Residents’ Perceptions of Quality in Supported Housing for People with Psychiatric Disabilities

Abstract

The residents’ perspective of the quality of housing support for people with psychiatric disabilities living in congregate supported housing has been studied and a comparison has been made with the findings from those from a previous study in ordinary housing with outreach support. One-hundred and seventy-eight residents from 27 supported housing facilities in eight Swedish municipalities completed the Quality of Psychiatric Care–Housing (QPC-H) instrument. The highest quality ratings were found for: Secluded Environment, Encounter and Support, while Participation, Housing Specific and Secure Environment were rated at lower levels. Despite relatively high ratings, a majority of items did not attain the 80% cutoff point deemed as defining satisfactory quality of service. The residents in ordinary housing with outreach support rated higher levels for the majority of the QPC-H dimensions in comparison with those in supported housing. A conclusion is that the quality of care in supported housing facilities has a number of deficiencies that need to be addressed. Supported housing is generally rated as having a lower quality of care than in ordinary housing with outreach support. Suggestions for the content of staff training are made based on the results.

Introduction

A Mental Health reform took place in Sweden in 1995 that generated the transfer of many of the long-term mentally ill from hospital-based treatment to municipal, supported accommodation services (Swedish National Board of Health and Welfare, 1999). People with severe mental illness may develop a psychiatric disability, which in Sweden has been defined as a lasting incapacity to manage everyday life due to mental illness (Swedish Government Official Reports, 2006). The accommodation services provided for those deemed to have a psychiatric disability by the municipal social services in Sweden can be divided into two clearly differing types of facilities. The decision as to which of these facilities is most suitable is based on an assessment of each individual’s needs of support to manage everyday life (Brunt, 2002). The first of these, supported housing (SH), which was previously often called a group home is a congregate residential solution with on-site staff support from office hours to 24/7. The residents in SH generally have greater needs of support than those in the second type of facility, termed ordinary housing with outreach support (OHS), where a person lives in his/her own home and receives support from a professional with the aim of managing that situation. The support in OHS is generally provided from once a week to once or twice per day depending on individual needs. Supported accommodation services in Sweden are mainly staffed by nurse assistants and licensed mental nurses.

Early research into housing for people with psychiatric disabilities included a focus on comparisons between hospital-based and community-based services where improvements were revealed in terms of a higher level of social functioning, larger social networks and symptom reduction for those in housing solutions (Leff & Trieman, 2000; O’Driscoll & Leff, 1993). Satisfaction with housing was also expressed in terms of essential needs being met: such as privacy, food, a roof over one’s head, and affordability (Owen et al, 1996). Recent research has revealed satisfaction with some aspects of housing across a spectrum of housing models in western countries; including a generally high level of satisfaction with housing (Brolin, Rask, Syrén, Baigi, & Brunt, 2015),with safety and security (Brunt & Hansson, 2002; Harvey, Killackey, Groves, & Herrman, 2012), with opportunities for privacy (Tsemberis, Rogers, Rodis, Dushuttle, & Skryha, 2003), with fewer restrictions (Hanrahan, Luchins, Savage, & Goldman, 2001) and control and choice about housing support (Nelson, Sylvestre, Aubry, George, & Trainor, 2007). Negative aspects have concerned similar areas, including a lack of privacy and choice found in a national survey of housing needs in Australia (Harvey et al., 2012) and greater restrictions (Tsemberis et al., 2003), while an enforced togetherness was reported among those living in SH in Sweden (Bengtsson-Tops, Ericsson, & Ehliasson, 2014).

Research into housing for people with psychiatric disabilities in Sweden has shown some significant differences between those living in the two major types of housing solutions. The residents in SH reported greater satisfaction with their social life but less satisfaction with the performance of support than their counterparts in OHS (Brolin et al., 2015). In a comparison of residents in the two housing types focusing on activity, both groups rated themselves as under-occupied while those in SH were more satisfied with everyday activities and rated their housing higher on possibilities for social interaction and personal development (Eklund, Argentzell, Bejerholm, Tjörnstrand, & Brunt, 2017). Differences between the housing types could also be found in two studies with Grounded Theory methodology, where the main concern of those living in SH was “being deprived of self-determination” as a consequence of the organization and structures in the setting. The main concern of those living in OHS was, on the other hand, “the impossible mission in everyday life”, with a focus on their difficulties in trying to cope with a complex everyday existence (Brolin, Brunt, Rask, Syren, & Sandgren, 2016a, 2016b ).

Quality in psychiatric services has been described as a multidimensional concept and been described by patients (Schröder, Ahlström, & Wilde-Larsson, 2006), staff (Schröder & Ahlström, 2004) and relatives (Schröder, Wilde-Larsson, & Ahlström, 2007) as something positive and normative, i.e. what the care should be like. Schröder et al. (2007) state that the following dimensions constitute quality of care from the patient’s perspective: the patient’s dignity, sense of security, participation, recovery and care environment. To ensure the best possible quality of care for people with mental illness, a process of continual assessment and improvement is thus important (Crow et al., 2002). The measurement of patients’ perceptions of the quality of care have become a significant component in psychiatric care evaluation (Schröder, Wilde-Larsson, Ahlström, & Lundqvist, 2010) as patients can identify areas in need of change, which can inform staff and management about improvements that can be made. Patients’ perceptions of the quality of care have also been found to be associated with lower admissions frequency (Priebe, Saidi, Want, Mangalore, & Knapp, 2009), and better treatment outcomes and improved quality of life (Blenkiron & Hamill, 2003).

Two questionnaires have been developed to study the quality of care in community-based services for people with psychiatric disabilities in Sweden, one has focused on day center services (QPC-DA; Lundqvist, Ivarsson, Brunt, Rask, & Schröder, 2016) and the other on housing (QPC-H; Lundqvist, Rask, Brunt, Ivarsson, & Schröder, 2016). The latter, used in the present study of SH, has previously been used in a study of the quality of care in OHS (Rask, Schröder, Lundqvist, Ivarsson, & Brunt, 2017), which revealed a generally high level of quality of care according to the residents in these housing services. The dimensions Encounter and Secluded Environment were the aspects that were then rated as the two with the highest quality of housing service, while the dimensions Participation and Secure Environment were rated as those with the lowest quality.

We have, as shown above, some knowledge about satisfaction with various aspects of housing services for people with psychiatric disabilities. On the other hand, we have little or no knowledge about how the residents perceive the quality of care in SH as no previous studies with this focus have been performed. There is thus a need for more research in this field using instruments with good psychometric properties that have been developed from the residents’ perspective. Furthermore, differences have been shown between the two housing types and significant differences may also exist between these two housing contexts in terms of the perceptions of the residents about the quality in services. The aim of the present study is thus to investigate the quality of housing support for people with psychiatric disabilities living in congregate supported housing from the resident’s perspective. A second aim is to compare these findings with those from a previous study in ordinary housing with outreach support.

Methods

Participants and procedure

The sample of 178 residents in the present study was from 27 SH units in eight Swedish municipalities and one privately run unit and consisted of residents receiving support in these facilities. The data collection was performed during an eight-week period in March and April 2016.

All residents, who could understand and express themselves in Swedish, were invited to participate in the study by a contact person at each unit, who also ensured that the residents were cognitively able to complete the questionnaire in a valid way. The residents were provided with both oral and written information about the study, including the voluntary nature of their participation, which they could terminate at any time and a guarantee of confidentiality. The Regional Ethical Review Board at Uppsala University, Sweden (ref. no. 2015/507) approved the study. The residents, who had given oral consent for their participation, were requested to complete the QPC–H questionnaire anonymously and place it in an envelope in a locked box at the housing facility.

Measures

Data were collected using the QPC–H questionnaire for measuring the quality of the housing support services from the residents’ perspective, where the respondents were requested to assess the quality of housing support for each item in a Likert-type scale ranging from 1 (totally disagree) to 4 (totally agree). The instrument is based on the original QPC for Inpatient Care instrument (QPC-IP, 30 items) for measuring quality in psychiatric care by Schröder et al. (2007, 2010), which now exists in several versions for use in different psychiatric contexts. The QPC-H instrument, which consists of 37 items, has a clear factor structure and demonstrates good reliability (Lundqvist, Rask, et al., 2016). The QPC–H contains the following dimensions: Encounter (8 items), Participation (8 items), Support (4 items), Secluded environment (3 items), Secure environment (3 items), and Housing specific (11 items).

Data analysis

Questionnaires with ≥30% missing items were excluded, and imputation was carried out by replacing the missing data points with the overall mean of that item in questionnaires with <30% missing items. Univariate regression analyses were performed on the QPC dimension scores with all background variables shown in Table 1. In Table 2, the responses (totally disagree, partly disagree, agree to a high degree and totally agree are presented in ascending order based on response 1, totally disagree. The corresponding QPC dimension scores, from a previous study of quality in the OHS housing context (Rask et al., 2017), are compared with the scores from the present study and are shown in Table 3. P-values less than 0.05 are regarded as statistically significant and all statistics were calculated with SPSS 24. Based on previous research on users’ perceptions of the quality of health services, we used 80% positive ratings (i.e. a score of at least 3 on QPC-H items) as a cutoff value defining satisfactory quality of service (Lundqvist, Ivarsson, Rask, Brunt, & Schröder, 2018).

Table 1. Characteristics of the study group (N = 178)

Background Variable	Residents n (%)	Background Variable	Residents n (%)
Gender		Living status
Men	110 (62)	Alone	170 (95)
Women	67 (38)	With partner	3 (2)
Missing data	1 (1)	Missing data	5 (3)
Age		Education
18-39	45 (25)	Not competed school	8 (5)
40-59	86 (48)	Compulsory school	59 (33)
60-75	38 (21)	Upper secondary school	75 (42)
Missing data	9 (5)	Higher education	32 (18)
		Missing data	4 (2)
Nationality		Daily occupation
Swedish	159 (89)	Vocational training	34 (19)
Other	17 (10)	Student	10 (6)
Missing data	2 (1)	Other	115 (65)
		Missing data	19 (11)
Living in current housing		Know responsible head of unit
0 – 3 months	67 (38)	No	30 (17)
3 – 12 months	67 (38)	Yes	145 (81)
12 – 24 months	21 (12)	Missing data	3 (2)
Missing	23 (13)	Know where to complain
		No	68 (38)
		Yes	108 (61)
		Missing data	2 (1)
Waiting time for current housing		Perceived time of waiting
No waiting time	51 (29)	Far too short	6 (3)
< 1 month	33 (19)	Too short	32 (18)
1-3 months	35 (20)	Neither short nor too long	86 (48)
> 3 months	43 (24)	Too long	24 (14)
Missing data	16 (9)	Far too long	12 (7)
		Missing data	18 (10)
Recommend current housing service		Patient Reported Diagnoses *
Certainly not	17 (10)	Anxiety disorders	5 (3)
Rather not	20 (11)	Affective disorders	17 (10)
Maybe	49 (28)	Neuropsychiatric	17 (10)
Willingly	49 (28)	Personality disorder	1 (1)
Absolutely	39 (22)	Psychosis/Schizophrenia	71 (40)
Missing data	4 (2)	Missing data	67 (38)
Know one’s diagnosis		Number of reported diagnoses
Yes	125 (70)	One	103 (58)
No	33 (19)	Two	8 (5)
Not got a diagnosis	12 (7)	Three or more	2 (1)
Missing data	8 (4)	Missing data	65 (37)
Current contact with outpatient clinic		Frequency of contact with outpatient clinic
No	44 (25)	≥ 2 times/week	5 (3)
Yes	129 (72)	1 time/week	11 (6)
Missing data	5 (3)	1-2 times/month	33 (19)
		2-6 times/year	37 (21)
		Missing data	92 (52)
Current mental health		Current physical health
Very bad	6 (3)	Very bad	12 (7)
Bad	21 (12)	Bad	28 (16)
Neither good nor bad	56 (32)	Neither good nor bad	59 (33)
Good	63 (35)	Good	57 (32)
Very good	29 (16)	Very good	20 (11)
Missing data	3 (2)	Missing data	2 (1)

* Residents can have more than one diagnosis.

Table 2. Perceptions of residents of the quality of psychiatric care in Supported Housing, percentage of answers per response category.

	Supported housing
	Totally disagree	Partly disagree	Agree to a high degree^a	Totally agree^a
Encounter
The staff treat me with kindness	1.7	14.0	30.3	53.9
The staff respect me	2.2	16.3	32.6	48.9
The staff have time to listen to me	2.2	20.8	33.1	43.8
The staff are dedicated	2.2	13.5	32.0	52.2
I get support when I need it	3.4	17.4	31.4	47.8
The staff understand my feelings	3.4	20.8	30.3	45.5
The staff care about how I perceive the housing service	4.5	16.3	28.6	50.6
The staff care about why I am angry	8.4	18.0	28.6	44.9
Participation
My previous experience of running a home is optimally utilized	2.2	15.7	35.9	46.1
I have influence over the service provided	5.6	25.8	39.4	29.2
My opinion on what is correct service is respected	6.2	12.9	38.2	42.7
I learn about my psychiatric disabilities	6.7	20.8	33.7	38.8
I received information in an understandable way	7.9	18.0	31.4	42.7
I am involved in decisions about my housing	8.4	11.2	36.5	43.8
I have been informed about the various support measures so that I can decide what is best for me	10.1	19.7	32.6	37.6
I get to learn to recognize signs of deteriorating mental health	11.8	25.8	34.2	28.1
Support
The staff prevent me from exposing my surroundings for discomfort	3.4	8.4	49.5	38.8
The staff help me to understand that guilt and shame should never have to stop me from seeking care	5.1	13.5	37.7	43.8
The staff help me to understand that it is not shameful to have suffered from psychological disorders	5.6	19.1	26.4	48.9
The staff prevent me from exposing myself to injury	6.2	15.2	42.7	36.0
Secluded Environment
I have privacy	2.2	13.5	27.5	56.7
I have a secluded place	3.4	9.0	25.2	62.4
There is a secluded place where I can receive visits from my family and friends	9.0	10.7	19.6	60.7
Secure Environment
I can feel safe with the other residents	3.9	18.0	33.2	44.9
I feel safe in my own home	5.1	16.3	26.4	52.2
I’m not disturbed by the other residents	9.6	20.2	29.8	40.4
Housing Specific Dimension
The staff have the ability to listen to me	2.2	19.7	31.4	46.6
The staff help me to create order in my daily life	3.4	16.9	33.7	46.1
The staff talk with me about what I perceive as difficult	3.9	24.2	31.5	40.4
The staff can help me to create order when I feel bad	4.5	19.1	32.6	43.8
I monitor and evaluate the support I get together with the staff	5.1	21.9	25.9	47.2
The staff encourage me to develop new skills	5.6	20.8	34.9	38.8
I have control over what the staff and I do together	6.2	12.4	38.7	42.7
The staff try to get to know me	6.2	18.5	29.8	45.5
The staff speak with me and not just each other	6.2	24.2	29.8	39.9
The staff talk to me about what I’m feeling and thinking	7.9	30.3	25.3	36.5
The staff encourage me to keep in touch with family and friends	11.2	19.1	32.6	37.1

^a Items meeting an acceptable quality level of 80% agree/totally agree are underscored.

Table 3. Perceptions of residents of dimensions of the quality of psychiatric care. Comparison between Supported Housing and Ordinary Housing with Outreach Support.

Dimensions of QPC	Supported Housing N = 178	Ordinary Housing with Outreach support N = 174	P-level*
	M(SD)	M(SD)
Encounter	3.25 (0.68)	3.48 (0.58)	<0.001
Participation	3.05 (0.66)	3.26 (0.61)	0.002
Support	3.23 (0.68)	3.33 (0.59)	NS
Secluded Environment	3.41 (0.66)	3.62 (0.44)	<0.001
Secure Environment	3.16 (0.73)	3.28 (0.49)	NS
Housing Specific	3.10 (0.73)	3.33 (0.64)	0.003
QPC Total	3.20 (0.58)	3.38 (0.45)	0.001

* T-test.

Results

Demography

The mean age of the participants was 48 years (SD = 13.9) and there was a majority of male residents (62%). A great majority was born in Sweden and over 90% had at least completed compulsory or upper secondary school. None of the participants was gainfully employed, while approximately 25% of them were in vocational training or were students and a majority had some form of daily occupation at day care centers (Table 1).

Most of the residents were living alone and had lived in their current housing facility for a mean period of 6 months (SD = 5.4). Approximately two-thirds had received a place in their housing facility within 3 months and about 50% had perceived the waiting time as being neither too short nor too long.

A majority of the residents reported that they knew their diagnosis and nearly 60% only reported having one diagnosis. The most frequently reported diagnoses were psychosis/schizophrenia, neuropsychiatric disorder and affective disorders. The levels of self-reported current mental and physical health were generally positive, although approximately one-third reported neither good nor bad ratings. Most of the residents had a current contact with an outpatient psychiatric clinic; the appointment frequency varied from twice a week to twice a year. Additionally, a majority of the residents knew who was in charge of the unit and where to complain.

Quality in supported housing

Secluded Environment, Encounter and Support were the dimensions that were rated having the highest quality of housing service (Table 3), while Participation, Housing Specific and Secure Environment were rated at lower levels. The three individual items with the highest number of residents who totally agreed with the statements are all the items in the Secluded Environment dimension: ‘I have a secluded place’, ‘I have privacy’ and ‘There is a secluded place where I can receive visits from my family and friends’. Two of the three individual items with the highest number of residents, who totally disagreed with the statements, are from the Participation dimension: ‘I get to learn to recognize signs of deteriorating mental health’ and ‘I have been informed about the various support measures so that I can decide what is best for me’.

Only 12 of the 37 items in the QPC-questionnaire attained the 80% cutoff value of a satisfactory level of quality. Three items in each of the Encounter, Participation and Secluded Environment dimensions were rated at this satisfactory level of quality. In the dimension Secluded Environment, all the three items, named above, having ‘a secluded place’ and ‘privacy’ had a satisfactory level of quality, although a considerable minority, slightly less than 20%, questioned the availability of a secluded place to receive visitors.

Three items in the Encounter dimension were assessed as having a satisfactory level of quality: ‘the staff treat me with kindness’, ‘the staff are dedicated’ and ‘the staff respect me’, while more than one quarter did not agree that ‘the staff care about why I am angry’. In the Support dimension, the residents rated the items: ‘the staff prevent me from exposing my surroundings for discomfort’ and ‘the staff help me to understand that guilt and shame should not stop me from seeking care’ at satisfactory quality levels. On the other hand, almost one quarter of the sample disagreed mostly with ‘The staff helps me to understand that it is not shameful to have suffered from psychological disorders’.

Only one item in the Housing Specific dimension attained a satisfactory level of quality: having ‘control over what the staff and I do together’. However, more than 30% of the residents disagreed with: ‘the staff talk to me about what I’m feeling and thinking’, ‘the staff encourage me to keep in touch with family and friends’ and ‘the staff speak with me and not just each other’. In Secure Environment the residents agreed mostly with: ‘I feel safe in my home’, while about 30% disagreed with the statement ‘I’m not disturbed by the other patients’, but none of the three items were rated at a satisfactory level.

In Participation, the dimension with the lowest mean score (Table 3), the residents rated highly: ‘previous experience of running a home being optimally utilized’, ‘opinions about what they perceive as the correct care being respected’ and them being ‘involved in decisions about my housing’. On the other hand, ‘to learn to recognize signs of deteriorating mental health’ and ‘I have been informed about the various support measures so that I can decide what is best for me’ were the statements about which more than 30% of the residents disagreed.

The results of the univariate regression analysis showed that gender differences were found with women perceiving the quality of Encounter, Secluded Environment and the total QPC score lower than men (Table 4). The participants born in foreign countries rated the dimension Secure Environment higher than their Swedish counterparts had done. There were, however, no significant differences in terms of the quality of housing services with regard to the residents’ age, nationality, living status, educational level, or whether having a daily occupation or not. Furthermore, there were no significant differences in the perceived quality of housing for how long one had lived in the current accommodation, how long the resident had waited for current housing and the residents’ perception of the waiting time.

Table 4. Regression coefficients (β) for background questions (independents) association with QPC total and domain scales (dependents) among people with psychiatric disabilities with supported housing.

Background variables	QPC total	QPC Domain Scales
		1. Encounter	2. Participation	3. Support	4. Secluded Env.	5. Secure Env.	6. Housing Spec.
Age	−0.11	−0.11	−0.10	−0.10	−0.12	−0.03	−0.08
Gender	−0.16*	−0.17*	−0.14	−0.11	−0.15*	−0.13	−0.12
Nationality	0.02	0.02	0.00	−0.01	−0.05	0.16*	−0.03
Living status	0.07	0.08	0.08	0.05	0.08	−0.03	0.08
Education	−0.08	−0.07	−0.06	0.14	−0.02	−0.02	−0.11
Daily occupation	−0.02	−0.02	−0.06	−0.00	0.06	−0.02	−0.03
Waiting time for current housing	−0.06	−0.09	−0.09	−0.05	−0.01	0.00	−0.08
Perceived waiting time	0.02	−0.02	−0.01	−0.01	−0.02	0.10	−0.00
Living in current housing	0.05	0.01	0.05	0.02	0.13	0.00	0.06
Know one’s diagnosis	0.00	−0.01	0.04	−0.02	0.07	0.06	−0.13
Type of diagnosis
Psychosis/Schizophrenia	Ref	Ref	Ref	Ref	Ref	Ref	Ref
Anxiety disorders	0.09	0.18	0.01	0.19	0.03	−0.07	0.13
Affective disorders	0.17	0.16	0.20*	0.11	0.08	0.10	0.21*
Neuropsychiatric	0.11	0.17	−0.00	0.07	0.17	0.10	0.04
Personality disorder	0.01	0.01	0.03	0.00	−0.05	−0.09	0.14
Number of Diagnoses	0.08	0.11	−0.01	0.14	0.01	0.01	0.07
Current contact with outpatient clinic	0.01	0.04	−0.02	0.09	0.01	−0.08	0.03
Current mental health	0.32***	0.30***	0.31***	0.18*	0.19*	0.35***	0.26***
Current physical health	0.23**	0.20**	0.22**	0.07	0.20**	0.30***	0.17*
Know responsible head of unit	0.20**	0.12	0.14	0.07	0.33***	0.19*	0.17*
Know where to complain	0.27***	0.21**	0.31***	0.21**	0.22**	0.12	0.29***
Recommend current housing service	0.30***	0.28***	0.37***	0.25***	0.21**	0.18*	0.24**

N = 178

*p < 0.05,

**p < 0.01, and

***p < 0.001.

A number of significant differences were, however, found in terms of the clinical characteristics. The residents who rated having a better mental health also rated higher housing quality for all dimensions. A similar result was found for physical health for all dimensions except Support. The association between these health variables and the quality of housing services was strongest for the dimension Secure Environment. There was no difference for the frequency of contact with a psychiatric outpatient clinic. There was no relationship with perceived quality of housing in regard to the number of diagnoses (comorbidity) or whether the residents knew their diagnosis/es. When looking at specific diagnoses, only affective disorder was related with quality of care in the Participation and Housing Specific dimensions (β = 0.20, p = 0.040 and β = 0.21, p = 0.035, respectively).

Furthermore, knowing who is in charge of the unit was positively related to the residents’ perception of quality of care for the Secluded and Secure Environment dimensions and the Housing Specific dimension, while knowing where to complain was positively related to the perception of quality in all dimensions except for Secure Environment. Finally, the residents who perceived a high level of quality for all dimensions were more positive to recommending the housing service.

In a comparison of the perception of the quality of housing services between SH facilities and those in OHS, the residents in the latter rated higher levels for the total QPC-score and four of the dimensions: Encounter, Participation, Secluded Environment, and Housing Specific (Table 3). However, there were similarities between the two housing services in the ranking of the quality dimensions, where Secluded Environment and Encounter were rated the highest and Participation the lowest in both housing contexts. A satisfactory quality of service for the SH facilities according to the residents was attained for 12 of the 37 items (32%), while the corresponding figure for OHS was 20 of the 37 items (54%).

Discussion

The highest quality ratings in supported housing facilities were found for the dimensions: Secluded Environment, Encounter and Support, while Participation, Housing Specific and Secure Environment were rated at lower levels. The three individual items with the highest level of total agreement according to the residents were all from the Secluded Environment dimension, while two of those with the highest level of total disagreement were from the Participation dimension. Despite relatively high ratings, a majority of items did not attain the 80% level of positive ratings that was deemed as defining satisfactory quality service. Gender differences were found with women perceiving the quality of Encounter, Secluded Environment and the total QPC score lower than men and residents born in foreign countries rated the dimension Secure Environment higher than their Swedish counterparts. Residents with a better health status generally rated a higher housing quality. Knowing who is in charge of the unit and where to complain were positively associated with a number of dimensions. The residents in ordinary housing with outreach support rated higher levels for the majority of the QPC–H dimensions in comparison with those in supported housing.

The two environmental dimensions in the QPC-questionnaire focus on different aspects; one addresses the concept of seclusion and the other that of security. All the items in the former attained a satisfactory level of quality according to the residents, while none of the items in the latter achieved this level. This contrast repeats the findings in the corresponding study using the QPC–H questionnaire focusing on OHS (Rask et al., 2017). In a study of satisfaction with housing for this group, privacy/seclusion and security/safety were found to be the greatest predictors of satisfaction (Brolin et al., 2015). Satisfaction with housing can be construed to be a constituent of the quality of care in that context (Schröder et al., 2006) and these two environmental aspects of housing are thus of great significance for quality of care. The different levels of quality rated by the residents for these two dimensions in the present study may perhaps be seen as contradictory as seclusion and security are not widely varying concepts. They may, however, represent two sides of the same coin reflecting the complex nature of a SH facility that has to meet the residents’ varying individual needs in a congregate environment. The results from previous studies support this interpretation, where living in SH facilities has been experienced as safe and secure (Harvey et al., 2012; Roos, Bjerkeset, Søndenaa, Antonsen, & Steinsbekk, 2016), while on the other hand limiting privacy and integrity through enforced companionship (Harvey et al., 2012) and generating insecurity when conflicts between residents occur (Roos et al., 2016). To improve quality of care in this context staff need to be aware of this dilemma and work towards improving the security aspects of the SH environment whilst maintaining a high level of seclusion.

The Participation dimension, containing items concerning information and resident involvement in the care at the housing facility, is rated the lowest of the six dimensions by the residents in SH facilities. This is a common feature of studies using the QPC instrument across varying psychiatric contexts: inpatient care (Schröder et al., 2010), outpatient care (Schröder, Ahlström, Wilde-Larsson, & Lundqvist, 2011), day-center services (Lundqvist et al., 2018) and housing-OHS (Rask et al., 2017). This common denominator of a low level of consumer participation in psychiatric services may well reflect the general hierarchical structure that has previously been prevalent in these services with low levels of patient autonomy and of involvement in decisions about care. Brolin et al. (2016a) have reported that a lack of self-determination was the main concern of residents in SH. Furthermore, Dadich, Fisher and Muir (2013) have found that residents, in services promoting “consumer-driven care plans”, had little or no involvement in drawing these up or were able to challenge the goals that the staff had suggested. Kirkpatrick and Byrne (2011) have proposed ways to promote greater participation and involvement by an emphasis on the need for creating possibilities for participation at different levels in SH, building confidence and a gradual involvement in planning and organizing activities. Based on a recent review of housing studies, Krotofil, McPherson, and Killaspy (2018) maintained that all housing services for this target group share a similar primary goal of facilitating recovery and independence, where participation and involvement in care processes are essential features. An increased focus on recovery in the staff training at SH units could help to improve the levels of participation and thus quality of care. From a forensic psychiatric care context, Selvin, Almqvist, Kjellin, and Schröder (2016) maintain that patient participation can be improved by active communication and the encouragement of patients taking responsibility, which is also applicable in a supported housing context.

The items in the QPC-H with the greatest level of disagreement are also interesting and important for staff to focus on. Two items in the Participation dimension; concerning support measures and learning about signs of deteriorating mental health and three in the Housing Specific dimension; concerning feelings, keeping in touch with family and friends and staff talking to me and not each other, had the largest number of totally disagree responses (30–38%). These items are similar in content to those that had the highest level of disagreement in the corresponding study in the OHS context (Rask et al., 2017), i.e. information and relational aspects. Skills development in the performance of these aspects for the support staff is needed to improve the quality of housing services, confirming the suggestions made by Brolin et al. (2015) about the importance of the social aspects of the staff’s work and the provision of information to the residents.

A second aim of this study was to compare the findings in the present SH context with those from the OHS context (Rask et al., 2017). The residents in the latter rated a higher quality of care than their counterparts in the SH context for four of the six dimensions as well as the QPC total score. A basic distinction in terms of an independent or congregate living situation may well explain this difference, as indicated by Warren and Bell’s (2000) study of housing preferences, where residents with previous experience of hospitalized care appreciated more the qualities of independent living in comparison to that of congregate living. Moreover, residents in SH reported less satisfaction with the performance of support than their counterparts in OHS (Brolin et al., 2015).

Women perceived the quality of Encounter, Secluded Environment and the total QPC score lower than men, while in the corresponding study in the OHS context women perceived the quality of Encounter, Participation and Housing Specific higher than men had done. One reason for these differences could be that women may feel they are more vulnerable in SH facilities, where there is generally a male predominance and thus perhaps have a greater need for privacy. The staff in SH facilities are generally predominantly female, and it is possible that staff expect more from female residents than they do from male residents thus leading to the female residents rating the quality of Encounter lower than men do. This reasoning is of course somewhat speculative and further research is needed in this context to gain greater knowledge and understanding about these gender differences in order to improve services.

Only 12 of the 37 items (32%) in the SH context were judged to attain the 80% positive rating cutoff value for a satisfactory quality of service. The corresponding score for the OHS context was 20 of 37 items (54%), and in the day-center context 25 of the 35 (71%) items were assessed as attaining that level of quality (Lundqvist et al., 2018). This may reflect the inherent difference between congregate and other services as discussed above but also confirming the findings from other studies where SH facilities have been criticized. The critique has concerned: restrictiveness of the environment (Henwood, Derejko, Couture, & Padgett, 2015), substandard accommodation (Forchuk, Nelson, & Hall, 2006) and disempowerment (Brolin et al., 2016a).

Some interesting results are revealed in a visual comparison of the ranking of the mean ratings for the five generic dimensions with three other studies using the QPC-questionnaires: inpatient care (QPC-IP; Schröder et al., 2010), community-based day center services (QPC-DA; Lundqvist, Ivarsson, et al., 2016) and ordinary housing with outreach support (QPC-H; Rask et al., 2017). Secluded Environment is rated highest in the two housing contexts but lower in the day center and inpatient care settings. The Encounter dimension is rated highly in all four settings, while the Participation dimension is lowest in three of the four settings. Some of these differences may well be explained as a consequence of the specific nature of the settings, for example in terms of the environmental aspects. This can, however, not explain the generally high ratings for the staff-resident interaction (encounter) and the lowest ratings by the residents/patients for the opportunities for participating in their own care in three of the contexts. A further analysis of the results from the studies using the QPC-questionnaires can shed more light on these apparent differences in quality of care in various psychiatric settings.

Limitations and strengths

The instrument used in this study, the QPC-H, is based on the original QPC-IP questionnaire and adjusted to suit the housing context. There is a family of QPC-questionnaires designed to measure the quality of psychiatric services from the patient/resident perspective, all of which have demonstrated a distinct factor structure and good reliability, which can be seen as a strength. The quality of care as rated in the study could be said to be fairly high although still lower than in other psychiatric contexts. This level of quality may be linked to the original process for identifying items for the questionnaire, which was based on the perceptions of patients in psychiatric care about positive aspects of quality of care; the items are thus positively phrased. Quality of care is generally rated highly by the consumers and a high cutoff point of 80% was chosen for defining satisfactory service quality based on previous research on users’ perceptions. This provided good possibilities for differentiation as seen in the comparison of results of the number of items that attained this cutoff point in different contexts.

Implications for practice

The results of this study of the quality of care in supported housing facilities for people with psychiatric disabilities have revealed a number of deficiencies, indicated by the majority of items not attaining the cutoff point for satisfactory service quality, and which thus need to be addressed in order to improve the quality of care. The clear differences between the two housing types, the SH and OHS, together with the complex nature of congregate SH facilities indicate the need for administrators and staff to pay greater attention to the specific needs of the residents in this housing type and to improve in-house training to achieve this aim. Some aspects that are indicated in this study and which should feature in such training include a focus on: recovery, environmental aspects, the promotion of autonomy, self-determination and participation as well as relational aspects and information.

Conclusions

The quality of care in supported housing facilities has, according to the residents, a number of deficiencies that need to be addressed. A comparison between the two major types of housing in Sweden for people with psychiatric disabilities has revealed that a majority of the aspects of quality of care as studied with the QPC-H questionnaire are rated lower in supported housing facilities than in ordinary housing with outreach support. Staff training with a focus on, for example, recovery, environmental aspects and participation is needed to improve the quality of care in these housing facilities.

Disclosure statement

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

Additional information

Funding

We thank NSPH (National Partnership for Mental Health) for their administrative support.