determine, and the IHOPG group showed a significantly higher rate.

Mental state at beginning of current hospitalization

As for Positive and Negative Syndrome Scale scores at the beginning of the current hospitalization, positive scores were 31.5 ± 8.15 in IHOPG and 25.7 ± 4.95 in HMCP; IHOPG showed a significantly higher value. Negative scores were 23.3 ± 3.9 in IHOPG and 26.1 ± 4.92 in HMCP, and the latter showed a significantly higher value.

Risk factors leading patients to IHOPG (Table 4)

Table 4.  Social risk factors driving subjects to be admitted under Involuntary Hospitalization Ordered by Prefectural Governor (IHOPG). Logistic regression analysis for all subjects (44 schizophrenic patients under IHOPG and 61 admitted under Hospitalization for Medical Care and Protection (HMCP) with IHOPG as dependent variable
VariablesBetaSEχ2Odds ratio95%CIP
Urban area with high population density 1.950.59710.6  7.012.18–22.6<0.01
Agricultural area with high population density 2.821.26 5.0216.81.43–199<0.05
Intimidation 2.600.61817.713.4   4–45.2<0.001
The family did not act to solve the problematic behaviors 1.870.629 8.79  6.471.88–22.2<0.01
Constant−2.820.56724.7  <0.001

Logistic regression analysis was performed for the 44 schizophrenic patients admitted under IHOPG and the 61 admitted under HMCP, with IHOPG as a dependent variable. The following items were clearly risk factors leading patients to IHOPG: living in a high-density urban area (β = 1.95, P < 0.01), living in a high-density agricultural area (β = 2.82, P < 0.05), intimidation as a problematic behavior (β = 2.60, P < 0.001), family members’ inability to solve the problematic behaviors (β = 1.87, P < 0.01).

DISCUSSION

First, as for the duration of morbidity up to the beginning of the current hospitalization, we found some relationships between ‘duration of morbidity’ and ‘problematic social behaviors’. Tsuji and Kato reported that schizophrenic individuals who had severely injured others while being overwhelmed in a hallucinatory paranoid state tended to be relatively old and to have had long-duration morbidity.5 Humphrey et al. stated that a schizophrenic individual with a longer illness was more prone toward life-threatening behavior than was one with a shorter duration of the illness.6 As mentioned above, admission under IHOPG is closely related to problematic social behaviors, including criminal offenses. Our study also presented that the IHOPG group had a longer duration of morbidity than the HMCP group.

As for dwelling situations, the IHOPG group showed a higher rate of living in high-density areas than the HMCP group did; indeed, living in high-density agricultural areas, not just living in high-density urban areas, was a risk factor driving patients to IHOPG. To our knowledge, no other study has discussed the relationship between population density and involuntary hospitalization. Therefore this study is the first to draw a clearly significant relationship between high population density and IHOPG admission.

As for financial situations, most of the patients in both groups earned less than 200 000 yen annually. Therefore both groups had tight economic conditions. In fact, no IHOPG patient earned the 1.5 million yen that could make them financially independent. The results of our statistical analysis made it clear that it was more difficult for patients admitted under IHOPG to support themselves than it was for those admitted under HMCP. Some reports have touched on the relationship between financial situations and problematic social behaviors. Estroff et al. reported that the tight financial situations of schizophrenic individuals were related to violence against their family members,7 and Wessely et al. reported that the likelihood of a schizophrenic person committing a crime increased the lower his or her status in society was.8 Our study revealed the difficulty of patients admitted under IHOPG to support themselves economically. Also, as was found in the previous reports, our study revealed that poor financial situations were related to a schizophrenic person's danger to hurt him/herself or others (DHHO).

As for treatment situations, HMCP patients were more likely than IHOPG patients to have continued receiving psychiatric treatment after their first examinations, to have continued consulting with doctors, and to have continued taking their medication, even if the encouragement of family members influenced these results. On the other hand, patients admitted under IHOPG were less likely to have continued receiving treatment, and more likely to have refused consultation with doctors or to have taken their medication; also, they tended not to have been receiving treatment just prior to their current hospitalization. It is supposed that a schizophrenic patient's acceptance or rejection of psychiatric treatment affects the likelihood of subsequent outbreaks of DHHO.

As for problematic behaviors, these items showed significant differences between the two groups: injury to others, assault, intimidation, attempted suicide, arson or playing with fire, destruction of personal property, insult, blackmail, intrusion into a house, and making noise. All of these are categorized as DHHO. This is a foregone conclusion because a psychiatric examination and IHOPG are applied after problematic behaviors like these happen. Some items had no significant differences between the groups (Table 3): homicide, self-injury, uncleanliness, theft, robbery, wandering, abnormal sexual behavior, staying indoors, and staying awake at night. Some of these behaviors meet the criteria for DHHO. In our analysis of these items, however, the HMCP group never showed significantly higher rates than the IHOPG patients.

Family situations considerably influence the treatment situations of patients. Each item of ‘family does not cooperate with patient's treatment’, ‘family has poor relationship with the patient’ and ‘family did not take action to solve the patient's problematic behaviors’ was a factor in deteriorating treatment situations and in exacerbating the condition of the disease. The IHOPG group showed a significantly higher rate than the HMCP group for each item. This indicates that the root causes of what were ultimately IHOPG behaviors had existed under the surface for some time. It has been reported that relapse of schizophrenia is more likely to occur when the patient is in a stressful family situation.9,10 Our study presented in further detail how families influence IHOPG patients, either directly or through the treatment.

It often happened that the IHOPG group had no obvious trigger that exacerbated the disease. This made it difficult to identify the cause of the exacerbation in its early stage. In other words, a patient's deteriorating condition may go unnoticed by the people around him or her. This situation should link with the patient's problematic social behaviors.

As for the relationship between such behaviors and the condition of the disease, it is needless to say that these behaviors and the avoidance of psychiatric treatments easily emerge when positive symptoms are severe. Lindqvist et al. reported that the violent crime rate among schizophrenic persons was four times higher than that of the general population.11 Planansky and Johnstone12 and Smith and Taylor13 stated that schizophrenics’ aggressions occurred during the active phases of psychosis. Smith and Taylor pointed out that the majority of male schizophrenics who had committed serious sex offenses did so after the onset of their schizophrenia. The present study showed that patients admitted under IHOPG scored higher on the positive syndrome scale than those admitted under HMCP. This result is in agreement with earlier studies. Forty-two of the 44 IHOPG subjects in the present study were mentally disordered offenders as defined by Asada,2 and many of their crimes were violent ones (Table 1).

A report by Baxter was unique.14 He stated that the delusions of schizophrenic persons were not responsible for all of their observed excess assaults, even if their assaults were related to their delusions. Baxter's report might indicate that their psychogenic reactions15 deriving from exacerbation of the disease contributed to their problematic social behaviors. Similarly, Nester mentioned that four fundamental personality dimensions (impulse control, affect regulation, narcissism, and paranoid cognitive personality style) operate jointly, and in varying degrees, as clinical risk factors for violence among persons with mental disorders including schizophrenia.16 Hodelet found, in a secure hospital population, a highly significant association between psychosis and violence, but the strength of the association was not increased by the presence of imperative hallucinations or delusions.17

In addition to what has already been said, Walsh et al. pointed out that comorbid substance abuse considerably increases risks for violence in schizophrenic persons.18 We anticipate that these points of view about the relationship between schizophrenia and violence will be explored further.

ACKNOWLEDGMENTS

We thank Dr Takashi Shinozaki (Shinozaki Clinic) and Dr Hiroshi Shiina (Associate Professor, School of Economics, Shinshu University) for their valuable advice.

Ancillary