The current state of mental health care in Italy: problems, perspectives, and lessons to learn
After legislative changes in 1978, Italian psychiatry underwent a thorough overhaul, with the gradual closure of all Mental Hospitals. A nation-wide network of Departments of Mental Health now deliver outpatient and inpatient care, but also run semi-residential and residential facilities (the latter with 2.9 beds per 10,000 inhabitants). Hospital care is delivered through small psychiatric units (with no more than 15 beds). There are also many private inpatient facilities operating in Italy, and the number of private inpatient beds per 10,000 inhabitants exceeds the number of public beds; overall there are 1.7 acute beds per 10,000 inhabitants—one of Europe’s currently lowest numbers. There is marked quanti- and qualitative variation in the provision of out- and inpatient care throughout the country, and service utilization patterns are similarly uneven. Studies examining quality of life report a fairly high degree of patient satisfaction, whereas patients’ families frequently bear a heavy burden. In conclusion, the Italian reform law led to the establishment of a broad network of facilities to meet diverse care needs. Further efforts are required to improve quality of care and to develop a more effectively integrated system. Greater attention must be paid to topics such as quality of care and outcomes, public and private sector balance, and the coordination of various resources and agencies.
Keywordshealth services researchmental healthmental health servicesquality of care
Twenty-eight years have elapsed since a mental health law, called “Law 180”, radically changed the architecture of psychiatric care in Italy. This law had a far-reaching international impact, as shown by the large number of related papers and monographs published in international journals (Burti, 2001; de Girolamo, 1989; de Girolamo and Cozza, 2000; Piccinelli et al., 2002).
The present paper aims at providing an overview of the current state of mental health care in Italy nearly 30 years after the introduction of the reform law and deriving some general lessons from the Italian experience.
The background for law 180
In Italy, a country with 56,305,568 inhabitants (ISTAT, 2002), health care is provided to the entire population by the National Health Service (NHS), which has a similar structure to that of the British NHS. All citizens have access to unlimited health care coverage through “Local Health Units” (LHUs), each of which is responsible for a geographically defined catchment area. Access to health services is generally free of charge, although some fees are charged for specific medical examinations; medicines for major mental disorders are generally free of charge (except for benzodiazepines, in order to limit consumption). Not only does the NHS provide psychiatric care through the public health care system, it also generally covers access to private inpatient psychiatric facilities (but not outpatient private consultations).
It should be noted that the drug and alcohol abuse sector, both in terms of prevention and care, is managed outside the mental health sector: the data presented herein therefore does not refer to patients with only substance abuse problems.
The law and its developments
The 1978 psychiatric reform established four principal components: (1) the gradual phasing out of Mental Hospitals (MHs) through the cessation of all new admissions; (2) the establishment of General Hospital Psychiatric Units (GHPUs) for acute admissions, with a maximum of 15 beds each; (3) more restrictive criteria and administrative procedures for compulsory admissions; and (4) the setting up of Community Mental Health Centres (CMHCs) providing psychiatric care to geographically defined areas. Mental health services are currently organized through 211 Departments of Mental Health (DMHs), covering the entire country, each of which is responsible for a geographically defined area, usually corresponding to that of a LHU.
The law was essentially a “guideline” law, and Italy’s 21 Regions were entrusted with the specific tasks of drafting and implementing detailed norms, methods, and timetables for the organizational translation of the law’s general principles. It is difficult to determine the extent to which there have been any regional differences in the amount of national funding received to implement the reform, because Italian regions commonly receive governmental funding for mental health collectively with the rest of health care funding. Moreover, each region is divided into several LHUs and has a large degree of autonomy in allocating its overall health budget. These conditions have led, over time, to a rather uneven national situation, with different regions adopting different standards in terms of service provision and different organizational frameworks (de Girolamo, 1989; Piccinelli et al., 2002; Tognoni and Saraceno, 1989)—a phenomenon that sometimes makes it difficult to obtain an overall picture of the state of mental health care on a national scale.
In an effort to overcome this lack of nation-wide homogeneity and to provide quantitative standards for services, the Ministry of Health launched a multi-year “National Mental Health Plan” (NMHP, Ministero della Salute, 1999), which is still operational. For the first time ever, a set of (mainly structural) standards to be achieved were spelled out; yet, regions rarely consider the NMHP standards as mandatory.
The present paper is partly based on data from studies conducted at three different analytical levels: the first level includes nation-wide, large-scale health services research projects (e.g., PROGRES and PROGRES-Acute; a national day-centre survey) (de Girolamo et al., 2002, 2005; de Girolamo et al., in press; Maone et al., 2002). The results of these studies have been described and, wherever possible or deemed necessary, have been compared with similar studies conducted in other countries. In addition, we have also considered a survey on mental health facilities conducted in 2001–2002 by the Ministry of Health (Ministero della Salute, 2002) as well as routine statistics gathered by the Ministry of Health (Ministero della Salute, 2001); finally a survey of all DMH Directors, promoted by the Italian Psychiatric Association (Bassi et al., 2003) has also been taken into account. Rates were re-calculated by using the most recent national demographic census (ISTAT, 2002) figures as denominator.
The second data source is represented by multicentric international studies involving one or more Italian centres, such as the EPSILON project (Becker et al., 2002a), which compared patients in treatment at selected services in various European countries. The third source is that of results from studies conducted in distinct Italian catchment areas, which investigated specific service functioning modalities and can be useful for examining specific issues related to the overall implementation of community care.
Structure and process data: national surveys
This section discusses the current structure and provision of mental health services in Italy, and is largely based on the findings of the two PROGRES surveys. These studies have been carried out with a similar methodology: there was a first phase consisting of a national survey of all non-hospital Residential Facilities (RFs) (e.g., PROGRES) and all acute inpatient facilities (PROGRES-Acute). The second phase evaluated a random sample of facilities and patients living or admitted to these facilities in greater detail.
Mental health services in Italy, Circa 2006 (Progres Surveys And Ministry Of Health Surveys)
Beds per 10,000 inhabitants
Number per 10,000 inhabitants
Number per 150,000 inhabitants
(1) General Hospital Psychiatric Unitsa
Overall number of beds
Average number of beds
(2) University Clinicsa
Overall number of beds
Average number of beds
(3) 24-h Community Mental Health Centres
Overall number of beds
Average number of beds
(4) Other Public Facilitiesa,b
Overall number of beds
(5) Private Psychiatric In-patient Facilitiesa
Overall number of beds
Average number of beds
Overall Number of Acute BEDS (1–5)b
Non-hospital Residential Facilities
Overall number of beds
Average number of beds
Community Mental Health Centres
NHS Full-Time Staff
The closure of all MHs and the new care system
All 76 Italian MHs have now been shut down; when the reform law was approved in 1978, there were 78,538 MH residents: the current number of elderly, long-stay patients still living in only very few of the old psychiatric establishments does not exceed 2,000 individuals. Most former long-stay, elderly patients have been transferred to different types of RFs, nursing homes, etc.
During the deinstitutionalization process, there was apparently no increase in severe crimes committed by the mentally ill, as the actual number of people placed in the six Italian forensic MHs has not increased since 1980 (Priebe et al., 2005). Yet, the exact prevalence of mental health problems within the prison population outside of forensic MHs is completely unknown.
An important set of data has been recently collected by the Italian Psychiatric Association through the DMHs (Bassi et al., 2003); it was found that 1.1% of the Italian population, at January 1, 2001, was on active treatment in mental health services. On average every year 27.7% of people in treatment were first-contact users. In each DMH there were on average 17.2 psychiatrists and 45.4 nurses; each patient reported on average 4 contacts (e.g., outpatient visits, admissions, etc.) per year.
Acute inpatient facilities
The PROGRES-Acute Project (PROGetto RESidenze, Residential Care Project for Acute patients) found that, overall, Italy (except Sicily) has 4,113 public inpatient beds available. There are 266 GHPUs, with a total of 3,498 beds; 23 University Psychiatric Clinics with 399 beds, as well as 16 24-h CMHCs, located in the Friuli-Venezia Giulia (N = 10) and Campania (N = 6) regions, with a total of 98 beds. On the whole, in Italy there are 0.78 public inpatient acute beds for 10,000 inhabitants—approximately 20% less than the official national standard (1 bed per 10,000 inhabitants).
In the 20 regions covered by the present survey, there are also 4,862 private beds in 54 private inpatient facilities (private facilities, however, have no defined catchment area), with 0.90 beds per 10,000 inhabitants. All these facilities were built long before the reform law, and there have been no substantial changes in the number of private inpatient beds before and after approval of the reform law. Hence, there is a total number of 8,975 acute, short-term psychiatric beds (public and private) in Italy (excluding Sicily)—i.e., 1.72 per 10,000 inhabitants. Some areas of the country (particularly in the South) present an inverse relationship between the number of public (lowest) and private (highest) beds, such that the regions with the fewest public acute inpatient beds for 10,000 inhabitants (Lazio, Campania, and Calabria) also show the highest concentration of private inpatient beds. The number of overall (public and private) inpatient beds across different regions presents considerable variation, with a ratio of 8:1 for the regions with the highest (Calabria) and the lowest (Umbria) availability of beds, respectively. Of course, these figures do not match any demonstrated difference in the prevalence of severe mental disorders among these different regions.
In regions where public beds are scarce (as in the South), compulsory admissions are almost twice as frequent as they are in other areas of the country, in order to “oblige” hospitals to accept acute patients, at least in some instances. In any event, the percentage of compulsory admissions, as compared to the total of psychiatric admissions, has steadily declined from approximately 50% in 1975 (3 years before reform law introduction), to approximately 20% in 1984. Ten-years later, in 1994, this percentage had dropped to 11.8% of the total of public admissions (Barbato, 1998)—a percentage/figure that has since remained fairly stable, as revealed by the PROGRES-Acute study.
Different areas show considerable variability in length of stay, with the mean length of stay in the North-East being nearly twice as long in central and southern Italy, as also confirmed by regional studies (McCrone and Lorusso, 1999). Even the number of public beds varies greatly from the South to the North-East and Center (by nearly a 1:2 ratio), and the different availability of beds can account for the much shorter average length of stay observed for the South.
In 2001, 103,260 acute admissions to public facilities were recorded, for a total of 70,062 patients admitted, with a median number of days per admission of 11.4 in GHPUs, 17.8 in University Psychiatric Clinics, and 21.1 in 24-h CMHCs. Corresponding rates of psychiatric admissions and admitted patients per 10,000 population were 19.8 and 13.4, respectively. In that same year, private facilities yielded a rate of 6.9/10,000 private psychiatric admissions per year, with a median length of stay of 37.6 days. The percentage of compulsory admissions was 12.9% of all admissions to public facilities. Though most facilities have been operating for a relatively short time, many suffer from significant (sometimes very severe) logistic and architectural limitations.
Temporal trends in national admission rates have been investigated by Preti and Miotto (2000) who found that both total and first-contact admissions for mental disorders increased from 1984 to 1994. In particular, first-admission rates for schizophrenia, paranoia, affective psychoses, mania, and (to a lesser extent) major depression all increased. These findings are difficult to interpret and may be attributable to changes in diagnostic patterns, to varying degrees of diagnostic assessment accuracy, and to changes in cultural attitudes toward severe mental disorders, which occurred after a radical change in psychiatric care.
Phase 1 of the PROGRES (the national survey of all Italian RFs) has shown that there were 1,370 RFs in the year 2000 throughout all of Italy (each of which had at least four beds) for a total of 17,138 beds and an average of 12.5 beds each. Overall, there were 2.98 residential beds available for every 10,000 inhabitants, with considerable regional variability (ranging from a low of 0.67 to a high of 6.93). The RFs employed 18,666 professionals, 60% of whom worked full-time. The average number of staff per RF was 13.7 (range 6.9–21.0), with a ratio of patients to full-time staff of 1.4:1 (range 0.8–22.3) (Picardi et al., 2006a).
Three-quarters of the RFs presented no formal limitation on length of stay. In many respects, the environmental characteristics found in the survey suggested a home-like atmosphere in many RFs. For instance, average indoor space per resident was 36 square meters; there was often a garden available, and residents generally lived in two-bed rooms (Picardi et al., 2006a). This finding may account for a subjective quality of life that was observed to be somehow comparable to that of psychotic outpatients (Picardi et al., 2006b). Nevertheless, most facilities had restrictive rules regulating patients’ daily life and behavior (Santone et al., 2005), and some also suffered from substantial logistic limitations.
Characteristics of Italian Rfs: findings of Phase 1 of the PROGRES projecta
Intensity of care
Cover 24 h
Cover <24 h
Assistance as needed
Patients admitted in 1999
Patients discharged in 1999
Building shared with other health services
Divided into modules
Location of the facility
On former MH grounds
On general hospital grounds
All regions participated in Phase 2 of the project, with the exception of Abruzzo. A random sample of approximately 20% of the RFs assessed in Phase 1 was selected for each region, with only a few variations in selected regions which under- or oversampled the facilities to be assessed. The final sample included 265 RFs, i.e., 19.3% of all Italian RFs, which hosted 2,962 patients.
Patients were mostly males, with low education, and 85% had a pension, generally a disability pension (de Girolamo et al., 2005). Admission to a GHPU in the year prior to the survey was reported for 22% of residents. Forty-five percent of the sample was totally inactive, not even assisting with their facility’s daily activities. Two-thirds presented a diagnosis of schizophrenia, but co-morbidity or primary substance abuse were infrequent. RF managers judged nearly three-quarters of the patients surveyed as being appropriately placed in their facilities and considered that very few had short-term prospects for discharge.
Community mental health centres
According to the Ministry survey, 707 CHMCs were operating in 2001—an average of 1.88 per 150,000 general population. These centres deliver the bulk of outpatient and non-residential care, mainly through a network of 1,107 outpatient clinics. They provide individual consultations and visits, organize a variety of daytime and domiciliary care activities for the most severely and disabled patients, establish and maintain contacts with other health and social agencies, and provide emergency interventions. In most Regions, CMHCs operate 12 h a day, 5–6 days a week; most have a multidisciplinary team, including psychiatrists, psychologists, social workers, nurses, and educators.
Of the “intermediate structures”, the Ministry survey found 309 day-hospitals and 612 day-centres (1.6 per 150,000 population) for 2001. A national survey was conducted in 481 day-centres, with a response rate of 49.5% (Maone et al., 2002). The authors estimated an average of 1 day-centre place available per every 10,000 population. They also noted, however, that many CMHC-operated activities can be regarded as true “day treatment”, although they frequently do not receive this official denomination. Among the 238 day-centres surveyed, approximately 1/4 were organized according to general principles of milieu therapy. At the opposite end of the spectrum, many day-centres selectively focused on the acquisition of vocational skills, maintained strong ties with social cooperatives, and assisted patients in finding a job. Approximately 39% of patients attending day-centres had been in treatment for three or more years; indeed, nearly 60% of the participating facilities regarded patient discharge as one of the most challenging problems. In summary, this large survey of day-centres revealed points of strength (e.g., the possibility of engaging and maintaining even very severe patients in day treatment), but also highlighted areas needing improvement.
In 2001, 30,711 workers were employed with unlimited contracts in NHS mental health facilities (with a rate of 0.54 workers per 1,000 inhabitants). A futher 3,735 workers were employed by DMHs with time-limited contracts, and an unspecified number of employees worked in mental health services with time-limited contracts (e.g., many employees of RFs). According to the latest estimates available, there are 5,561 psychiatrists employed by public mental health services, 1,850 psychologists, and 14,760 nurses (with varying qualifications). In Italy there are also many psychiatrists or psychologists working in private practice only, but there are no official estimates on the number of private practitioners, type of practice, etc.
Quality of care: multi centre studies
Results of several surveys (Barbui et al., 1999; Munizza et al., 1995; Tibaldi et al., 1997; Tognoni, 1999; Tomasi et al., 2006) showed that in new services that cater both in- and outpatients, drug polytherapy is common, and prescription patterns are inconsistent and frequently at odds with current guidelines and recommendations, as found in similar surveys conducted abroad (Harrington et al., 2002).
Moreover, evidence-based psychosocial treatments (e.g., cognitive-behavioral techniques, psychoeducational programmes, interpersonal psychotherapy, social skills training, etc.) have not achieved wide diffusion (Morlino et al., 1993; Magliano et al., 2002).
Lastly, studies investigating patterns of patients’ contacts with services over time reveal frequently high drop-out rates (ranging from 46 to over 90%), especially among patients suffering from non-psychotic disorders (Percudani et al., 2002; Morlino et al., 1995; Rossi et al., 2002). This finding supports the assumption that mental health services focus much of their interventions on patients with psychotic disorders or severe personality disorders. Whether this latter situation has a positive or negative impact on community-based services remains an unresolved issue and deserves careful scrutiny.
Outcome data: local, multicentre, and national studies
Outcome studies are crucial for evaluting the results of the radical mental health care system change occuring after 1978. Unfortunately, however, only a relatively low number of such studies have been conducted. We can only summarize here the most important findings of these studies, and the interested reader is advised to consult the original sources for further details.
Continuity of care is often difficult to achieve. Of the 1,070 consecutive patients discharged from 21 GHPUs located in eight different Italian regions, only approximately half (N = 559) of the original cohort could be traced, which indicates a widespread loss of contacts after discharge. Moreover, a fairly high readmission rate (43%) was found (Barbato et al., 1992a, b) in a 6-month follow-up of this cohort of patients. In the entire Lombardy Region, only 1 in 6 of nearly all the first-ever admitted patients for the year 2000, who had had no previous contact with any type of public service, was in contact with public mental health services the following year (Lora et al., 2002).
Conversely, a study conducted with a representative cohort of 495 patients seeking care in the South-Verona CMHC showed that only 17% of these patients (and none with a diagnosis of schizophrenia) had dropped-out. Another interesting finding was that patients who were less satisfied with staffing skills and behaviours were more likely to drop-out (Rossi et al., 2002).
An innovative 3-year follow-up study has been recently published; it examined a 1-year-treated prevalence cohort of 107 patients with schizophrenia attending the South Verona community-based mental health service (Ruggeri et al., 2004). This study was one of the few measuring several types of outcome simultaneously, and it was found that mean symptom severity was stable in 74% of patients, worsened in 18%, and improved in 8%. While the total number of care needs did not significantly change, social and health needs did significantly decrease, and functioning needs increased. The met/unmet needs ratio tended to worsen in all domains except for basic needs, indicating that on-going needs tend to worsen. Quality of life was rated as medium–high by the majority of participants at baseline and did not change over time. Functioning worsened in 47% and improved in 30% of participants. Depending on the definition of “good” and “poor” outcome, the percentage of patients with a “poor” outcome ranged from 31% to 3%, and percentages of “good” outcome ranged from a high of 24% to a low of 0%. Hence, these results suggest that only a minority of patients with schizophrenia can be placed at the extremes of the range of possible outcomes, and that caution should be taken in extrapolating the results of studies based on single outcome indicators.
Trends in suicide rates
Suicide rates have been regarded by some authors as a proxy indicator of the quality of services. Yet, to date there have been conflicting results. Although several researchers did find a suicide rate on the increase over the last 20 years, the phenomenon is probably not only the result of changes in the psychiatric care system, but may be due to a variety of ecological factors (De Leo et al., 1997; Platt et al., 1992; Williams et al., 1986, 1987). Overall, it is difficult to draw any general conclusions about trends in suicide rates over time, or to make meaningful correlations with the reform law, given that these trends are likely due to a variety of socioeconomic and cultural factors. In any event, suicide rates in Italy remain among the lowest in the world (Levi et al., 2003).
Family burden is the only outcome dimension that has been studied in a large, nation-wide survey; in 30, randomly selected, DMHs (Magliano et al., 2001, 2002, 2005, 2006a, b) most family members in the schizophrenia group came out significantly affected by the patient’s condition: 97% reported feelings of loss, and 83% stated they frequently cried or felt depressed. Moreover, 73% had neglected their hobbies and 68% were virtually unable to afford going on any type of holiday due to the patient’s situation. Eighty-percent of the families were in regular contact with mental health services; 59% attended general informative sessions on the patient’s illness, and its treatment. Yet, only a very small percentage (8%) received any structured psychoeducational intervention.
Family burden resulted to be somehow lower in Northern Italy although different studies Reported different rates of burden of care in different Italian areas, from high-range (Samele and Manning, 2000; Lora et al., 2001) to mid-range (as shown in the Epsilon Study of Schizophrenia) in comparison with other European countries (van Wingaarden et al., 2003).
The changes that have taken place in Italy in the mental health sector have generally accompanied, and sometimes preceded, parallel changes internationally, showing a clear trend towards the downsizing of MHs and a growth in community-based models of care (de Girolamo, 1989; de Girolamo and Cozza, 2000). The total number of psychiatric beds (e.g., acute and residential beds) has decreased by 68% since the year of the reform (1978). Reductions of similar magnitude have occurred in the United States (Manderscheid et al., 2000) and in England (Glover et al., 2004), as well as in other developed countries.
With regard to the new services, the implementation of community-oriented models of mental health care has been successful in many parts of Italy and has made mental health care accessible to numerous individuals with various mental health needs, who in the past might have refrained from any contact with the old-fashioned asylum system. In several areas of the country, however, the quality of care delivered is still questionable, and there is much room for improvement.
Public versus private facilities
Although a large number of public beds were closed after Law 180, the number of private beds did not substantially change. Hence, a remarkable and possibly unforeseen consequence of the change in the overall architecture of the system is that private facilities currently provide as much as 54% of all acute, short-term psychiatric beds. For comparison purposes, the overall proportion of private medical and surgical hospital beds in Italy for the year 2000 was only 18% out of a total of 271,744 hospital beds (Ministero della Salute, 2001).
Unfortunately, no reliable, systematic data concerning type and quality of care provided by private psychiatric facilities is to date available. Moreover, no real accreditation system is in operation in Italy to guarantee the quality of these facilities. More detailed information will soon be available from the PROGRES-Acute project.
General hospital psychiatric units and residential facilities
A comparison of GHPU bed number per 10,000 inhabitants in 13 western European countries (excluding Luxembourg) using figures from the WHO Atlas on Mental Health Resources (WHO, 2005) shows that the number of acute (public and private) inpatient beds in Italy is lower than in the other 11 countries (the only exceptions being Greece and Spain), and much lower than the European median (2.25 beds per 10,000 inhabitants). This shortage is particularly severe in some areas of the country (e.g., especially the South) and should be urgently addressed. We have already noted herein that the shortage of public inpatient beds is associated with a shorter length of stay and a higher proportion of compulsory admissions.
Residential care for the most severely-ill patients requires special attention, and the PROGRES study has yielded a large body of information. The provision of residential beds shows high variability across regions, and variations in the provision of residential beds, at least in part, reflect regional differences in health planning (de Girolamo and Cozza, 2000; Fioritti et al., 1997; Tognoni and Saraceno, 1989). Moreover, a recent survey conducted in the Emilia-Romagna region in 2005 (and therefore 5 years after the first census) found a marked 50% increase in the number of RFs: this result perhaps suggests that a replication of the national survey today would reveal many more residential beds available (Regione Emilia-Romagna, 2006).
The paucity of data on the provision of RFs in other countries makes comparisons difficult. For instance, a survey carried out in eight districts surrounding London yielded 9.5 residential beds per 10,000—a substantially higher number than that found for Italy (Lelliott et al., 1996). Yet, in another UK survey of 35 districts, the figure was 4.3 (Faulkner et al., 1993). More recently, the Adult Mental Health Service Mapping has found figures similar to the Italian ones. For example, there were 2.6 residential beds per 10,000 population in the in some areas of England (Glover et al., 2004).
The PROGRES survey results also suggest that many RFs provide mostly long-term accommodation, with very low patient turnover, as also observed in the US (Geller and Fisher, 1993) and the UK (Trieman et al., 1998). For many chronic, disabled patients, RFs represent “homes for life” (Trieman et al., 1998), rather than rehabilitation sites, and this is perhaps their primary mission. At the same time, however, several studies have shown that living in small, domestic-like environments is associated with better quality of life and higher patient satisfaction than in traditional MH wards (Barry and Crosby, 1996; Lehman et al., 1986; Picardi et al., 2006b).
Variability of the reform implementation
In the first years after the law approval, several researchers noted a marked variation in service provision for different areas of the country, especially between the more wealthy areas of Northern and Central Italy and the poorer Regions of the South and the islands (e.g., Sicily and Sardinia) (Bollini et al., 1986; de Girolamo, 1989). Yet, variation in structure and process indicators can also be found in neighboring areas (Barbato et al., 1992a, b; de Girolamo et al., 1988; Salvador-Carulla et al., 2005). Hence, if Italy is to achieve uniform, evidence-based standards of care throughout the country, this imbalance between different areas and different services must be addressed.
Unresolved public health issues
A more difficult issue confronting policy planners and clinicians is the quality of care delivered in existing mental health services. Both drug utilization studies and a variety of other data point to an unsatisfactory situation, although many problems encountered in Italy today are also observed in most other European countries (e.g., poor prescribing practice, limited availability of effective psychosocial programmes, etc.) (Kohn et al., 2004). More in-depth evaluations are needed, some of which, however, will soon be available.
Additional information will also be required if we are to understand whether the relatively more favorable outcome of severely-ill patients (as compared to the past) can be attributed to more effective forms of treatment, the elimination of an iatrogenic environment (e.g., the MH), or to non-clinical factors (e.g., changes in the socio-economic environment, or wider availability of family support).
Conclusions: a few lessons
What main lessons, if any, can be learned from the Italian experience? The first is that the transition from a predominantly hospital-based service to a predominantly community-based service cannot be accomplished simply by closing MHs. Appropriate alternative facilities must be provided and this requires adequate time for planning and implementation. Once the initial enthusiasm for community care as a sort of “magic box” has faded, there is growing awareness of the fact that community care “is a service delivery vehicle. It can allow treatment to be offered to a patient, but is not the treatment itself. This distinction is important, as the actual ingredients of treatment have been insufficiently emphasized” (Thornicroft, 2000). The second point is that political and administrative commitment is essential. Community care is not, and will never be, a cheap solution (although MHs with minimally acceptable standards are expensive), but indeed is much more expensive! In fact, if community care is to be effective, investments must made in buildings, staff, staff training, and in backup facilities.
The third point is that “monitoring and evaluation are important aspects of change: planning and evaluation should go hand in hand and evaluation should, wherever possible, have an epidemiological basis” (Tansella and Williams, 1987). Hence, large studies conducted nationally and multicentric studies comparing Italy and other countries over the most recent years represent important (but only preliminary) achievements in this direction. At that point, a strong evidence-base should make further improvements in the system possible—although nearly 30 years of experience in mental health reform in Italy provides no guarantee that this will happen!