Differences in Personal Recovery Among Individuals with Severe Mental Disorders in Private and Supported Accommodations: An Exploratory Study

1. Introduction

Severe mental disorders (SMD) are long-term conditions characterized by significant impairments in daily functioning, including challenges in work, education, social relationships, and self-care, ultimately limiting full participation in society [1,2]. These disorders encompass a range of mental health conditions such as schizophrenia, bipolar disorder, major depressive disorder, and other disorders [3,4]. Although they represent a relatively small portion of the population—approximately 4% of individuals with mental health conditions—people with SMD often experience a high level of need despite low service coverage, a phenomenon described by Killaspy as ‘low volume, high needs’ [5]. Because of their complex and persistent nature, SMDs require integrated and comprehensive approaches to care that combine medical treatment with psychosocial and community support [1,6,7,8].

Since deinstitutionalization, European recommendations [9,10,11,12] suggest to approach SMD using the personal recovery model, recognized as a key component of effective rehabilitation for people with complex needs [9]. Personal recovery is a strengths-based approach that supports individuals with severe and long-term mental disorders in leading meaningful lives, even with ongoing symptoms and functional challenges [3,13,14,15,16,17,18]. It emphasizes hope, self-determination, and empowerment through collaborative, person-centered [15], and evidence-based care [18,19], including shared decision-making [14,15,16,17]. Recovery involves recognizing individual strengths and goals, fostering equal partnerships between service users and professionals [14,15,16,17,20], and often includes both progress and setbacks along the way [21,22]. This approach upholds human rights and is linked to better symptoms, functioning, quality of life, satisfaction with care, and reduced service needs [14,23,24,25,26].

In detail, personal recovery care is recommended to be developed not in hospital settings but in community-based care that should have to be prioritized to avoid people with SMD marginalization and discrimination [9,10,11,12] . The implementation of the recovery model in Community Mental Health Services (CMHSs), particularly through evidence-based recovery-oriented practices [14,17,27], has resulted in improved user self-management, self-efficacy, autonomy [15,28], as well as enhanced health and social outcomes [25,29,30]. These practices are considered essential for adapting CMHS to the needs of the current century and correlate positively with treatment outcomes improvement and reduced health costs [14,31,32,33].

In the European context, Italy stands out for its progressive mental health policies, which emphasize a balanced integration of community-based and hospital services [34]. This approach contrasts sharply with many other countries that continue to face challenges in effectively implementing deinstitutionalization strategies [35,36,37]. Notably, Italy and Iceland are the only European countries to have fully eliminated psychiatric hospitals [38].

In countries where psychiatric hospitals still exist, many individuals with psychosocial disabilities remain institutionalized rather than being integrated into their communities [35,36,37]. Italy, however, offers a different model. People with SMD in Italy live either in private housing or accommodation (PA) or in supported accommodation (SA) specifically designed for mental health needs. This system of supported housing is fully established only in Italy and England, both of which have developed comprehensive services aimed at promoting independent living. These services follow a progressive care model that allows individuals to gradually move from more intensive to less intensive forms of support as they acquire the skills necessary for autonomy and social inclusion. Support is tailored to the individual's evolving needs, with the ultimate goal of helping them move toward less supported—or fully independent—living arrangements. While this model provides personalized assistance and clear recovery goals for both service users and professionals, it also involves moving between different housing settings as individuals progress in their recovery journey [39,40].

This small, exploratory study aimed to investigate, for the first time in Italy, differences in the recovery process of people with SMD living in PA and of people with SMD living in SA in the CMHS of South Verona. The recovery process has been evaluated with the use of the standardized tool Mental Health Recovery Star (MHRS) [41,42,43,44,45,46].

2. Materials & Methods

2.1. Study Design

All study procedures adhered to the ethical standards of the relevant national and institutional committees on human research, as well as the principles outlined in the Declaration of Helsinki (1975, revised 2008). The procedures involving patients were approved by the Research Ethics Committee of the University Hospital Trust of Verona (reference 34950, dated 30/05/2018). Written informed consent was obtained from all participating patients and mental health professionals.

This small exploratory study was conducted from May 2017 to October 2018 at the South Verona CMHS, part of a system grounded in evidence-based care and the bio-psycho-social model [47]. CMHSs, which became central to Italy’s mental health care following deinstitutionalization, provide diagnosis, treatment, and support. Each district-based CMHS serves approximately 100,000 residents, enhancing accessibility [48].

Data were collected at recruitment (baseline -BL) and at a six-month follow-up (FU), aligned with the typical evaluation timeframe used in Italian rehabilitation settings, such as day centers and SA. Given the limited sampling pool, a modest number of participants was anticipated, and the study was therefore designed to be exploratory.

2.2. Participants

Eligible participants included mental health professionals from the South Verona CMHS who had been trained in the MHRS, were willing to complete assessments at two time points, and could recruit at least one service user under their care. Of the 45 professionals trained between May and October 2017, 19 met these criteria. The remaining 26 were excluded because they did not work at the South Verona CMHS (n=15), declined participation (n=8), or were unable to recruit a person with SMD (n=3). Most participants were female (78.9%), over one-third were medical doctors (psychiatry residents) (36.8%), and the majority (73.6%) worked in multidisciplinary community teams. Their average tenure was 137 months (SD = 122.2).

To support consistent MHRS use, professionals attended monthly supervision and educational sessions led by certified trainers. A total of 12 meetings, with a mean of 12 attendees each, covered topics including data collection, recovery-oriented practices, MHRS implementation, motivational interviewing, and shared decision-making. Trainers also offered individual support when needed.

People with SMD were considered eligible based on clinical and functional criteria consistent with international definitions of SMD [3,5] . Specifically, individuals were eligible if they were under the care of a trained key professional at the South Verona CMHS, had a confirmed diagnosis of a severe and persistent mental disorder (such as schizophrenia, bipolar disorder, or other disorder with significant functional impairment), lived within the service catchment area, were aged 18–65, and provided informed consent for participation in assessments at two time points. Functional severity and chronicity were assessed by the key professionals, who also verified eligibility. Individuals were excluded if they had moderate to severe intellectual disability [49] or were experiencing acute psychiatric hospitalization or severe psychopathological symptoms at the time of recruitment, with severity assessed using tools such as the Health of the Nation Outcome Scales (HoNOS). A total of 25 individuals meeting these criteria were identified and recruited by their key professionals.

2.3. Clinical Tools and Their Characteristics

Socio-demographic, service use, and clinical data for service users were retrieved from the Verona Department of Mental Health (DMH) database and the South Verona Psychiatric Case Register [50]. Assessment tools were chosen collaboratively by the research team, certified MHRS trainers, and experienced rehabilitation staff to ensure consistency with standard evaluation practices in South Verona CMHS rehabilitation settings. After receiving specialized training and under the supervision of the research team, key professionals conducted standardized assessments at BL and FU, including the assessments described in Table 1.

The MHRS, created by Triangle Consulting in 2007 for the Mental Health Providers Forum, supports collaborative, "expert-to-expert" partnerships between users and professionals. [55,56]. Widely adopted in the UK and abroad [50], it was translated into Italian in 2013, with over 8,000 professionals trained [41,51].

Using a 10-point star representing key life areas enclosed in 4 main domains: Physical and mental health (Managing mental health, Self-care, Addictive behavior), Activities and functioning (Living skills, Work., Responsibilities), Self-image (Identity and self-esteem, Trust and hope), and Networks (Social networks, Relationships), users and professionals assess progress with the 'Scale of Change', based on the Transtheoretical Model's five recovery stages [57] —from feeling stuck, passing by accepting help, believing, and learning to achieving self-reliance.

After assessment, a joint care plan is developed with up to three goals. Despite some concerns about inter-rater reliability, the MHRS is valued for its strong internal consistency and its focus on recovery, shared decision-making, and user empowerment [41,50,58,59,60,61,62].

Table 1. Assessments used in the sample.

Assessments	Items	Scoring	Use
MHRS – Mental Health Recovery Star [57,58]	10 domains	Each domain rated from 1 to 10, reflecting recovery stages (e.g., Stuck to Self-reliance); visualized in a star-shaped diagram	Supports collaborative care planning and monitors recovery-oriented outcomes
GAF – Global Assessment of Functioning [59]	Single global rating	Scale from 0–100; higher scores indicate better functioning; divided into 10-point ranges with descriptive anchors	Summarizes overall psychological, social, and occupational functioning
HoNOS – Health of the Nation Outcome Scales [60,61]	12 items	Each scored from 0 (no problem) to 4 (severe problem); total score 0–48	Assesses symptom severity and social functioning
MPR – Monitoring of the Pathway of Rehabilitation [62]	10 items each with 4 subitems	Domain-specific scoring using Likert-type or categorical scales (e.g., autonomy, social skills, work ability); may vary by local adaptation	Tracks progress in psychiatric rehabilitation and functional recovery
CAN – Camberwell Assessment of Need [63,64]	22 need areas	Each area rated as no need, met need, or unmet need, from both professional and patient perspectives; summary includes total, met, and unmet needs	Identifies clinical and social needs to inform individualized care planning

Table 1. Assessments used in the sample.

Assessments	Items	Scoring	Use
MHRS – Mental Health Recovery Star [57,58]	10 domains	Each domain rated from 1 to 10, reflecting recovery stages (e.g., Stuck to Self-reliance); visualized in a star-shaped diagram	Supports collaborative care planning and monitors recovery-oriented outcomes
GAF – Global Assessment of Functioning [59]	Single global rating	Scale from 0–100; higher scores indicate better functioning; divided into 10-point ranges with descriptive anchors	Summarizes overall psychological, social, and occupational functioning
HoNOS – Health of the Nation Outcome Scales [60,61]	12 items	Each scored from 0 (no problem) to 4 (severe problem); total score 0–48	Assesses symptom severity and social functioning
MPR – Monitoring of the Pathway of Rehabilitation [62]	10 items each with 4 subitems	Domain-specific scoring using Likert-type or categorical scales (e.g., autonomy, social skills, work ability); may vary by local adaptation	Tracks progress in psychiatric rehabilitation and functional recovery
CAN – Camberwell Assessment of Need [63,64]	22 need areas	Each area rated as no need, met need, or unmet need, from both professional and patient perspectives; summary includes total, met, and unmet needs	Identifies clinical and social needs to inform individualized care planning

2.4. Statistical Analysis

Descriptive data were reported as frequencies, means, and standard deviations. The normality of continuous variables was confirmed using the Kolmogorov–Smirnov test, thus permitting the use of parametric tests. Comparisons between people with SMD living in PA and in SA at BL were made using t-tests for independent samples (continuous characteristics) and Fisher’s exact tests (dichotomous characteristics). Changes between scores on the standardised assessment tools between BL and FU were investigated using the t-test for repeated measurements. All tests were bilateral with a significance level set at 0.05. Statistical analyses were performed using the SPSS 22.0 program.

3. Results

3.1. Mental Health Professionals’ Assessments

A total of 19 mental health professionals conducted the BL assessments. As shown in Supplementary Table 1, evaluations in private accommodations were more frequently carried out by medical doctors (42.9%), whereas in SA, support workers were the most frequent evaluators (36.4%). The average length of professional experience was higher among those working in SA compared to those in private accommodations, though this difference was not statistically significant (p=0.262).

3.2. Service User Characteristics at Baseline

As presented in Table 2, 25 people with SMD were included in the study: 14 in PA and 11 in SA. No significant differences were observed between groups in terms of age, marital status, education, employment status, age at first psychiatric contact, or clinical diagnosis.

People with SMD in PA demonstrated higher mean scores than those in SA in a specific MHRS domain, including " Physical and mental health" (p=0.039) and, particularly, in "Addictive behavior" (p=0.049). as shown in Table 3, at BL people with SMD in PA prioritized goals on “Physical and mental health” and “Networks, while those in SA prioritized goals in “Activities and functioning”.

In terms of clinical assessments, people with SMD in PA scored significantly higher on the GAF compared to those in SA (p=0.040), indicating better overall functioning.

People with SMD in SA had significantly higher levels of clinical and social needs, as indicated by both the CAN-Patient and CAN-Staff ratings. These included higher total needs (p=0.041 and p=0.008, respectively) and met needs (p=0.021 and p=0.006, respectively).

Table 3. Prioritizing area of the MHRS work plan by accommodation at recruitment and follow-up.

	BL Private accommodation (N=14)	FU Private accommodation (N=9)	p-value Fisher’s exact test	BL Supported accommodation (N=11)	FU Supported accommodation (N=11)	p-value Fisher’s exact test
Area/s of intervention
Physical and mental health	5 (35.7%)	2 (22.2%)	0.011	3 (27.3%)	5 (54.6%)	0.176
Activities and functioning	3 (21.4%)	3 (33.3%)		6 (54.6%)	3 (27.3%)
Self-image	1 (7.1%)	1 (11.1%)		1 (4.0%)	0 (0.0%)
Networks	5 (35.7%)	3 (33.3%)		1 (9.1%)	2 (18.2%)

Table 3. Prioritizing area of the MHRS work plan by accommodation at recruitment and follow-up.

	BL Private accommodation (N=14)	FU Private accommodation (N=9)	p-value Fisher’s exact test	BL Supported accommodation (N=11)	FU Supported accommodation (N=11)	p-value Fisher’s exact test
Area/s of intervention
Physical and mental health	5 (35.7%)	2 (22.2%)	0.011	3 (27.3%)	5 (54.6%)	0.176
Activities and functioning	3 (21.4%)	3 (33.3%)		6 (54.6%)	3 (27.3%)
Self-image	1 (7.1%)	1 (11.1%)		1 (4.0%)	0 (0.0%)
Networks	5 (35.7%)	3 (33.3%)		1 (9.1%)	2 (18.2%)

3.3. Clinical and Functional Changes of People with SMD from Baseline to Follow-Up According to Accommodation

Table 4 summarizes changes in clinical and functional outcomes of people with SMD according to accommodation from BL to six-month FU. Although both PA and SA patients showed improvements in romantic relationships and occupational status, these changes were not statistically significant. In the PA group, significant improvements were observed in all MHRS domains, alongside a significant increase in the overall MHRS score (p<0.001). Symptom severity decreased significantly (p<0.001), and functional autonomy also improved (p=0.041) as for functioning (p<0.001). Needs assessment indicated a significant reduction in total needs and unmet needs for both users (p=0.023 and p=0.003) and staff (p=0.006 and p=0.002) in the PA group. The ratio of met to unmet needs also improved.

Among those in SA, significant improvements were found in the overall MHRS score (p<0.001) and in domains such as "Activities and Functioning" (p<0.001) and "Networks" (p=0.001). Functional autonomy also showed a modest but significant increase (p=0.005). Significant increases were found in met needs from both user (p=0.008) and staff (p=0.015) perspectives, though reductions in unmet needs were not statistically significant.

Changes in recovery goal prioritization across accommodation types are shown in Table 3. Among users in PA, the proportion prioritizing "Physical and mental health" in their MHRS work plans significantly decreased from 35.7% at baseline to 22.2% at follow-up (p=0.011). In contrast, this domain became more prominent in SA, increasing from 27.3% to 54.6%, although this change was not statistically significant (p=0.176).

4. Discussion

The findings of this study offer an insightful perspective on recovery trajectories among individuals with SMD living in different types of accommodation in Italy. At baseline, both groups were broadly similar in terms of socio-demographic and diagnostic characteristics, supporting comparability. However, individuals in PA demonstrated better overall functioning and fewer clinical and social needs compared to those in SA. These differences suggest that, although all participants met criteria for SMD, those in SA were likely experiencing a more severe or complex form of the disorder at study entry.

Interestingly, despite these BL disparities, individuals in PA showed more consistent and widespread improvements across nearly all measured domains from BL to FU. This included significant gains in recovery outcomes, reduction in symptom severity, and improvements in autonomy and overall functioning, as well as notable decreases in both total and unmet needs with improved ratio of met to unmet needs suggesting a positive shift in perceived care adequacy. Functional improvements, such as social functioning and autonomy, are critical aspects of recovery. Individuals in PA might experience these improvements more rapidly due to fewer constraints and greater initial capacity for independence [51,52].

In contrast, individuals in SA also improved but their gains were more modest and concentrated in specific domains—particularly in areas they had prioritized at BL such as "Activities and Functioning". Reductions in unmet needs were not statistically significant in this group, though met needs did increase. This pattern suggests that while both groups benefited from recovery-oriented CMHS, individuals in PA may have had more capacity to translate these supports into broader functional improvements.

The greater severity at BL among SA residents likely influenced both their starting point and their rate of recovery over six months. SA residents often start with greater severity of symptoms, which can influence their initial recovery trajectory and rate of improvement. Higher initial severity can slow the rate of recovery, but individuals with severe symptoms can still show substantial improvements over time [53,54]. Furthermore, SA provides essential stability and support, which is crucial for those with higher initial severity, but their progress might be more gradual [54,55].

The shift in goal prioritization—from "Physical and Mental Health" in PA (which became less dominant at FU) to increasing emphasis on this domain in SA—highlights a positive alignment between perceived need and intervention focus, particularly among those starting from a more impaired BL. This alignment is facilitated by multidisciplinary approaches, recovery-oriented practices, and consumer-centered goal setting, ultimately leading to improved health outcomes [56].

4.1. Strengths and Limitations

While this study provides initial insights into potential differences in recovery trajectories between people with SMD living in private accommodation (PA) and those in supported accommodation (SA), further research is needed to fully understand the observed differences. As a small and exploratory study, these findings should be interpreted with caution. Improvements measured through the MHRS and other outcome tools may reflect natural changes over time, including possible regression to the mean, rather than the specific impact of the collaborative care planning approach. Additionally, the study was conducted within a single mental health service, limiting the generalizability of the results to other contexts.

Another important limitation is the absence of service user involvement in the study design, which may have affected the relevance and applicability of the research. Lastly, the use of multiple statistical tests may reduce the overall power of the study, meaning that the findings should be viewed as preliminary indications rather than definitive conclusions.

5. Conclusions

Our findings reinforce that personal recovery is possible across diverse levels of need, but also underscore the importance of tailoring intensity and duration of support. While those in PA showed more generalized recovery, people in SA improved in the specific areas they prioritized—suggesting the recovery model was effectively adapted to meet different stages and severities of need. However, continued and possibly longer-term support may be required in SA settings to close the gap in overall recovery progress.