Design
This is a cross-sectional study, analyzing baseline data from the Norwegian research project A Pairwise Randomized Study on Implementation of Guidelines and Evidence-based Treatments of Psychoses (ClinicalTrials NCT03271242). The study was approved by the Regional Committee for Medical and Health Research Ethics (REK Sørøst B 2015/2169), and followed the principles of the Declaration of Helsinki.
Setting and sample
A total of 325 mental health service users from six health authorities across Norway, including three university hospitals, were recruited. Thirty-nine clinical units and hospital departments with outpatient clinics, day units, mobile teams, and inpatient wards participated. Further details about the participating units are available in the study protocol (ClinicalTrials NCT03271242). Inclusion criteria were: mental health service user diagnosed with psychosis (ICD-10 F20-29) (World Health Organization, 1992), and aged 16 years or older. The only exclusion criterion was being unable to understand and answer the questionnaires in Norwegian. Thirty-three service users with missing data were excluded, reducing the final study sample to N = 292, for whom sociodemographic and clinical characteristics are shown in Table 1.
Table 1
Participants (N = 292) sociodemographic and clinical characteristics
Characteristics
|
|
Gender N (%)
|
|
Female
|
122 (42)
|
Ethnicity N (%)
|
|
Norwegian
|
255 (88)
|
Other
|
34 (12)
|
Age Mean (SD)
|
40 (12.7)
|
Diagnosis N (%)
|
|
Schizophrenia
|
145 (53)
|
Schizoaffective disorder
|
54 (20)
|
Other
|
74 (27)
|
GAF symptoma Mean (SD)
|
53 (13)
|
GAF functionb Mean (SD)
|
51 (11.3)
|
Community treatment order N (%)
|
|
Yes
|
40 (14)
|
Depression (BASIS-24)c Mean (SD)
|
1.3 (0.92)
|
Personal recovery (QPR)d Mean (SD)
|
41 (10.2)
|
Perceived support (INSPIRE)e Mean (SD)
|
66 (17.6)
|
Quality of life (MANSA)f Mean (SD
|
41 (10.2)
|
Satisfaction with services (CSQ-8)g Mean (SD)
|
26 (4.7)
|
- Range from 0 to 100, higher scores indicate less severity
- Range from 0 to 100, higher scores indicate higher function
- Range from 0 to 4, higher scores indicate more severe symptoms
- Range from 0 to 60, higher scores indicate higher level of personal recovery
- Range from 0 to 100, higher scores indicate more perceived support
- Range from 1 to 7, higher scores indicate higher quality of life
- Range from 8 to 32, higher scores indicate higher satisfaction
Measures
Outcome measure
The Client Satisfaction Questionnaire-8 (CSQ-8) (Table 2) (25) 8 is an eight-item questionnaire used to measure patient’s global satisfaction with services, which has shown good psychometric properties (26). The CSQ-8 measures general satisfaction on eight scaled items from 1 (= poor) to, 4 (= excellent) resulting in a total score range of 8–32. Level of satisfaction is classified as low (8–20), intermediate (21–26), or high (27–32).
Covariates
The Questionnaire about the Process of Recovery (QPR) (27) was used to examine personal recovery level. The QPR is a 15-item self-report measure of recovery developed through collaboration between clinicians and service user researchers, which has shown adequate psychometric properties (28). Items are rated on a five-point Likert scale from 0 (Disagree strongly) to 4 (Agree strongly). Total sum score ranges from 0 (low recovery) to 60 (high recovery).
Perceived support for personal recovery was examined using the 20-item support subscale from the INSPIRE measure of staff support of personal recovery (29). Each service user-rated subscale item is first rated on whether it is important for the participant’s recovery (e.g. “An important part of my recovery is…feeling hopeful about my future, (Yes/No). If yes, the participant rates the support they receive from their health service provider for this item (“I feel supported by my worker with this”) on a five-point Likert scale from 0 (Not at all) to 4 (Very much). A total support score is calculated for each participant as described in the INSPIRE scoring instruction guide (30) and ranges from 0 (low support) to 100 (high support).
Quality of life was assessed using a single item from the Manchester Short Assessment of Quality of Life (MANSA) (31): “How satisfied are you with your life as a whole?” which was rated on a seven-point scale from 1 (Couldn’t be worse) to 7 (Couldn’t be better). The variable was named Quality of life.
Confounders
User and service characteristics that have been considered as potential confounders in studies on satisfaction with services, such as illness severity, depressive symptoms, age, and legal status of treatment (1), were included.
Illness severity was assessed using the Global Assessment of Functioning Scale (GAF) (32). Level of functioning and severity of service users’ symptoms are rated by clinicians on a scale (1–100), with lower scores indicating more severe symptoms and lower levels of functioning. The split version of the scale used in this study has symptom (GAF-S) and function (GAF-F) subscales (33).
Depression was assessed using the ‘depression/functioning’ domain of the Behavior and Symptom Identification Scale (BASIS-24). BASIS-24 is a brief service user self-report measure of psychopathology and functioning, which was developed to assess mental health treatment outcomes. This 24-item scale assesses six symptom and functioning domains: Depression/functioning, Interpersonal relationships, Self-harm, Emotional lability, Psychosis, and Substance abuse. BASIS-24 has shown good validity and reliability for assessing mental health status and functioning from the perspective of service users (34, 35). Scores were calculated as described in the BASIS-24 instruction guide (36), providing a score between 0 and 4 with higher scores indicating more severe problems.
Information on whether participants were on a CTO (Yes/No) at the time of participation in the study, gender, and age were also included as confounders.
Table 2
Items of the Client Satisfaction Questionnaire (CSQ-8) (Range 8–32)
1. How would you rate the quality of service received?
2. Did you get the kind of service that you wanted?
3. To what extent has our program met your needs?
4. If a friend were in need of similar help, would you recommend our program to him or her?
5. How satisfied are you with the amount of help you have received?
6. Have the services you received helped you to deal more effectively with your problems?
7. In an overall, general sense, how satisfied are you with the service you have received?
8. If you were to seek help again, would you come back to our program?
|
Procedure
Clinicians at the participating mental health units recruited eligible service users who were in contact with the clinic during the study period, and newly referred service users assessed to have psychosis. Clinicians performed clinical ratings and questionnaires were administered to service users by the secretary or other clinic personnel. Service users were either provided with a place to sit in the clinic to complete the questionnaires, or took them home. When finished, the questionnaire was sealed in an envelope, and returned to the clinic. The recruitment period lasted from June 2016 until March 2017, and only participants who gave written informed consent were included.
Analysis
Factor analysis with principal component extraction method was performed on CSQ-8 items. Cronbach’s alpha was estimated to assess the reliability of the scale.
As participants came from different clinical sites, a hierarchical structure may have been present in these data. Intra-class correlation coefficient was used to assess possible cluster effects on site level. Unadjusted and adjusted linear mixed models with random effects for units were estimated to assess the association between service satisfaction (CSQ-8) and three covariates (QPR, INSPIRE, MANSA) controlled for confounders (GAF-symptoms, GAF-Function, Depression/functioning, CTO, Age, and Gender). Multicollinearity was assessed by inspecting correlations among covariates. Standard residual diagnostics was performed.
As an exploratory analysis, the interactions between being on a CTO and quality of life (MANSA) and personal recovery (QPR) were entered into the model, to determine whether the CTO-variable moderated their relation with CSQ-8. All tests were two-tailed, and results with p-values below 0.05 were considered statistically significant.
Imputation of missing values on the GAF (n = 31) and the MANSA (n = 7) were performed by generating the empirical distributions for each variable and drawing a random number from that distribution to replace the missing value. The process was repeated until all missing values were imputed.