Participation in leisure activities and quality of life of people with psychosis in England: a multi-site cross-sectional study

Study design and participants

A cross-sectional survey was conducted in NHS Community Mental Health Team (CMHT)s in England. CMHT is an umbrella term used to describe a multi-professional team involved in the delivery of mental health care and it’s formed of community psychiatric nurses, occupational therapists, social workers, psychologists, psychiatrists and health care support workers.

From June 2017 to May 2018, participants were recruited in six participating NHS Trusts covering a range of geographical areas, in both urban and rural contexts: Cornwall Partnership NHS Foundation Trust; Devon Partnership NHS Trust; East London NHS Foundation Trust (covering East London, Luton and Bedfordshire); Oxford Health NHS Foundation Trust (covering large areas of Oxfordshire and Buckinghamshire) and Somerset Partnership NHS Foundation Trust; Tees, Esk and Wear Valleys NHS Foundation Trust (covering County Durham, Darlington, Teeside, York and North Yorkshire).

Participants were identified by clinicians or clinical study officers from CMHT caseloads. Whilst CMHTs look after patients with different diagnoses, this study focussed on out-patients with a diagnosis of a psychotic disorder according to the International Classification of Disease-10 (ICD-10) codes F20-29.

Participants were eligible for inclusion if they met the following criteria:

•Adults aged 18–69 years old.

•A clinical diagnosis of a psychotic disorder according to the International Classification of Disease-10 (ICD-10) codes F20-29, as identified in clinical records.

•Receiving care from outpatient secondary mental health services or primary care services.

•Have the capacity to provide informed consent.

•Able to communicate in English.

Exclusion criteria:

•A current and primary diagnosis of substance use disorder (ICD-10 F10-19).

•Had been hospitalised in the previous week (although these potential participants could be re-approached at a later time).

•Their postcodes could not be obtained because they were homeless or living in temporary accommodation at the time of the survey.

Procedures and measures

Eligible participants were identified by clinicians or clinical study officers and asked for their agreement to speak to a member of the research team. Participants then completed the study questionnaires and researchers accessed participant clinical records to retrieve clinical and sociodemographic characteristics. All participants who agreed to take part in this study were interviewed in quiet rooms in community mental health teams, primary care settings, or at participant’s homes using standardised case report forms. All interviews were face-to-face and took about 45 min to complete. Several measures were used during the assessments:

First, the UK Time Use Survey (TUS) [20] as adapted by Priebe et al. [21] to focus on activities outside of home (See Additional file 1: Table S1 Time Use Survey Leisure Activities), was used to assess participation in leisure activities lasting more than 10 min during the previous week. The adapted TUS was selected for its focus on social leisure activities which take place outside the home and include the following: Been to a museum/art gallery, Been to a place of entertainment (e.g. dance, club, bingo and casino), Been to an event as a spectator (e.g. sports event, theatre and live performance), Been to outdoor trips (e.g. picnics, beach), Been to a library, Been to a community social group/day centre, Been to a shopping centre, Been to the cinema, Attended a religious group/activity/service, Been visited by friends, Visited friends and Been out to eat/drink at a café/restaurant/pub. If they participated in an activity that was not on the list, they were also asked to specify the activity they completed. For each activity they had completed, participants were asked to report (i) the number of times they completed the activity (i.e. only taking short breaks in between constituted one activity), (ii) the duration to the nearest 10 min, (iii) whether participation took place alone or with someone else, (iv) and if with someone else, to define their relationship to this individual: parent, sibling, friend, partner or other. Participants were asked not to double-count time spent in activities (e.g. going out for a meal and visiting a friend) but select the activity which best describes the event.

Second, the number of self-reported social contacts in the last 7 days was measured with the Social Contact Assessment (SCA) [22]. The instrument asks participants to list the initials of social contacts with who they have been in contact in the last 7 days to generate a total number of social contacts. For ‘being in contact’, we mean that they can name them and have had a chat (more than just greetings) in the last week. Participants were asked not to include people they were living with, health professionals or people they worked with, unless their contact that took place outside of and was unrelated to work. For each contact, participants were asked to define the type of relationship (1 = parent, 2 = sibling, 3 = friend, 4 = partner, 5 = other), on how many days in the last week they had been in face-to-face contact, whether the meeting was one-to-one, in a group or both, on how many days they used voice or video call, mail or text message, whether they can talk to them about their personal feelings/worries and whether they did something for them and vice-versa.

Third, participants reported satisfaction with different aspects of their life using the twelve items of the Manchester Short Assessment of Quality of Life (MANSA), rated on a scale from 1 (very dissatisfied) to 7 (very satisfied). Participants were asked to rate their satisfaction with: life as a whole, job situation, financial situation, friendships, leisure activities, accommodation, personal safety, the people they live with, sex life, family relationships physical health and mental health. MANSA has been widely used to assess the quality of life of people with psychosis and its psychometric properties have been well established [23,24,25].

Researchers collected additional participant characteristics, such as age, gender (male/female), marital status (single/in a relationship), country of birth (born in the United Kingdom/born in a different country), an education level (primary/secondary/further), living situation (living alone/living with someone), accommodation (living independently/living in supported accommodation), employment (employed/not employed), receipt of welfare benefits (or not) and length of illness (calculated in the number of years from the day of the first contact with mental health services). These were collected from participants’ assessments and checked against available data in medical records.

Description of the sample

Figure 1 presents the flow of participant recruitment (see Fig. 1). Once inclusion criteria were applied, 2888 were eligible. Of those, we were able to contact and explain the study to 1720 people of whom 613 agreed to take part representing a consent rate of 35%. For a small number of consented individuals, it only became apparent that they did not meet certain eligibility criteria (e.g. capacity to provide informed consent, been hospitalised in the previous week) upon meeting.

Fig. 1figure 1

Flow diagram describing recruitment of community participants with psychosis

A sample of 588 participants enrolled in the study, of whom 533 reported their participation in leisure activities during the previous week. The majority of participants (N = 407) responded to all 12 questions related to quality of life with the MANSA. All analyses involving the outcome variable (i.e. quality of life) used the final sample (N = 407). The missing data was handled and the list-wise deletion was used because data was missing completely at random [26].

Ethics committee approval

The West Midlands—Solihull Research Ethics Committee (17/WM/0191) approved the study. All participants were given the information sheet and written informed consent was obtained for all participants.

Statistical analysis

Descriptive statistics (i.e. number of participants and percentage) for socio-demographic factors are reported. Mean and Standard Deviation are reported for quality of life, leisure activities and social contacts (see Table 1). Quality of life is the mean score (on scale of 1 to 7) of the 12 items on the MANSA. Leisure activities are the number of activities participants attended in the last 7 days. Social contacts are the total number of face-to-face social contacts with different individuals in the last 7 days.

Table 1 Descriptive statistics: socio-demographic informationStructural equation model (SEM)

Three basic concepts are commonly used by researchers for SEM reporting: (1) a well-developed theoretical model, (2) operational definitions of the observed/unobserved variables and (3) graphical representation of the theorised model. A checklist for SEM reporting is used [26, 27].

SEM has been widely used as a series of statistical methods because it allows analyses, evaluation and interpretation of complex relationships between one or more independent variables and one or more dependent variables [28, 29].

SEM has observed variables, also called measured variables that are represented by a square or rectangle in the traditional graphic. The unobserved variables, also termed latent variables are represented by large circles, and the small circles represent the measurement errors in the observed variables. Two other terms associated with SEM are exogenous, similar to independent variables, and endogenous, similar to dependent/outcome variables. Single-head arrows (paths) represent directional effects from 1 variable (latent or observed) to another. In our structural model, we used the path analysis where leisure activities, social contacts and quality of life were measured (see Figs. 2 and 3). In the diagram, the regression coefficient, the direction and the magnitude of the relationship, is set at 1 for each path [26, 27].

Fig. 2figure 2

Relationship between the Participation in leisure activities and the Quality of life. The number of social contacts is mediated the association. Standardised coefficients and values

Fig. 3figure 3

Relationship between the Participation in leisure activities and the Quality of life, after controlling for socio-demographic and other factors such as diagnosis. Standardised coefficients and values

We explored the relationships between participation in leisure activities and the quality of life through the proposed SEM. This study hypothesises that participation in leisure activities has a positive influence on quality of life. In addition, it is hypothesised that social contacts are thought to have mediating effects between participation in leisure activities and quality of life. This reflects the general logic model of the study ‘social engagement—> social contacts—> quality of life’ [30]. In other words, in our model, participation in leisure activities is an independent variable, social contacts are a mediator and quality of life is a dependent variable. The fit of the model is evaluated using SEM. The relationships are graphically presented in Figs. 2 and 3.

We conducted the statistical analyses in two steps. The first step looked at the relationship between participation in leisure activities, social contacts and quality of life without involving socio-demographic factors (see Fig. 2 and Table 2). The second step was built from step one by adding into the model all socio-demographic factors and other factors such diagnosis (see Fig. 3 and Table 3).

Table 2 Correlations, means and standard deviations (SD) for individual items (quality of life score, leisure activities attended and social contacts made)Table 3 SEM investigating the association between the participation in leisure activities and quality of life; the mediating effect between social contacts and quality of life

Dummy variables were created for socio-economic factors to investigate the association of some variables of interest with the quality of life. For example, whether there is gender difference in the way male and female participate in leisure activities [10]; whether being unemployed [17], single [10], living alone or having a specific diagnosis [14, 19] influence the quality of life of people with psychosis. Dummy variables created (e.g. 1 = female and 0 = otherwise; 1 = single and 0 = otherwise; 1 = young adults and 0 = otherwise; 1 = up to secondary education and 0 = otherwise; 1 = independent accommodation and 0 = otherwise; 1 = living alone and 0 = otherwise; 1 = unemployed and 0 = otherwise; 1 = on benefits and otherwise; 1 = schizophrenia and 0 = other diagnoses).

We reported odds ratios (OR) with their corresponding 95% confidence intervals (CI) and P-value significant of 0.05. The final sample size was reported and we explained how missing data was handled. For example, the list-wise deletion was used because data was missing completely at random [26].

We also checked whether the quality of life and other factors, such as age and education, were normally distributed (See Additional file 2: Distribution Graphs). All statistical analyses were conducted using Stata 17.0 [31].

留言 (0)

沒有登入
gif