Social Psychiatry and Psychiatric Epidemiology

, Volume 42, Issue 9, pp 747–752

A 15-year national follow-up: legislation is not enough to reduce the use of seclusion and restraint

Authors

    • Vanha Vaasa Hospital
  • Eila Sailas
    • National Research Centre for Welfare and Health
    • Kellokoski Hospital
  • Markku Eronen
    • Vanha Vaasa Hospital
  • Anna-Maija Koivisto
    • Tampere School of Public Health, University of Tampere
    • Tampere University Hospital, Research Unit
  • Jouko Lönnqvist
    • Dept. of Mental Health and Alcohol ResearchNational Public Health Institute
    • Dept. of PsychiatryUniversity of Helsinki
  • Riittakerttu Kaltiala-Heino
    • Tampere School of Public Health, University of Tampere
    • Psychiatric Treatment and Research Unit for Adolescent Intensive Care (EVA)Tampere University Hospital
ORIGINAL PAPER

DOI: 10.1007/s00127-007-0219-7

Cite this article as:
Keski-Valkama, A., Sailas, E., Eronen, M. et al. Soc Psychiat Epidemiol (2007) 42: 747. doi:10.1007/s00127-007-0219-7

Abstract

Background

Seclusion and restraint are frequent but controversial coercive measures used in psychiatric treatment. Legislative efforts have started to emerge to control the use of these measures in many countries. In the present study, the nationwide trends in the use of seclusion and restraint were investigated in Finland over a 15-year span which was characterised by legislative changes aiming to clarify and restrict the use of these measures.

Method

The data were collected during a predetermined week in 1990, 1991, 1994, 1998 and 2004, using a structured postal survey of Finnish psychiatric hospitals. The numbers of inpatients during the study weeks were obtained from the National Hospital Discharge Register.

Results

The total number of the secluded and restrained patients declined as did the number of all inpatients during the study weeks, but the risk of being secluded or restrained remained the same over time when compared to the first study year. The duration of the restraint incidents did not change, but the duration of seclusion increased. A regional variation was found in the use of coercive measures.

Conclusion

Legislative changes solely cannot reduce the use of seclusion and restraint or change the prevailing treatment cultures connected with these measures. The use of seclusion and restraint should be vigilantly monitored and ethical questions should be under continuous scrutiny.

Key words

seclusionrestraintcoercion legislation

Introduction

Seclusion and restraint are frequently used coercive measures in psychiatric treatment. Despite the scanty evidence for the beneficial effects of these measures [23], it is often stated that treating disturbed psychiatric patients is impossible without seclusion or restraint [12, 25]. In recent years clinical, ethical and legal debate has increased awareness of the controversial nature of these procedures. International recommendations and ethical codes have started to emerge [5, 6, 27], and the European Committee for the prevention of torture and inhumane or degrading treatment or punishment (CPT) has shown a particular interest in the use of seclusion and restraint [20].

Existing data indicate that the rates of seclusion and restraint vary from 0% to 66% among psychiatric hospitals [4]. The average duration of seclusion and restraint has also been observed to vary from 1.5 to 50.6 h across studies. Some indication exists that seclusion and restraint rates are fairly unchanged over time in the same hospital [25] and do not react to changes in the size, function or policies of the hospital [7]. However, successful programmes to reduce the use of seclusion and restraint have started to emerge at the individual hospital level [9, 11, 14, 18].

The reasons for the variation in the use of seclusion and restraint in different hospitals are unclear, but several clinical, non-clinical and methodological explanations have been suggested. Variation in the characteristics of the patients seems to have an insufficient explanation for the variation in the use of seclusion and restraint [12]. Divergent policies, treatment philosophies, staffing levels and training and organisational structure and climate of psychiatric units have been suggested to have substantial influence [17]. Comparing seclusion and restraint rates internationally is difficult due to different definitions of seclusion and restraint and differences in studied patient populations [12].

In many countries, the lack of consistency in the policies of using and monitoring the seclusion and restraint practices has evoked concern. New legislation, recommendations, professional guidelines and some court cases have started to emerge to control the use of coercive measures in psychiatry [24]. For example in the US, new federal regulations concerning the use of seclusion and restraint were introduced in 1999. The new legislation is considerably more restrictive than the previous one. The effects of its implementation have been reported in two separate studies. The overall number of seclusion and restraint incidents decreased by more than 50% during 3 months in three adult units and in one child unit at a medical centre [8] and 26% during a 2-year period in an adolescent hospital [10].

Despite an increasing number of legislative changes aimed to reduce the use of coercion in different countries, there is no research on the effect of these changes on a national level. The aim of the present study was to investigate the trends in the use of seclusion and restraint in psychiatric care in Finland during a period of 15 years characterised by legislative changes and by an explicit deinstitutionalisation policy in psychiatry, complicated by economical recession. Specifically, the study set out to investigate whether the legislative and policy changes succeeded, during a period of 15 years, in bringing reductions (1) in the total number of the secluded and restrained patients; (2) in the risk for being secluded and restrained; (3) in the duration of seclusion and restraint incidents; and (4) in the regional variation in the number of secluded and restrained patients.

Materials and methods

Setting

In Finland, the last two decades have indicated a remarkable change in how the legal system treats the patients’ right of self-determination. Foundation for this change was formed when Finland joined the European Convention of Human Rights in 1990 and implemented the fundamental rights in 1995. The restriction of patients’ freedom and the use of coercion in psychiatry are regulated by the Finnish Mental Health Act. In relation to using coercion, the Finnish Mental Health Act has been revised three times during three decades so that it has become more specific each time.

The first revision in 1978 [19] confined coercion solely to the patients in involuntary treatment. In 1991, the Mental Health Act [19] tightened the regulations regarding involuntary hospitalisation. Besides these changes in the Mental Health Act, the Act on the Status and Rights of Patients [1] went into effect in 1993 and strongly emphasised the inherent right of patients’ self-determination.

However, the Mental Health Acts in 1978 and in 1991 did not include explicit regulations about the seclusion and restraint practices. In practice, local instructions regulated the use of seclusion and restraint at the hospital level until the partly revised Mental Health Act [19] went into effect in 2002. Since then, the Act [19] has focused on regulating the use of coercive measures during inpatient care. The intent of the reform was to specify the reasons for limiting the fundamental rights of the involuntarily treated patient as well as to clarify and standardise coercive practices.

Finnish psychiatric care underwent a period of rapid deinstitutionalisation in the late 1980s. The aim was to increase community care resources to meet the needs created by the decreasing psychiatric bed capacity, but due to a serious economic recession in the early 1990s, these aims were not fulfilled [15]. The deinstitutionalisation policy resulted in decreases in psychiatric inpatient treatment days, although the number of treated patients did not decrease [22].

Procedure

The data were collected during a predetermined week in December of the years 1990, 1991, 1994, 1998 and 2004 using a structured postal survey of Finnish psychiatric hospitals by official request of the National Public Health Institute and the National Research and Development Centre for Welfare and Health (STAKES). The hospitals were reached on the basis of the Register of Institutions maintained by STAKES.

A letter was sent to the medical directors of the hospitals presenting the study and including the study protocol as well as structured forms for collecting information on seclusion and restraint incidents. The medical directors distributed the forms and instructions to all the wards permitted to use seclusion and/or restraint for working-aged (18–64 years) patients. The form was completed after each seclusion and restraint incident. Only the first seclusion and restraint incident of each patient in the study week was documented in detail. If the same patient had more than one incident during the study week, the number of consecutive incidents was recorded. If two incidents occurred in less than 1 h, they were considered as one incident. A reminder letter was sent to all the hospitals that had not returned the forms at the end of the study year.

Measures

In the structured form, seclusion was defined either as moving the patient to a locked seclusion room or locking up the patient in his or her own room. Restraint was defined as a mechanical restraint, i.e., confining the patient to a restraint bed. Physical or manual restraint, an order for isolation in an unlocked room, treatment on a locked ward or restraining the patient because of weak somatic health did not qualify as seclusion or restraint. The form included structured questions regarding the type (seclusion or restraint) and the duration of the patient’s first incident in the study week. The total number of incidents for each patient during the study week was reported.

In Finland, specialist level health services are divided into five tertiary-level catchment areas authorised by university hospitals. The tertiary-level catchment areas of each secluded and restrained patient was registered in the study form.

The hospital discharge register (HILMO) data

The number of all hospitalised psychiatric patients during the study week was obtained from the national HILMO-register authorised by STAKES. The number represents all patients in psychiatric hospital treatment because they all are in the risk of being secluded or restrained: a patient placed in a ward that has no seclusion or restraint can be transferred temporarily to a ward with facilities, if so required. All hospitals in Finland are obliged to supply a detailed document to the Register at the completion of each patient’s treatment period. Thus, it is possible to obtain information on the number of patients in inpatient treatment at any given time.

Response rate, missing data and study population

The response rate was 92.3% in 1990, 98.1% in 1991, 98.3% in 1994, 100% in 1998 and 98.2% in 2004.

The type (seclusion or restraint) of coercive measure was missing in one case (0.7%) in 1991. The duration of the patient’s first incident during the study week was missing in 10 cases (5.9%) in 1990, in eight cases (5.3) in 1991, in seven cases (4.5%) in 1994, in five cases (5.1%) in 1998 and in 12 cases (12.6%) in 2004.

In total, the material comprised 671 working-aged (18–64 years) secluded/restrained patients. The average age of the secluded/restrained patient was 39.1 years (SD 11.4) and 55.7% of them were male. In 64.2% of the cases, the main diagnosis was psychosis, in 9.5% substance use-related diagnosis, in 7.6% affective disorder, in 2.2% personality disorder, in 1.9% organic disorder, in 1.5% mental retardation and in 1.2% other main diagnoses. A diagnosis was missing in 11.9% of the cases.

Statistical analyses

The data were analysed using the SPSS statistical software version 11.5 and Confidence Interval Analysis (CIA) version 2.0.0 software. The national HILMO-register data were used to calculate the relative risk (RR) for being in seclusion and restraint during the study week at a psychiatric hospital. RR was calculated per study year and the year 1990 was used as the reference year. The Kruskall-Wallis test was used to evaluate the differences in the duration of seclusion and restraint incidents between the study years. To study a regional variation in the use of seclusion and restraint, the prevalence of the secluded and restrained patients per 100,000 Finnish inhabitants in the five tertiary-level catchment areas was used. The regional variation and its persistence over time were studied by comparing prevalence among these areas separately for each study year using χ2-tests.

Results

The number of seclusion and restraint incidents

The total number of seclusion and restraint incidents during the study week was 263 in 1990, 242 in 1991, 217 in 1994, 161 in 1998 and 129 in 2004.

The risk for being secluded or restrained

The total number of all hospitalised patients and the total number of the secluded and restrained patients decreased, but the risk for being secluded did not changed during the study time when compared to the first study year (Table 1). In the use of restraint, there was a slight decrease from index year 1990 to 1991, 1998 and 2004, but no linear trend was formed.
Table 1

The proportion (%) of and the risk (RR, 95% CI) for being secluded (S) or restrained (R) during the study week for all patients in psychiatric hospitals

Study year

Number of all psychiatric in-patientsa

Number of S patients

S patients of all psychiatric in-patients (%)

RR (95% CI)

Number of R patients

R patients of all psychiatric in-patients (%)

RR (95% CI)

1990

6,417

94

1.5

1.0

75

1.2

1.0

1991

6,103

107

1.8

1.19 (0.91–1.58)

43

0.7

0.60 (0.42–0.88)

1994

5,785

71

1.2

0.84 (0.62–1.14)

86

1.5

1.27 (0.94–1.73)

1998

5,170

59

1.1

0.78 (0.56–1.08)

40

0.8

0.66 (0.45–0.97)

2004

4,589

59

1.3

0.88 (0.64–1.21)

36

0.8

0.67 (0.45–0.99)

aNumber of all psychiatric inpatients is obtained from the HILMO-data

The duration of seclusion and restraint incidents

The duration of the seclusion incidents increased over the study time (χ2(4) = 36.111, p < 0.001). The median duration of seclusion was 335 min (min  = 15, max = 5,040, sum = 54,495) in 1990, 540 min (min  = 15, max = 7,515, sum = 71,049) in 1991, 672.5 min (min  = 40, max = 10,080, sum = 75,145) in 1994, 727.5 min (min  = 55, max = 8,520, sum = 61,265) in 1998 and 1,025 min (min  = 120, max = 7,590, sum =  71,081) in 2004. The duration of the restraint incidents did not change over the 15-year study period (χ2(4) = 2.455, p = 0.653). The median duration of restraint was 425 min (min  = 30, max = 3,045, sum = 47,249) in 1990, 322.5 min (min  = 50, max =  3,950, sum = 23,040) in 1991, 400 min (min  = 55, max = 10,080, sum = 63,055) in 1994, 442.5 min (min  = 45, max = 5,159, sum = 31,997) in 1998 and 420 min (min  = 30, max = 3,705, sum = 20,728) in 2004.

The regional variation in the number of secluded and restrained patients

The five areas differed from each other in the use of seclusion each study year, except in 1990 and in 2004. There were, however, systematic differences in the use of restraint among the five tertiary-level catchment areas across the study period (Table 2). Areas 4 and 5 consistently used less restraint than the other areas. While area 4 seemed to compensate a low use of restraint with a higher use of seclusion, area 5 displayed quite low figures in both. Area 3, however, seemed to prefer restraint over seclusion.
Table 2

The regional variation in the population standardised rates (per 100,000 inhabitants) of secluded (S) and restrained (R) patients between the five tertiary-level catchments areas according to study years

Study year

Number of S patients

Number of R patients

Area 1

Area 2

Area 3

Area 4

Area 5

χ2(4)

p-Value

Area 1

Area 2

Area 3

Area 4

Area 5

χ2(4)

p-Value

1990

1.8

2.7

1.7

1.6

1.6

4.2

0.375

1.6

2.7

2.6

0.4

0.3

30.6

<0.001

1991

1.9

2.5

0.8

2.9

2.4

10.4

0.034

1.2

0.9

1.4

0.6

0

10.6

0.032

1994

1.4

0.3

0.3

2.9

1.3

32.5

<0.001

2.1

2.6

2.3

0.4

1.0

20.6

<0.001

1998

1.1

1.0

0.3

1.9

1.0

10.3

0.036

1.0

1.3

1.0

0.3

0.3

10.4

0.034

2004

1.1

0.9

0.6

1.3

1.7

5.2

0.267

1.2

1.1

0.6

0.08

0.1

17.8

<0.001

Discussion

In Finland, the period between 1990 and 2004 was characterised by legislative changes resulting in more restrictive and specified regulations in the use of coercive measures. Regardless of these legal changes, the risk for being secluded or restrained during psychiatric hospital care did not decline over time. This result is both surprising and concerning.

Previously some smaller studies have reported that changing laws to be more explicit has had a positive impact on lessening seclusion and restraint practices [8, 10]. Furthermore, successful programmes to reduce the use of seclusion and restraint have started to emerge at the individual hospital level [9, 11, 14, 18]. However, in these earlier studies, only a few hospitals or wards were included and the follow-up times were relatively short. Moreover, studying the use of seclusion and restraint only in some hospitals or in a restricted area is vulnerable to misinterpretations because the research itself may influence the everyday practices in the units under scrutiny. The authors of the present study are aware of no previous similar nationwide studies concerning the impact of the long-term system-wide legal changes on the use of seclusion and restraint with such a long follow-up time.

In Finland, the issued laws consisted of principles about patients’ right to self-determination and instructions on how to report the seclusion and restraint incidents, but they did not contain explicit restrictions in the use of seclusion and restraint. Laws with such universal content could require a longer time to have an effect on the staff attitudes and treatment practices in psychiatric hospital treatment. Moreover, all the laws were implemented without any national educational programme and the personnel were not given any practical guidelines. Therefore, the prevailing treatment cultures were not really challenged. This study had a 15-year follow-up time, but even a longer one might have been required to detect the expected changes in attitudes.

In Finland, a rapid deinstitutionalising process meant a significant decline in the number of psychiatric hospital beds [15]. At the same time, the number of patients in psychiatric treatment did not decrease, but the average time spent in a psychiatric hospital shortened considerably. Hence, it can be assumed that the patients in psychiatric treatment have become more acutely ill and therefore are more challenging to treat. The new evidence from international reports of continuous problems with violence in psychiatric hospitals suggests the same [13]. Therefore, it can be assumed that without the legislative changes there has actually been an increase in the risk for being secluded or restrained when in psychiatric hospital treatment. To determine this, more knowledge concerning the changes in patients’ diagnosis and acuteness is needed.

While the number of the seclusion and restraint incidents decreased, the duration of restraint remained the same, and the duration of seclusion increased even threefold over the 15-year study period. This is in contrast to previous studies in which the duration of seclusion and restraint decreased after the introduction of stricter regulations [8, 10]. In this study the increase in the duration of seclusion in spite of the decrease in the number of incidents themselves can be an unintended side-effect of the stricter regulations in applying this measure: when the use of coercion becomes more complicated, patients are not released from seclusion as easily as before, because re-seclusion requires more complicated and time-consuming practices of registration. The result is alarming because it suggests that if the overall attitude to the use of restraint and seclusion does not alter, the coercive practices only take new forms, but do not diminish. This confirms the earlier assumption that it is the atmosphere on the psychiatric wards that has a great impact on the seclusion and restraint practices [2]. Further study is obviously needed to clarify the connection between the duration of coercive measures and the acuteness of the patients over time because of the deinstitutionalising process discussed already above.

Similar to previous studies [4, 16], there was a significant regional variation in the use of seclusion and restraint over the 15-year study period. One of the five tertiary-level catchment areas seemed to use seclusion instead of restraint, and another area preferred to use restraint rather than seclusion, while one area displayed quite low use of both. There is no evidence that this could be traced back to epidemiological differences in Finland [16]. Furthermore, legislation is uniform throughout the country, thus a regional variation in the use of coercion cannot be explained by different legal regulations like in previous findings [3]. Okin [21] studied the use of seclusion and restraint in seven state hospitals which had similar admission and discharge policies and operated under similar regulations in Massachusetts. He found that the hospitals used seclusion and restraint in widely varying degrees. Way and Banks [26] also found considerable variation in the rates of seclusion and restraint in New York State hospitals. Authors in both studies stated that factors relating to individual hospital practices and conditions have a significant effect on the seclusion and restraint practices. It can be concluded that legislation alone is not enough to equalise differences in coercive treatment traditions in different hospitals as they appear to be fairly permanent.

The data of the present study had the advantages of nationwide coverage, an exceptionally long follow-up period, and an unusually high response rate. Furthermore, special attention was paid to the collection of data so that it was performed in exactly similar fashion each study year. Due to Finnish registration practices, the authors were able to assess the entire country for the exact number of patients treated in the psychiatric facilities during the study week and report the coercion incidents in relation to patients at risk. The authors were not able to ensure that all the wards using seclusion or restraint in the participating hospitals responded. There is, however, no reason to assume that this would cause any systematic bias. The study was mostly accepted with interest and cooperation by all the psychiatric hospitals. As the data collection took place during only 1 week per year and only the first coercive incident of each patient in the study week was documented in detail, there was hardly any risk of the study wards growing slack. The two forensic psychiatric hospitals and the only Prison Mental Hospital in Finland all situated in different tertiary-level catchment areas were included. The authors ensured that this inclusion did not bias the results.

Conclusions

The use of coercion in psychiatry is a serious human right issue. Legislation, deinstitutionalisation policy and discussion in the society have all emphasised that the coercive measures should be used less in psychiatric care, but this is not reflected in practice. The reduction of the use of seclusion and restraint cannot be achieved solely through systemic measures like legislation. The regional variations in a country with one legislation show that the treatment traditions overpower the law in different hospitals. The increase in the duration of seclusion incidents indicates that the coercive practices should be vigilantly monitored and ethical questions should be under continuous scrutiny. The authors’ suggestion is to offer educational programmes for the staff of psychiatric hospitals together with laws with explicit regulations for the use of coercive measures.

Copyright information

© Springer-Verlag 2007