Compulsory Commitment to Care of Substance Misusers: International Trends during 25 Years

Law on compulsory commitment to care of alcohol and drug misusers (CCC) is a fairly common legislative option worldwide and a number of studies have indicated just how common. Porter et al. [1] reported to the WHO from 43 countries out of which 23 had CCC in mental health or social legislation (civil CCC). In addition, three federal countries had such law on state level. Thus, civil CCC existed in 26 (60%) of all 43 countries. They reported also on legislations on diversion to treatment from the criminal justice system, which mostly concern court orders, although they may at times include consent by the individual. Of the 43 countries, 18 (42%) had such diversion to treatment, either on country or state level. In all, 34 countries and territories (79%) had at least one of these options, i.e. either compulsory civil commitment or some coerced diversion to treatment from criminal justice, or both. In a second report, Porter et al. [2] reported on 79 countries (on country or state level). They found civil CCC in 51 of the 79 countries (65%), while coercive legislation associated with criminal justice legislation was found in 58 countries (73%). (Although Hong Kong and Macau are not separate countries, they are here handled as ‘countries’. After de-colonization they still have different legal systems from that of mainland China, based on the Chinese policy ‘one country, two systems’.) WHO Europe [3] studied 16 countries, including 10 newly independent states, previously parts of the Soviet Union. All reported law on CCC. Israelsson and Gerdner [4] combined these data for 90 countries. Within criminal justice legislation, only those laws mandating through court orders were counted, not laws that concerned diversion to treatment with full consent including possibility to withdraw. With these added to compulsory civil commitment, Israelsson and Gerdner [4] concluded that 82% of the 90 countries had some law on CCC at the eve of the 20th century. In a more recent study on 38 European countries, Israelsson [5] found that 74% had some such law. Room [6] reports from 147 countries that 42.5% have law on ‘special legislation for the compulsory treatment of substance use disorder’, while 20.5% have drug courts, and 52.2% have programs on diversion to treatment. The report does neither specify countries having each of these options, nor does it provide legislative texts or clear definitions of ‘special legislation on compulsory treatment’ or drug courts. Therefore, a total number of CCC, combining civil commitment and mandated care under criminal justice legislation, cannot be calculated from the report. Thus, although all studies seem to agree that CCC legislation is present in many countries, they vary somewhat in how common this phenomenon is.

These reports present the legislative situation at different times. Porter et al. [1] report on legislations up to 1982, while Israelsson [5] report on legislations up to 2009. Thus, there is a time span of at least 27 years between data collections. Using these reports and other available data, it should be possible to explore trends in whether or not countries choose mandated care as a legislative option, and if so, which types of CCC legislation these trends concern. Obviously, the different proportions in the reports mentioned may be related to variation in sampling, which in turn is due to the great linguistic, cultural or administrative difficulties involved in gathering information from many countries. There may also be inconsistencies in the definitions of the phenomenon – i.e. are legislations on CCC categorized in the same way? The focus of this study is primarily on trends during these years, but in order to study that, definitions and selection must be considered first.

CCC refers to the situation when the misuser is by law mandated to enter and/or to remain in care or treatment. It does not refer to forced medication. Porter et al. [1] state the following: ‘Compulsory civil commitment means the involuntary admission by judicial or administrative order, usually to an inpatient facility, for treatment of drug or alcohol dependence on the grounds stated in civil law’ [p. 45]. In addition, Porter et al. [1] note that legislation establishing a national treatment program for drug and alcohol dependence will not be complete without provisions for diversion of drug dependent persons from the criminal justice system. They also stated ‘one of the major policy questions facing legislators is whether diversion to treatment should be mandatory or merely made available for persons charged with certain serious offences, such as crimes or violence’ [p. 57]. Thus in theory, Porter and colleagues differentiate between mandated and voluntary diversion to treatment. The fact that the alternative is prison, instead of living as usual, does not merit calling this compulsory or involuntary. The individual still has a choice to accept care, even if the alternative options are limited. Although Porter et al. did not sum up mandated vs. voluntary categories, they provided legal texts from the various countries that make such categorization possible. The next WHO study [2] reported in a similar manner. It provides information on which specific countries that have civil CCC and diversion to treatment from the criminal justice legislation respectively, and it provides legal texts on both, making it possible to differentiate mandated from voluntary diversion to treatment. WHO Europe [3] describes compulsory commitment in a way that is applicable to both civil commitment and to mandated care in criminal justice legislation: ‘compulsory treatment comprises the nonvoluntary involvement of a drug- or alcohol-dependent person into a treatment process for supervision, care and treatment, which implies deprivation of the person’s freedom of behaviour’ [p. 19]. Still however, the tables provided in the report did not clearly differentiate between mandated and voluntary diversion to treatment from the criminal justice system, but legal texts on most countries made this possible.

In a study on CCC legislation based on the three WHO reports, Israelsson and Gerdner [4] used the available legal texts to distinguish mandated court orders from voluntary treatment as two different forms of diversion to treatment. Only the mandated court orders were seen as compulsory, i.e. CCC, while open contracts on spending time in treatment instead of in prison were not included. The definition used is: ‘CCC means that misusers are not given any legal choice to avoid and/or to leave care or treatment, i.e. they are by law mandated to care or treatment’ [p. 118]. The data collection for Israelsson [5] is based on the same definition. The demarcation line in this definition is congruent with the philosophic analysis by Tännsjö [7], who mentions a specific type of legal restrain, which is in fact based on consent by the individual at an initial stage to be held against his or her will later, if he/she then wants to leave treatment prematurely. It is called Homeric or Ulyssian coercion, based on the writings by Homer on Ulysses, who let himself be tied up in order to resist the songs of the sirens. This type is currently in use in Norway and Denmark within civil compulsory commitment to residential care, and was previously used in Sweden (before 1981). It was reported by Porter et al. [2] to be in use in Hong Kong and within criminal justice legislation in Qatar. These examples concern residential care, authorizing institutions to restrain the person from leaving prematurely. The same definition would also include the drug courts system that first developed in Miami-Dade County, Fla., USA, in 1989 as a response to an enormous increase of drug-related arrests [8]. These programs gained popularity in the USA and in 2010 over 2,500 drug courts were operational [9]. Since the 1990s, the drug court system has been established in many other countries [8,10,11,12,13]. For criminal offences to be handled in drug courts, initial consent by the offender is often needed, but the sentence takes the form of a court order in which the person is mandated to treatment. If he or she is not granted the right to withdraw consent, this court order is a form of CCC, but if he or she has the full right to withdraw consent at any time, it is not. The latter situation, when the person has the legal right to withdraw consent and leave treatment, is sometimes called ‘quasi-compulsory treatment’, due to the constrained choice when the only alternative to treatment is prison [14,15]. The orders from drug courts include not only residential treatment, but also placement in outpatient programs. Such programs lack the physical attributes to restrain the patient from leaving. Instead, violations to comply can be punished by the court by sanctions and penalties, which are intended to motivate the offender to progress through the program [8,12,16].

Thus, the definition can be applied when comparing data from previous WHO reports [1,2,3], as well as the data collected by Israelsson [5]. The latest WHO report and its chapter on compulsory care [6] provides a summary based on aggregated figures, but neither lists of countries with various legislations, nor legal texts that make it possible to distinguish diversion to treatment with full consent from mandated treatment. Looking at the questionnaire used for the latest WHO study, and provided to us by the Swedish respondents at the National Board on Health and Welfare, we find that the questions on diversion from criminal justice system do not distinguish between mandated care (court orders) and voluntary contracts. Thus, it will not be possible to make such a distinction without access to the actual legislative texts to analyze. In order to secure functional equivalence, we must, unfortunately, choose not to include this report despite the large number of countries included.

Obviously, the use of aggregated data of different samples of countries in different times is problematic if used to estimate trends, since that would invite selection bias. The ideal sampling would be comparing data on CCC laws for the very same countries at different times. Very few countries however are included in all reports. A discussion on international trends based on very few countries is likewise dubious. The variation in quantity of countries included in the reports is partly due to their special foci. Some studies had a worldwide scope in sampling, while WHO Europe [3] had its special attention to post-communist countries, and Israelsson [5] collected data from European countries. Even the largest report [6] lacks 15 countries that are present in others. Since sampling differs between reports, it is reasonable to combine them to get more complete datasets, provided that the same definitions on CCC can be applied. Information about when a law was first enacted, and if it has been changed or not since then is also used, is provided in Porter et al. [2] and in the data collection of Israelsson [5]. Additional information is presented in case studies on legislations, published in national reports or peer-reviewed articles. Further, the European Monitoring Centre for Drugs and Drug Addiction (EMCDDA) provides information on a number of European countries [17,18,19], and the WHO on a number of Asian countries [20], while Cooper et al. [12] give more information on drug courts. With such a combined strategy applied only when functional equivalence can be obtained, it will be possible to use available data for a large number of countries in order to detect trends in enactment, abolishment, or amendment of law on CCC.

The main aim is to explore how changes in CCC legislation developed in various countries, and whether it is possible to recognize trends over time, with three time points of observation – the mid-1980s, the late 1990s, and about a decade after the millennium. Did laws on CCC increase or decrease, or were there changes in types of legislation on CCC? Can changes be noticed in the length of mandated time or (for a European subsample) in the quantitative application of CCC? What was the contextual situation of changes in legislations among countries in the subsample, and were they accompanied by political debates?

The legislations in focus concerns compulsory care of adult alcohol and drug misusers (CCC). The definition of CCC implies ‘misusers are not given any legal choice to avoid and/or to leave care or treatment, i.e. they are by law mandated to enter and/or to remain in care or treatment’. Hence, laws on diversion to treatment implying full consent of offenders, i.e. with right to withdraw consent at any time, or treatment demanding full approval of prisoner during prison sentences, are not included. The laws explored are categorized in two types: CCC under criminal law and civil CCC. CCC under criminal law implies that a misuser is sentenced to treatment due to a criminal offence, i.e. when use, misuse or possession for own use is criminalized; due to other crimes related to use or misuse, e.g. dealing drugs, or due to other crimes committed by an addict but not necessarily related to his or her misuse. CCC laws outside criminal justice legislation – i.e. in mental health law or in social or special legislation are called civil CCC.

Three points of measure are used for observation, i.e. 1986, 1999 and 2009. Many data on law on CCC derive from three WHO reports [1,2,3]. In addition, external information on CCC law from various sources completes the dataset, such as case studies on legislations, national reports or peer-reviewed articles [12,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58].

Data on European countries for 2009 are obtained from the data collection on European countries conducted by the first author [5]. In that study, countries that are members of the European Council, or geographically surrounded by such (i.e. Belarus and Kosovo), were included. For countries represented in EMCDDA, the appointed National Focal Points (NFP) were addressed. For other countries, the governmental departments or agencies in respective country, or the national embassies situated in Sweden, were addressed to get contact information on sources for information. The participants were provided country-specific questionnaires, asking for information on CCC both in criminal justice legislation and in civil legislation (mental and social laws). The form was based on available previous knowledge about the country’s legislations, and included questions both on past and present laws on CCC such as: type of law, criteria for admission, statistics or estimates on number of applications, length in compulsory care, amendments in law, if political debates within or between political parties occurred in relation to legislative changes and topics of these debates. Data collection took place between March 2008 and October 2009. Eventually 38 European countries were included [see [5]].

Thus, combining all datasets, this study includes 104 countries. In addition, 21 states in federal countries are included. The stated year of enactments has guided completion of the dataset. In total, data on CCC legislation is accessible for 71 of the 104 countries for 1986, for 96 in 1999 and for 58 in 2009. For full review of sources for information on law each year, see table 1.

International trends in law on CCC are explored from the 3 time points of observation, and significance of change is tested using nonparametric tests, i.e. the sign test and Cochran’s Q test. The sign test is a test of two related samples which distribute a ‘sign’, either positive or negative, in the comparison between paired samples, here 2 years of observation, e.g. whether a dichotomy (yes or no on existence of a type of CCC legislation) changes in positive or negative direction. This change is tested for significance. In addition, trends in existence of law on CCC are analyzed among panels of countries providing data for all 3 observed time points, using Cochran’s Q test in exploring changes in existence of legislation within the panel at three points of measurement. Similarly, trends in parametric data on the application of these laws (number of cases) are tried with general linear models (GLM), repeated measures. GLM explores changes in means over time in same panels of subjects (here countries), using eta squared to estimate within-subject effects, and interpreted as follows: >0.01 is a small effect, >0.06 is middle, and >0.14 is a large effect [59]. All are provided in SPSS version 19. The 5% level of significance is applied.

An overview on reported law on CCC for each of the 104 countries or territories at the three times of observation, i.e. 1986, 1999 and 2009, is shown in table 1. In federal countries, these laws may be present on state level. Examples of legislations on state level are given in the table.

It can be noticed that among the explored countries, in all three times of observation, existence of some type of CCC of substance misusers is more common than not having such, i.e. 60 of 71 countries providing data (85%) in 1986, 83 of 96 (86%) in 1999, and 49 of 58 (84%) in 2009. CCC under criminal law is the most common type, and was present in 37 of 71 countries (52%) in 1986, in 55 of 92 (60%) in 1999, and in 41 of 59 (69%) in 2009. However, civil CCC is almost as widespread since 36 of 61 countries (59%) in 1986, 50 of 92 (54%) in 1999, and 22 of 52 (42%) in 2009 report such laws (mental health or social legislation). Many of these laws changed during these years. Laws on CCC have been enacted, abolished or replaced with new similar law. Nevertheless, not all types of legislative changes are seen in this table. From 1986 to 1999, 18 countries enacted new law on CCC, 6 abolished with no replacement, while 4 replaced one law for another type. In the same period, 20 countries revised old law. From 1999 to 2009, 11 countries abolished old law without replacement, 4 replaced old law with new of similar type, 4 replaced one law for another type, while another 14 revised old law. In all, 68 of the countries with CCC legislation enacted, abolished, revised or amended such law. The intensive legislative activities in most countries on CCC should illustrate that most of these laws are indeed active.

Whether or not the net outcome of these legislative activities results in more or less CCC laws cannot easily be seen in table 1. The stability in quantity of CCC legislation is therefore shown in table 2, using the available data on countries with more than one observation time.

Table 2 shows that the international presence of CCC legislations (any type) was apparently stable from 1986 to 1999. Only 2 countries removed CCC altogether while 5 enacted CCC without replacing older similar law. In the period 1999–2009, somewhat more countries removed than implemented CCC laws. When the whole period 1986–2009 is observed, a reduction is shown, although not reaching statistical significance (p = 0.070). Looking at civil CCC and criminal justice CCC separately, the tendencies appear more clearly. The net civil CCC was stable from 1986 to 1999, but sharply decreased in the period 1999–2009 when 12 countries removed and no country enacted new law (p < 0.001). Fewer countries were observed all the time 1986–2009, but still a significant reduction could be traced (p = 0.021). Looking instead at CCC in criminal justice legislation, the opposite significant trend is observed, with increases noted in all three time frames (p < 0.022).

This analysis uses all countries available for comparison. But the samples differ between time frames, with fewer data in later times. Trends are therefore also studied among panels of fewer countries from which data could be obtained at all three times. Stability over the years is tested with Cochran’s Q test (table 3).

Looking at any type of CCC a trend is apparent, with a reduction from 93 to 79% (p = 0.032), especially after the millennium. This is attributed to an even stronger trend in reduction of civil CCC from 68 to 44% (p = 0.003), which occurred between 1999 and 2009. CCC under criminal law shows a stable linear increase from 50 to 71% (p = 0.017). Thus, the two types of analyses, one using data from all countries with two points of comparison (table 2) and another analysis using panels with all three points of comparison (table 3), are in agreement that civil CCC is replaced with CCC in criminal law.

Changes do not only concern type of legislation, but also content. One such aspect is the maximum duration of time a person can be mandated to care according to the law. Some laws do not include any maximum time. Without a time limit, there is the possibility that persons are kept in mandated care for very long. Table 1 gives the available information about maximum times.

In 1986, 33 civil CCC laws stated the maximum time varying from 2 days to 10 years in defined time, while three laws had undefined maximum time. The median – when undefined time is at the upper end of the scale – was 1 year. In 1999, data on maximum time was stated in 56 laws on civil CCC, now varying from 2 days to 4 years, plus six laws with undefined maximum time. The median then was 6 months, i.e. half the previous length. In 2009, data on maximum time within mandated care cover 28 civil CCC laws, varying from 15 days to 6 years, plus 7 with no upper limit. The median is still 6 months.

CCC in criminal law more seldom has any defined maximum time, implying that mandated care may have very long duration. In 1986, maximum time was stated in 17 laws, varying from 3 months to 3 years. Since another seven laws had an undefined time limit, the median maximum time was in fact 3 years. In 1999, 35 laws had stated maximum time, varying from 14 days to 5 years. Another 14 laws had no defined maximum time. The median then was 5 years. In 2009, 32 laws defined maximum time, varying from 6 months to 5 years, and with ten laws lacking upper limit, the median still is 5 years. Thus, while we can notice a decline in mandated time within civil CCC, the opposite tendency is shown within CCC under criminal law. In fact, the median time of the latter is ten times that of civil CCC.

In the data collection on CCC in Europe 2009, the respondents were requested to provide data on how many persons were compulsorily committed in accordance with each law during the years 2002–2006. If national statistics were available, such were preferred, but if not the respondents were asked to give their best estimates. Only a limited number of countries were able to present such statistics or estimates, and, if available, it varied in forms. Some gave numbers of mandated persons per year, while others presented mandate decisions per year. Therefore, there is no point in trying to aggregate numbers from these different figures. It is, however, possible to use these numerical estimates of cases (i.e. persons or decisions) to explore if they are increasing, stable or decreasing.

Fifteen countries provided figures on CCC provisions according to 19 laws, i.e. some countries provided statistics from more than one law. Of these 19 estimates, three failed to give repeated measures, i.e. for more than 1 year. Another two had to be excluded due to invalid data (i.e. figures on all compulsorily committed according to mental health law, but not specifically due to substance use disorders). Thus, repeated data on numbers of applied cases were available for 14 laws from 12 countries: Azerbaijan, Croatia, Czech Republic, Finland, Germany, Ireland, Moldova, the Netherlands, Norway, Slovak Republic, Slovenia and Sweden, all but one based on official government statistics. For Germany and Sweden, statistics were provided separately for the two types of CCC. Data on persons in CCC within mental health legislation from the German state Nordrhein-Westfalen was obtained from LZG.NRW [37]. For 12 of these 14 laws, data were provided for all 5 requested years, and for the other two, data were available for 3 years only. The first and last of these were tried with nonparametric paired tests, showing 3 countries with decreasing numbers and 11 with increasing numbers, thus some tendency of an increase which approached statistical significance (sign test: p = 0.057). GLM analysis was applied in order to include measures at three points of measurement. The mean of available middle years were combined to one middle measurement. In order to make data more comparable with spherical attributes, all were transformed to cases per 1 million inhabitants. The time series of the three points of measurement (first, middle, last) showed sphericity (Mauchly’s W = 0.54; p = 0.025), and a clear tendency of an increase (within-subject effects: p = 0.015; partial eta squared = 0.275 with an observed power of 0.76). Thus, we found a strong increasing trend in numbers of cases during these years. When former communist Eastern states were compared to Western states in multivariate analysis, the between-group effect did not approach significance (p = 0.62; partial eta squared = 0.02). Thus, this tendency applied equally to East and West.

The data collection on European countries included a question if legislative changes had been in focus of political debates. Fifteen countries stated that such had occurred. The three top issues were: ethical considerations (10 of 15 countries), questioning the benefits of compulsory care (9), or the content of provided care (8). Obviously, the main topics may be combined. If the content of care does not have quality, the benefits of care might be reduced and thus the ethics in committing persons involuntarily to such care may be more in doubt. Interestingly, neither the costs of CCC, nor the legal security of committed persons, were debated that often. The issues for debate do not differ particularly on type of law of CCC, i.e. civil CCC (9 laws) or CCC under criminal law (6 laws) with exception of the most common topic, ethics, which is more often discussed concerning civil CCC. The result shows that law on CCC is a politically much discussed legal option. A closer look into some examples provides more contextual insight into these discussions. The information is based on the questionnaire replies (responders are mentioned in Acknowledgements) if not otherwise mentioned.

East Europe. Since the fall of the iron curtain, East European countries experienced rapid changes in social, economic, health and legislative respects. In former Yugoslavia, political changes implied actions of war between several neighboring countries. Croatia reports that the War of Independence in the 1990s rapidly increased the number of drug addicts [17]. In 1994, Croatia undertook steps to form nongovernmental organizations dealing with prevention, counseling and measures of harm reduction. Before and after the war, CCC under criminal law was in use, with a maximum of 5 years in treatment. Before a revision in 1997, the political debate questioned the content of CCC and if it was beneficent for misusers, but the decision was to keep it.

Macedonia was spared from outbreaks of drug epidemics and had through the 1990s, and still, a low prevalence of HIV/AIDS in spite of opioids being the primary abused drug [18,60,61,62]. The Yugoslav Penal Code of 1976, which included CCC, was replaced in 2004 by a new criminal law on CCC. The law’s enactment rendered a political debate on the benefits and legal security for misusers as well as the content of care and economic considerations.

The political transition in Czech Republic was peaceful, but the drug scene altered since the country became a transit for drug traffic. CCC under criminal law has been present in the country since the 1960s and is still in force; in addition, a law on civil CCC was introduced shortly after the velvet revolution, but reported as abolished in 2009. Before recodification of the criminal code in 1997, it was debated if CCC, called ‘protective treatment’, is beneficent. The content of care as well as organizational and economic considerations was also up for political discussions, but CCC remained intact after the revision.

The Russian Federation was seriously affected by a vast increase in drug consumption and the HIV/AIDS epidemic. Older RSFSR laws on CCC, still in force in 1986, were abolished after the political transformation, but new laws of both types were enacted in 1992 and 1996. From 2004 and onwards, several changes were made to the Russian Criminal Code, mainly introduced to cope with the overall deteriorated drug situation. Maximum time in treatment, which in older legislation could imply treatment for up to 10 years, is in the new laws not defined. In response to the increase of HIV and hepatitis C infections among injecting drug users (IDUs), Russia in 2003 adapted a note to article 230 directly aimed to cope with this situation. Despite extensive national political debate, the Russian legal position with presence of both types of law on CCC is intact, seemingly due to the epidemic increase in heroin misuse and blood-transmitted diseases among IDUs.

Nordic Countries. Sweden, Finland and Norway had previously, since the early 20th century, civil CCC within social legislation with long-term institutional placements (1 year, but up to 2 years under certain conditions). During the 1980s, treatment times were shortened – to begin with in Sweden in 1981, when it was first replaced with 2 months (4 months in certain cases), and later, in 1987, after an HIV epidemic outbreak among heroin IDUs in Stockholm, extended to 6 months [63]. In 1986, Norway and Finland shortened CCC in their social legislation – Norway to 3 months and Finland to 30 days (90 days in certain cases). The discussions commented on in the questionnaires refer to amendments and revisions to these later legislations.

Norwegian social law on CCC includes not only the 3 months of care mentioned above. In addition, there are two paragraphs which are – from a European perspective – less common. First there is the so-called ‘Homeric restrain’, i.e. an addict may apply to be kept later against one’s own will to leave treatment prematurely. The other is a special paragraph on compulsory care of pregnant misusers, to protect the fetus from harm. The latter was much discussed before it was introduced in 1996 [64]. Care of persons within social CCC increased in numbers, both concerning pregnant and other misusers. Despite that, CCC of substance misusers within a mental health act is more used than CCC in social legislation [65].

The Swedish stand on civil CCC has relatively strong acceptance within the society, according to governmental enquiries, but its content of care has been criticized for not living up to evidence-based standards despite high costs [66]. Recently, a government task force proposed integrating this type of care within psychiatric healthcare [67]. The minister of social affairs has suggested a new criterion for CCC to pregnant misusers, in order to protect the unborn child, but met opposition.

At the end of the 1990s, Finland experienced an increased use of amphetamines, but within a few years this outbreak stabilized [17]. Finland now reports a reoccurring national debate criticizing the non-use of its existing social legislation on civil CCC since the rate of alcohol-related deaths and liver diseases began to rise in 2004. Stenius [68] showed that civil CCC for substance misusers within psychiatric healthcare legislation is much more applied than CCC within social legislation. Beside the topic of well-being of misusers, Finnish debate on CCC has concerned the need for CCC for pregnant misusers to protect the unborn child [69]. A government appointed committee was in 2009 studying the possibility for such legal measures.

Mediterranean Countries. In Greece, addicts who committed drug-related offences are handled more lenient if their addiction contributed to the offence. Possession of small drug quantities for own use could imply that a prosecutor declares the addict as not punishable and that he should be invited to a therapeutic program. A prosecutor may also commit an offender to care as a probation order, i.e. CCC under criminal law. Greece is in great shortage of treatment facilities. Thus, whether compulsorily committed or not, the addict is seldom provided treatment.

Under the Cypriot penal law of 1992, convicted addicts may be sentenced to detoxification and rehabilitation in specific centers. Similar to Greece, however, Cyprus lacks treatment programs to exercise the full potential of its legislation on voluntary and compulsory treatment. In fact, the law on care of drug addicts is not formally implemented [17]. An update is planned in the current Cypriot National Strategy on Drugs 2009–2012 [70].

In 1993, Portugal abolished its law on civil CCC from 1983, which was seen as repressive and incriminating. Its benefits for the individuals were questioned. Instead harm reduction alternatives were introduced. A new legal direction was taken to decriminalize drug use and rely totally on voluntary care, emphasizing the responsibility of individual misusers to search alternatives to their drug dependence. Portugal experienced a significant increase both in drug use and in misusers seeking treatment, the latter from 23,654 in 1998 to 32,064 in 2001 [71]. Despite this and a threefold increase in diagnosed cases of HIV/AIDS from 1993 to 2000, Portugal continued its new policy.

West Europe. Switzerland experienced a heroin epidemic in the 1970s. The revision in 1975 of the federal narcotic legislation implied sanctions for use of illicit substances and measures to help drug addicts. However, in larger cities the drug situation deteriorated and both heroin misuse and HIV incidences escalated. Both types of CCC law (civil and criminal justice) were present at federal level in 1986, and legislation on civil CCC could also exist on state level. Civil CCC remained in 1999, but in 2009 no federal law on CCC exists. As with Portugal, the Swiss alternative aimed towards harm reduction alternatives [72,73,74]. Political debates concerned economic considerations and whether the new drug policy should be based on abstinence and/or harm reduction. An increase in treatment-seeking is reported [74].

The Netherlands is known for pragmatism and harm reduction strategies and the separation between cannabis and ‘hard drugs’, with permissions to sell cannabis through ‘coffee shops’. This pragmatism includes increased use of CCC. A recidivist addict offender can be placed in special closed treatment institution according to the penal code, and the mental health law implies compulsory care of misusers if they are in need for medical attention or dangerous to themselves or significant others [17,75]. The Netherlands also tried other legislative solutions. Between 2001 and 2004, the law Penal care facility for addicts (SOV) was in force. It was designed to commit recidivist addicts, who financed their addiction with theft, to placement in intensive treatment for up to 2 years in special institutions. Before SOV was enacted, it met political debate especially concerning the benefits of compulsory treatment and overall ethical considerations. Since 2004 this law is incorporated with the law Placement in an institution for prolific offenders (ISD) and is intended for routine adult offenders with addiction or psychiatric problems [17,76].

From 2000 to 2007, France experienced a sharp increase in cannabis use. The law that enables CCC has legal affinity to criminal law. It was object of parliamentary debates mainly concerning organization of care and ethical considerations of CCC. The increase in cannabis use did not influence national legislation on CCC but fuelled the debate on decriminalization as proposed by a slight majority of members in the Henrion state commission, reviewing the law in the mid-1990s [77,78]. However, the use of cannabis, as other narcotics, is still criminalized in France.

In the United Kingdom, the legislative framework for civil CCC is the Mental Health Act of 1983. The grounds for commitment are related to need for care criteria as well as danger for the misuser or significant others. The law cannot be applied solely due to substance dependence but is applicable in case of alcohol or drug-induced health problems, e.g. psychosis. The number of detentions yearly according to this law related to alcohol and drugs is not known but estimated as few. The law was questioned concerning ethics, the benefits of compulsory treatment and the content of care. After two decades of heavy increase of heroin use in the UK, first located to the London area in the late 1960s [79], a new legislative option to get misusers into care and to reduce criminality was introduced in England and Wales in 2004, by establishment of six pilot drug courts. The drug court system has been described as overall successful by authorities and misusers [80]. Drug courts are now operational in various locations in the UK.

In 2001, Ireland replaced its 1945 Mental Health Act to improve standards in care and treatment, and to comply to principles on human rights in detention of persons with mental illness. As in most mental health acts, the Irish act does not authorize commitment of misusers solely due to a person’s addiction, but it is applicable to prevent danger of the health of the person in need of care or the well-being of others. In addition, Ireland has joined the increasing number of countries introducing a drug court system. A pilot project in Dublin established in 2001 was described as a success, with declining recidivism in crime by 75%. Such programs on a national scale are now considered [12,17].

The study found indications of some international trends concerning legislation on CCC. There is still lack of data from many countries, and more such would surely give a more complete picture, especially concerning the later trends in non-European and non-Anglo-Saxon countries. Still, these findings represent a large part of the world, not least in number of inhabitants. The main findings are: (1) There is – on the one hand – a trend to decrease compulsory civil commitment to care for substance misusers, a trend that emerged after the millennium. (2) There is – on the other hand – an increase of compulsory care within criminal justice legislation. This tendency is slower, but steady since the last 25 years. (3) The total number of countries using any type of compulsory care decreased somewhat, especially after the millennium. (4) There is also a trend to reduce the maximum time within civil commitment to care, but contrary to this, (5) the maximum time an individual can be mandated to care within criminal justice legislation has increased a lot, and (6) according to available data in a sample of European countries, the number of persons mandated to care for substance use disorders increased strongly from 2002 to 2006.

This development is in fact contrary to the recommendations made by the WHO in 1967 [81]. At that time, the most common type of mandated care was within penal legislation, and substance use or dealing drugs were often the crimes for which the persons were punished. The WHO (1967) recognized that a growing number of countries saw these persons as sick and recommended treatment within healthcare organizations, if necessary ensured by civil commitment [81]. Although voluntary treatment was preferred, compulsory treatment was recognized as helpful in severe cases. How can we then understand this development? The study provides summaries on the contextual situation within some European countries at the time of legal amendments, which may shed some light on the main factors.

First we notice that amendments and other changes in legislation on CCC in many countries were introduced as responses to epidemics and aggravated misuse problems followed by increased criminality. The country descriptions contain many examples of various responses to drug-related epidemics. UK and Ireland introduced drug courts. Croatia, Russia, Sweden and the Netherlands amended legislation to increase the use of CCC. Finland now discusses actions in the same direction, while Greece and Cyprus had such ambitions but failed to implement them. There are also examples of alternative strategies, with Portugal and Switzerland relying on harm reduction measures as the clearest examples. Despite this, there seem to be tendencies of merging strategies. While the Nordic countries expand harm reduction programs within their traditional restrictive policies, the traditionally liberal Netherlands expands CCC laws within its strategy. Thus, it seems that harm reduction and drug-free treatment, including compulsory care for severe cases, can be combined within more pragmatic and comprehensive strategies by countries previously known for choosing opposite positions.

The relation between CCC and epidemics is well established historically and seen in many different cultural contexts. Civil commitment was introduced in many European countries to cope with severe drunkenness in the beginning of the 20th century [82,83]. Although this was called ‘care’, the content was often more directed to re-socialization of the ‘drunkard’ to the labor market [83]. After World War II, similar models of CCC expanded to cope with increased alcohol problems in East Europe [84], and with the drug epidemics in countries in East Asia, e.g. the opium epidemic in China, the amphetamine epidemic in Japan, and later the heroin epidemic in Singapore [85,86,87,88,89]. Often these drug epidemics occur in times of war, or in the afterbirth of war [90], as was also seen in Southeast Asia and in the USA after the Vietnam War, in Russia after the wars in Afghanistan and Chechnya, and as witnessed in the case of Croatia after the war of independence. Epidemics may also occur in peaceful times, as reported from a number of other countries. The association between CCC and drug-related epidemics often refer to increase in HIV/AIDS during the mid-1980s and 1990s [88,91,92,93,94]. At that time, injecting drug misusers (IDUs) played a role in spreading HIV in many European countries [95]. The responders from Russia and Sweden witness about such a connection. It is therefore possible that the stability of civil CCC in these years is explained by the HIV/AIDS epidemic. After the millennium, when new medicines helped to control this plague, CCC of addicts decreased.

A second finding is that ethical problems of compulsory care are claimed to be often discussed. Although some countries decided against CCC for substance misusers altogether (e.g. Switzerland and Portugal), other countries decided to accept CCC either within criminal justice legislation, civil (social or mental health) legislation, or both. The ethical debates seem to have been more frequent concerning civil CCC than concerning CCC under criminal law. There are important differences between these two legislative systems. It might be seen as less ethically problematic to sentence offenders to compulsory care. It can even be viewed as more humane to mandate them to placement in treatment than in prison. This is even truer when court orders are based on committed crimes other than illegal use or possession – the latter being less used in many countries. It is important to point out that court orders due to offences have not only rehabilitative intentions, but aim primarily to protect society. Civil CCC follows other logic, since persons are committed in order to secure provision of care more for the sake of the individuals.

According to the Madrid Declaration [96], the ground for psychiatric treatment without individual consent should be restricted to persons who are incapacitated or ‘unable to exercise proper judgment because of a mental disorder’. In most countries this applies first of all to severe psychosis and depression. The question whether or not this may also be applied to the severest cases of addiction seems to be a matter of some controversy. The mental health law of the UK rules out involuntary commitment based solely on diagnosis of substance dependence. This, however, is not untypical. Ireland has recently followed, and a somewhat similar policy has for long been the rule in – among others – Germany and the Nordic countries, where severe health problems or danger to self or others, etc. is needed for CCC, while dependence is not.

In the data collection on Europe for the present study, none of the countries could provide time series of CCC of substance misusers according to mental health acts on the national level, but such data obtained on state level from Nordrhein-Westfalen [37] indicate that severe substance misuse as a primary diagnostic category is about 20 times more frequently applied (per million inhabitants) than CCC within German criminal justice legislation. Denmark reports practically no cases with its Homeric social CCC, but Dahl Jensen and Paulsen [97] report that of those involuntarily committed to Danish mental healthcare, 5% were due to substance use disorders. Recently, scholars from Finland, Sweden and Norway discovered in all three countries – contrary to expectations – that substance misusers were more often compulsorily committed according to the mental health acts than according to the much more debated social legislations, and that these cases include many with severe substance problems as primary diagnosis [65,68,98]. Gerdner and Berglund [98] reported that the time in CCC within mental health law is sometimes very long. Since few countries produce national statistics on the diagnostic distribution of persons in CCC within mental health acts, it is not possible to know for sure if any, or how many, substance misusers are actually committed in those countries, including countries claiming not to have CCC for substance misusers.

The trend away from civil CCC seems mostly to affect social care legislation. A number of countries are also shifting type of civil CCC from social to mental health legislation, partly as a consequence of accepting substance use disorders as diseases that should be treated within the healthcare system, i.e. a form of normalization. It should be noted, however, that this is not without problems. Many countries with civil CCC within social legislation can present the statistical series that they lack concerning CCC in mental health legislation, probably since social (or special) legislation has been more in focus of controversy. A shift within civil CCC, from social to mental health legislation, may therefore be a shift from openly acknowledged CCC to hidden CCC. Additionally, the rehabilitative process in cases with severe addictions after initial treatment has to focus social aspects of life for the recovering persons for long time. We know from the debate in Sweden that there is a fear within the social work profession and within client associations that the shift will endanger this perspective.

The main trend, however, is a shift from civil CCC to CCC within criminal justice legislation, applied to greater number of persons and during longer times than before. Weisner [99] pointed more than 20 years ago at a tendency of ‘criminalization of deviance’ as a way of funding treatment. This might be seen as a manifestation of neoliberal policies, i.e. that governments would not take the same responsibility for its weakest citizens, but still has responsibility for upholding discipline [100,101]. The way it is implemented in many countries is, however, more than before influenced by modern knowledge on rehabilitation. There is a shift within penal legislation concerning nonviolent offenders in favor of the drug court system. It includes not only incarcerated treatment, but after initial residential treatment, the most part of rehabilitation is provided in community care and with a menu of relevant options. A major problem, however, concerns the question – Who will be treated? A shift from paternalistic treatment of those severe cases in need of treatment that cannot be provided voluntarily, to a protective treatment to discipline persons that would otherwise constitute a threat to others or society, has great implications for who is targeted by compulsory measures. This was shown many years ago by Segal [102] from time series of Italy, England/Wales and the USA. A shift from protective criteria focusing ‘danger’ tends to place young out-acting males in compulsory treatment, but do not include more vulnerable persons. These are better targeted with need-for-care criteria. There is therefore a risk that the shift from civil CCC to penal CCC will abandon the societal responsibility for the weakest substance misusers, more often elder, more often female and more often with severe somatic as well as mental health and severe social problems.

Addiction can be detrimental to the individual in terms of physical/mental health and in terms of social decline (housing, income, social relations) and it can be considered a criminogenic factor when the misuser engages in criminal behaviors that cause societal problems. The latter should warrant the misuser CCC in criminal law and the first civil CCC. How the two types of law correspond to various negative consequences will be explored in a forthcoming study.

We are grateful for the cooperation with EMCDDA REITOX network, and especially to the responders who provided information on the national debates and contexts, i.e. L. Vugrinec (Croatia), M. Savvidou (Cyprus), H. Gajdosikova (Czech Republic), K. Stenius (Finland), I. Obradovic (France), M. Markellou (Greece), S. Lyons and A. Daly (Ireland), P. Vaskova (Macedonia), M. Van Ooyen (The Netherlands), A. Strand (Norway), P. Vitoria (Portugal), M. Golichenko (Russia), M. Saraceni and M. Büechi (Switzerland), and G. Eaton (UK).