Counselling for young people and young adults in the voluntary and community sector: An overview of the demographic profile of clients and outcomes.

OBJECTIVES: To determine the demographic profile of clients accessing voluntary and community sector (VCS) counselling services for young people and young adults, and to undertake the first systematic evaluation of outcomes.
DESIGN: Naturalistic, non-experimental study design.
METHODS: A total of 2,144 clients were recruited from nine services in England. Data on gender, age, ethnicity, number of sessions, problem descriptions, and initial levels of psychological distress were collected from clients. Calculations of reliable and clinically significant change, repeated measures t-tests, effect size estimates, and moderation analyses were undertaken to determine whether pre-post change, over an average of 5.1 (SD = 5.3) sessions, was significant and which variables predicted outcomes. Levels of service satisfaction were also recorded.
RESULTS: The demographic profile of clients accessing VCS services was distinct to those accessing statutory and school-based counselling services, with a greater representation of females, 'older' clients, and clients from Black and Minority Ethnic (BME) groups. Clients accessing VCS counselling services presented with multiple and complex needs. Statistically significant reductions in psychological distress were observed with effect sizes ranging from .64 to .80. High levels of service satisfaction were expressed by clients.
CONCLUSIONS: This study establishes that clinical outcomes in the VCS are comparable to those reported in school-based and statutory mental health services in the United Kingdom. VCS services were perceived as being highly accessible and appear able to reach marginalized groups who may not be accessing other services. PRACTITIONER POINTS: VCS services in the United Kingdom may be more accessible to young people from marginalized groups, such as those from BME backgrounds, compared to statutory and school-based counselling services. Counselling in VCS services is associated with significant short-term reductions in psychological distress, at a level comparable with the statutory and educational sectors. Clients accessing counselling in VCS services report relatively high levels of service satisfaction.

Background

One in 10 young people in the United Kingdom suffer from a diagnosable mental health disorder (Green, McGinnity, Meltzer, Ford, & Goodman, 2004). Furthermore, one in eight reported symptoms of mental ill‐health between 2011 and 2012 (ONS, 2015), with conduct disorders, anxiety, depression, and attention deficit hyperactivity disorder (ADHD) being the most prevalent issues in the United Kingdom (Green et al., 2004). Estimates suggest that half of all cases of mental health disorders start by the age of 14 years and three‐quarters by the age of 24 (Kessler et al., 2005).

It is estimated that the global burden of mental illness accounts for 32.4% of years lived with a disability (YLDs) and 13% of disability‐adjusted life years (Vigo, Thornicroft, & Atun, 2016). Furthermore, global estimates suggested that mental health problems cost £1.6 trillion in 2010, with this expected to more than double by 2030 (Bloom et al., 2011). In the United Kingdom, the financial implications of mental health service use for 5‐ to 15‐year‐olds with an emotional or behavioural disorder have been estimated to cost public sector services almost £1.47 billion annually (Snell et al., 2013). Failure to address such issues in childhood can lead to increased risks of self‐harm, suicide, substance misuse and contact with the criminal justice system, and lower educational, employment, and financial achievements (Richards et al., 2009). These findings highlight the importance of early intervention and prevention.

Specialist Child and Adolescent Mental Health Services (CAMHS) are NHS‐funded statutory services in the United Kingdom which assess and treat young people with behavioural, emotional, and mental health difficulties (Young Minds, 2017). Such services typically employ multi‐disciplinary teams and offer a range of treatments to young people, including medication and psychological therapies, such as cognitive behavioural therapy, counselling, art therapy, and child and adolescent psychotherapy. Between 1 June and 31 August 2017, there were 36,069 new referrals to CAMHS (NHS Digital, 2017), suggesting that services receive around 200,000–250,000 new referrals annually. Consistently, reports have noted difficulties young people face in accessing CAMHS, such as an increase in the number of referrals, waiting times, and thresholds for entry (CYP MHT, 2015). The most recent CAMHS benchmarking report (NHS Benchmarking Network, 2015) highlighted an 11% increase in the number of referrals to CAMHS in England between 2013/2014 and 2014/2015, as well as an increase in the average waiting time for a routine appointment from 22 to 32 weeks.

School‐based counselling is an alternative non‐statutory mental health provision. In the United Kingdom, it predominantly takes a person‐centred or humanistic form (Cooper, 2009). A recent report published by the Department for Education (DfE, 2017) has estimated that 61% of schools and colleges in England provide counselling services for pupils. Unlike in CAMHS, which follows a more diagnostic model, many young people accessing school‐based counselling do not present with specific clinical disorders, such as anxiety or depression (Cooper, 2009). School‐based counselling has shown positive outcomes in pilot randomized trials (Cooper, 2013). However, it has been acknowledged that there are areas for improvement. These include the following: increasing evidence‐based practice, improving access for those from Black and Minority Ethnic (BME) backgrounds, and ensuring that services are equipped to meet the needs of young people from vulnerable groups, such as looked after children (DfE, 2016). In addition, research indicates that not all young people want to see a counsellor in their school setting, with approximately one quarter expressing a preference to see a counsellor elsewhere, such as a youth centre or doctor's surgery (Cooper, 2006). Reasons for why young people may not wish to see a counsellor within school include the perception that it is less discrete than other settings, and anxiety about peers knowing they are attending (Cooper, 2004).

Voluntary and community sector (VCS) counselling services for children and young people – typically termed Youth Information, Advice and Counselling Services (YIACS) – provide a non‐statutory provision outside of the school environment, which may address some of the challenges above Estimates suggest that such services typically see between 80,000 and 100,000 clients each year (Street, 2013). However, they have reported increasing demand which has been attributed to failures in statutory services and an increase in the number of young people presenting with complex and multiple issues (Youth Access, 2016). Furthermore, it has been suggested that VCS services may play an important role in engaging hard‐to‐reach and marginalized groups of young people (Milbourne, 2009).

A review by Street (2013) suggests that many VCS counselling services in the United Kingdom do evaluate the outcomes of the support they provide, with preliminary evidence that their interventions are associated with positive outcomes. Despite this, it has also been noted that outcomes data from this setting are sparse, with great variability in the validity and the type of data collected (Street, 2013).

The aim of the present study, therefore, was to address this gap in the literature by conducting the first systematic evaluation of the outcomes of counselling in the VCS in England. As a systematic study, we aimed to collect outcome data using a common set of measures in accordance with a pre‐defined protocol, and across a large number of participants at several VCS sites. In addition, using this sample, we aimed to establish the demographic profile of service users accessing VCS counselling services for young people and young adults.

Method

Design

This was a naturalistic study which collected data from young people and young adults who accessed YIACS across England between June 2014 and June 2016.

Setting

Youth Information, Advice and Counselling Services typically work with young people aged 11–25. The types of support available can vary across services, but typically include information, counselling, emotional support, access to health clinics, and advice and advocacy (Youth Access, 2017). Whilst participants recruited to the present study had access to the full range of interventions offered by the services they were seen in, data were only collected from those who received counselling.

Intervention

Counsellors working in YIACS typically practise a range of therapeutic orientations. However, estimates suggest that 91% provide person‐centred counselling and 71% provide humanistic counselling, with around half providing cognitive or psychodynamic therapies (Youth Access, 2008). All counsellors who provided the intervention for the present study worked on an individual, face‐to‐face basis and had completed a diploma in counselling or psychotherapy as a minimum. Counsellors typically worked with clients on a weekly basis.

Measures

Demographics were collected using a proforma designed specifically for the present study. The form captured gender, age (in years), and ethnicity.

Current View tool

The Current View tool (CAMHS Press, 2013) is a practitioner‐reported measure which is used to capture problem descriptions, complexity factors (e.g., being a looked after child or holding asylum seeker/refugee status), contextual factors, and education, employment, or training (EET) difficulties in young people presenting for therapy. Problem descriptors, contextual factors, and EET difficulties are rated in terms of severity of impact (‘none’, ‘mild’, ‘moderate’ or ‘severe’), whilst complexity factors are considered as being present or not. For the purposes of the present study, prevalence was defined as the sum of mild, moderate, and severe indications of problems.

Emerging research lends some support to the convergent validity of the Current View tool (Wolpert et al., 2015); however, preliminary indications of reliability are moderate at best (Wolpert et al., 2015). Further research to determine the reliability and validity of the Current View Tool has been recommended (Wolpert et al., 2015).

YP‐CORE/CORE‐10

The primary outcome measure was the Young Person's Clinical Outcomes in Routine Evaluation (YP‐CORE; Twigg et al., 2009) or the CORE‐10 (Barkham et al., 2013) depending on the age of the client; with those aged between 11 and 16 years old, at initial assessment, asked to complete the YP‐CORE and those aged 17 and over to complete the CORE‐10. Of the 1,299 clients who completed CORE‐10, 93 (7.2%) were younger than 17 years at intake. Conversely, 124 (16.5%) of those who completed YP‐CORE were older than 17 years at intake. As these proportions were small, and CORE‐10 and YP‐CORE are very similar measures, it was not deemed necessary to exclude these data from analyses.

Both the YP‐CORE and CORE‐10 are self‐report measures of psychological distress, with good levels of internal reliability, acceptability, and validity (Barkham et al., 2013; Twigg et al., 2009, 2016). Clients are asked to respond to 10 statements in terms of how often they have felt like this during the past week (‘not at all’, ‘only occasionally’, ‘sometimes’, ‘often’, or ‘all or most of the time’). Items include statements such as ‘I've felt able to cope when things go wrong’ and ‘I've felt unhappy’ and responses carry a numerical weighting of 0–4. Total scores on both measures range from 0 to 40, with higher scores indicating greater levels of psychological distress.

The clinical cut‐off for CORE‐10 is 11; scores equal to or greater than this indicate that a person is experiencing clinical levels of psychological distress (Barkham et al., 2013). A difference of six or more points between first and last session is considered ‘reliable change’ that is not change that has occurred by chance or through measurement error. Therefore, a client can ‘reliably improve’ by reducing their total CORE‐10 score by 6 or more points, or ‘reliably deteriorate’ by increasing their total score by the same number of points. The reliable change indices and clinical thresholds for YP‐CORE are gender and age band specific and are set out in Table 1 (Twigg et al., 2016).

Table 1

Reliable and clinical change indices – YP‐CORE (Twigg et al., 2016)

Gender/age	Reliable change indices	Clinical cut‐off point
Male, 11–13 years	8.3	10.3
Male, 14–16 years	8.0	14.1
Female, 11–13 years	8.0	14.4
Female, 14–16 years	7.4	15.9

In instances where it had not explicitly been indicated whether the CORE score obtained from a client was from CORE‐10 or YP‐CORE, the research team assumed that those aged 16 years and younger had completed the YP‐CORE and those aged 17 years and older had completed the CORE‐10. In total, paired outcome data from 179 clients were treated in this manner. For 81 clients with paired outcome data, no data were available on age or which CORE measure had been used, and therefore, they were excluded from further analysis.

Experience of Service Questionnaire

The Experience of Service Questionnaire (ESQ; Attride‐Stirling, 2002) is a 15‐item measure of service satisfaction, which can be completed by a young person or their parent/carer. For the purposes of the present study, the ESQ was used in self‐report form only. Twelve of the 15 items consist of statements such as ‘I feel that the people who saw me listened to me’, ‘The facilities here are comfortable (e.g., waiting area)’, and ‘Overall, the help I have received here is good’. Respondents are asked to identify whether they feel each statement is ‘certainly true’, ‘partly true’, or ‘not true’. The remaining three questions are open‐ended to elicit qualitative data on what respondents feel has been: ‘really good about their care’, what they feel ‘needs improving’, and if there is anything else they want to say about the service they received. Only the quantitative data are reported here. The mean response rate for individual questions on the ESQ was 98.1% (SD = 1.89, range = 92.4–99.7%).

Participants

Nine YIACS were recruited to the study through Youth Access, a registered charity and national membership organization, and eight saw the study through to completion. The services were in a variety of locations across England, employed between 8 and 60 full time equivalent staff, and had an annual turnover of between £250,000 and £2,000,000. Seven of the eight services (87.5%) were located in inner‐city areas.

A total of 2,155 young people and young adults were referred for counselling during the data collection period. Eleven participants were excluded from further analyses due to being outside of the 11–25 age range, leaving a total sample of 2,144. Paired outcome data were available from 1,448 participants (68.0% of the total sample; CORE‐10 n = 899, YP‐CORE n = 549). Of these, 1,241 (85.7%) cases were recorded as being ‘closed’ (CORE‐10 n = 789, YP‐CORE n = 452): counselling had either been mutually concluded or a client had not attended for 6 weeks or more. Analyses of pre–post outcomes were undertaken on closed cases only.

Experience of Service Questionnaire forms were available for 1,374 (64.1%) clients. Figure 1 outlines the flow of participants through the study.

Figure 1

Flow chart of study participants.

The mean number of sessions attended across the whole sample (n = 2,144) was 5.1 (SD = 5.3, range 1–47 sessions). Participants who completed the CORE‐10 measure (n = 1,299) attended an average of 5.6 sessions (SD = 5.9, range 1–47 sessions) and those who completed the YP‐CORE measure (n = 750) attended an average of 4.8 sessions (SD = 4.3, range 1–39).

Sample demographics

Table 2 provides an overview of the sample demographics. The majority of participants who completed the CORE‐10 measure were female (n = 848, 65.3%) and had an average age of 20.0 years (SD = 2.8); less than half (n = 579, 44.6%) were of white British ethnicity, and just under a fifth (n = 237; 18.2%) were of Black or Black British ethnicity. A slightly higher proportion of participants who completed YP‐CORE were female (n = 524, 69.9%), from a white British background (n = 466, 62.1%) and had an average age of 15.0 years (SD = 2.1).

Table 2

Participant demographics

Variable		Total sample (n = 2,144)	CORE‐10 (n = 1,299)	YP‐CORE (n = 750)
Gender (n, %)	Male	638 (29.8)	433 (33.3)	202 (26.9)
	Female	1,384 (64.6)	848 (65.3)	524 (69.9)
	Transgender	1 (0.0)	1 (0.1)	0 (0.0)
	Missing	121 (5.6)	17 (1.3)	24 (3.2)
Mean age in years (SD)		18.2 (3.5)	20.0 (2.8)	15.0 (2.1)
Age (n, %)	11–15 years	508 (23.7)	42 (3.2)	444 (59.2)
	16–20 years	957 (44.6)	675 (52.0)	284 (37.9)
	21–25 years	570 (26.6)	564 (43.4)	14 (1.9)
	Missing	109 (5.1)	18 (1.4)	8 (1.1)
Ethnicity (n, %)	White British	1,047 (48.8)	579 (44.6)	466 (62.1)
	White Other	109 (5.1)	73 (5.6)	43 (4.5)
	Asian/Asian British	126 (5.9)	96 (7.4)	28 (3.7)
	Mixed ethnicity	176 (8.2)	119 (9.2)	55 (7.3)
	Black/Black British	311 (14.5)	237 (18.2)	70 (9.3)
	Other ethnic background	55 (2.6)	38 (2.9)	17 (2.3)
	Missing	320 (14.9)	157 (12.1)	80 (10.7)

Percentages have been calculated as a proportion of the total number of participants in each group (i.e., 433 [33.3%] of the 1,299 participants who had completed the CORE‐10 were male).

Initial severity levels of psychological distress

For participants who completed the CORE‐10, there were no statistically significant differences in mean initial levels of psychological distress between clients who attended just one session (n = 400, M = 21.8, SD = 7.5) and those who attended at least two sessions, n = 899, M = 21.4, SD = 7.4; t(1,297) = 1.078, p = .281. However, for clients who completed the YP‐CORE, there was a statistically significant difference in mean initial levels of psychological distress between clients who attended just one session (n = 201, M = 18.9, SD = 8.5) and those who attended at least two sessions, n = 549, M = 20.5, SD = 7.9; t(748) = −2.481, p = .013. This indicates that younger clients (i.e., those aged 16 years and below) who only attended for one session were less psychologically distressed at intake than those who attended for at least two sessions.

Procedure

Initial contact with YIACS was made through Youth Access, inviting them to attend a meeting to discuss the present study. Expressions of interest were then formally followed up with service managers who were provided with an information sheet and consent form.

Service managers who consented to participate were provided with training in the consent procedures for young people and completing and scoring the outcome measures. Service managers were then required to cascade this training down to the counsellors employed in their services.

Informed consent was obtained from young people by the counsellor. Young people aged under 16 were assessed for eligibility to provide consent using the principles of ‘Gillick competence’ (Fallon, 2003). Here, a young person can be deemed able to give consent providing they have the intellectual and emotional competence to do so, as well as an understanding of the implications of participating.

Counsellors were asked to complete the demographics form and Current View tool during, or just after, the first session with a young person. The Current View tool could then be updated over the course of counselling to provide an accurate ‘picture’ of the young person at that point in time. The CORE‐10/YP‐CORE measure was administered at every session to ensure that an ‘end‐point’ measure was collected for each participant who attended two or more sessions.

Formal ethical approval was not sought for the present study due to it being an analysis of anonymised, routinely collected data for the purposes of service evaluation; hence, according to UK standards, ethical review is not required (Department of Health, 2011).

Data analysis

Descriptive statistics were calculated for demographic data, number of sessions (based on the number of completed CORE‐10/YP‐CORE measures per client), initial levels of psychological distress, problem descriptions, complexity factors (e.g., being a looked after child or holding asylum seeker/refugee status), contextual problems, and EET difficulties, and levels of reliable and clinically significant change (Evans, Margison, & Barkham, 1998). To determine the amount of reliable and clinical change on YP‐CORE, it is necessary to know both the age and gender of the client. Therefore, 349 (77.2%) of the 452 clients with paired YP‐CORE data were included in the analysis; for the remaining 103 clients, either age or gender was unknown, or their age was outside of the 11–16 year range for whom YP‐CORE is appropriate.

Descriptive statistics were also calculated for ESQ data to determine the response options for each question as a proportion of the total number of responses. In addition, research has shown that the summed score of ESQ items 1–7, 11 and 12 measure satisfaction with care to a good degree of precision (Brown, Ford, Deighton, & Wolpert, 2014). Therefore, satisfaction with care scores were calculated for each client, with higher scores indicating higher levels of satisfaction.

Inferential statistics, including repeated measures t‐tests, effect size calculations, and moderation analyses, were used to determine whether pre–post change on the CORE measures was significant and whether gender, age, or ethnicity predicted outcomes.

Data were analysed using the Statistical Package for the Social Sciences (SPSS) version 18.0 (SPSS Inc., 2009).

Results

Problem descriptions, complexity factors, contextual problems, and attendance/attainment issues in EET

In total, Current View forms were available for 1,445 (67.4%) clients. Tables 3 and 4 provide an overview of the responses to each item. Percentages are calculated as a proportion of those for whom Current View was available.

Table 3

Current View – problem descriptions, contextual problems, and attendance/attainment issues in education, employment, or training

	Problem description
	None (%)	Mild (%)	Moderate (%)	Severe (%)	Not known/Missing (%)
Anxious away from caregivers (separation anxiety)	488 (33.8)	504 (34.9)	76 (5.3)	29 (2.0)	348 (24.1)
Anxious in social situations (social anxiety/phobia)	437 (30.2)	293 (20.3)	266 (18.4)	149 (10.3)	300 (20.8)
Anxious generally (generalized anxiety)	274 (19.0)	346 (23.9)	449 (31.1)	185 (12.8)	191 (13.2)
Compelled to do or think things (OCD)	532 (36.8)	182 (12.6)	208 (14.4)	97 (6.7)	426 (29.5)
Panics (panic disorder)	579 (40.1)	207 (14.3)	175 (12.1)	74 (5.1)	410 (28.4)
Avoids going out (agoraphobia)	737 (51.0)	199 (13.8)	110 (7.6)	40 (2.8)	359 (24.8)
Avoids specific things (specific phobia)	758 (52.5)	107 (7.4)	87 (6.0)	27 (1.9)	466 (32.2)
Repetitive problematic behaviours (habit problems)	707 (48.9)	88 (6.1)	85 (5.9)	27 (1.9)	538 (37.2)
Depression/low mood	378 (26.2)	281 (19.4)	392 (27.1)	202 (14.0)	192 (13.3)
Self‐injury or self‐harm	494 (34.2)	247 (17.1)	283 (19.6)	145 (10.0)	276 (19.1)
Extremes of mood (bipolar disorder)	737 (51.0)	128 (8.9)	106 (7.3)	27 (1.9)	447 (30.9)
Delusional beliefs and hallucinations (psychosis)	898 (62.1)	50 (3.5)	38 (2.6)	18 (1.2)	441 (30.5)
Drug and alcohol difficulties (substance abuse)	854 (59.1)	74 (5.1)	48 (3.3)	14 (1.0)	455 (31.5)
Difficulties sitting still or concentrating (ADHD/hyperactivity)	847 (58.6)	106 (7.3)	60 (4.2)	32 (2.2)	400 (27.7)
Behavioural difficulties (CD or ODD)	846 (58.5)	82 (5.7)	72 (5.0)	22 (1.5)	423 (29.3)
Poses risk to others	1,013 (70.1)	64 (4.4)	37 (2.6)	17 (1.2)	314 (21.7)
Carer management of Children and Young People behaviour (e.g., management of child)	987 (68.3)	55 (3.8)	37 (2.6)	5 (0.3)	361 (25.0)
Doesn't get to the toilet in time (elimination problems)	1,113 (77.0)	11 (0.8)	14 (1.0)	6 (0.4)	301 (20.8)
Disturbed by traumatic event (PTSD)	695 (48.1)	92 (6.4)	155 (10.7)	63 (4.4)	440 (30.4)
Eating issues (anorexia/bulimia)	616 (42.6)	127 (8.8)	115 (8.0)	73 (5.1)	514 (35.6)
Family relationship difficulties	451 (31.2)	248 (17.2)	321 (22.2)	176 (12.2)	249 (17.2)
Problems in attachment to parent/carer (attachment problems)	431 (29.8)	185 (12.8)	254 (17.6)	173 (12.0)	402 (27.8)
Peer relationship difficulties	409 (28.3)	306 (21.2)	314 (21.7)	98 (6.8)	318 (22.0)
Persistent difficulties managing relationships with others (includes emerging personality disorder)	651 (45.1)	192 (13.3)	156 (10.8)	55 (3.8)	391 (27.1)
Does not speak (selective mutism)	1,195 (82.7)	60 (4.2)	62 (4.3)	17 (1.2)	111 (7.7)
Gender discomfort issues (gender identity disorder)	1,064 (73.6)	14 (1.0)	11 (0.8)	3 (0.2)	353 (24.4)
Unexplained physical symptoms	978 (67.7)	48 (3.3)	31 (2.1)	18 (1.2)	370 (25.6)
Unexplained developmental difficulties	1,019 (70.5)	23 (1.6)	15 (1.0)	6 (0.4)	382 (26.4)
Self‐care issues (includes medical care management, obesity)	1,095 (75.8)	59 (4.1)	36 (2.5)	6 (0.4)	249 (17.2)
Adjustment to health issues	980 (67.8)	63 (4.4)	55 (3.8)	21 (1.5)	326 (22.6)

Table 4

Current View – complexity factors

Complexity factor	Not present (%)	Present (%)	Not known/Missing (%)
Looked after child	1,171 (81.0)	80 (5.5)	194 (13.4)
Young carer status	1,217 (84.2)	58 (4.0)	170 (11.8)
Learning disability	1,137 (78.7)	67 (4.6)	241 (16.7)
Serious physical health issues (including chronic fatigue)	1,097 (75.9)	81 (5.6)	267 (18.5)
Pervasive developmental disorders (autism/Asperger's)	1,119 (77.4)	55 (3.8)	271 (18.8)
Neurological issues (e.g., Tics or Tourette's)	1,184 (81.9)	26 (1.8)	235 (16.3)
Current protection plan	1,177 (81.5)	14 (1.0)	254 (17.6)
Deemed ‘child in need’ of social service input	1,148 (79.4)	35 (2.4)	262 (18.1)
Refugee or asylum seeker	1,249 (86.4)	31 (2.1)	165 (11.4)
Experience of war, torture, or trafficking	1,260 (87.2)	10 (0.7)	175 (12.1)
Experience of abuse or neglect	840 (58.1)	162 (11.2)	443 (30.7)
Parental health issues	599 (41.5)	325 (22.5)	521 (36.1)
Contact with Youth Justice System	847 (58.6)	139 (9.6)	459 (31.8)
Living in financial difficulty	754 (52.2)	124 (8.6)	567 (39.2)

The most prevalent presenting problem was generalized anxiety (n = 980, 67.8%), followed by depression/low mood (n = 875, 60.5%), family relationship difficulties (n = 745, 51.6%), and peer relationship difficulties (n = 718, 49.7%). The most common complexity and contextual factors were parental health issues (n = 325, 22.5%), experience of abuse/neglect (n = 162, 11.2%), contact with the Youth Justice System (n = 139, 9.6%), living in financial difficulty (n = 124, 8.6%), school, work, or training issues (n = 765, 52.9%), and home issues (n = 658, 45.5%). Furthermore, 330 (22.8%) young people were experiencing attendance difficulties in education, employment, and training and 336 (23.3%) were experiencing attainment difficulties in these environments.

Clinical outcomes

CORE‐10

The mean pre‐therapy CORE‐10 score was 21.5 (SD = 7.4), and the mean post‐therapy CORE‐10 score was 14.9 (SD = 8.6), which indicates a statistically significant reduction in psychological distress, t(788) = 22.4, p < .001, mean difference = 6.5 (SD = 8.2, 95% CI: 5.96, 7.11). Cohen's effect size (d, Cohen, 1988), which was computed using the pooled SD and corrected for dependence between means (Morris & DeShon, 2002), was 0.80, which represents a large effect size.

Just over half of the sample (n = 412, 52.2%) showed reliable improvement, a small proportion showed reliable deterioration (n = 44, 5.6%), and the remainder (n = 333, 42.2%) showed no reliable change. A total of 232 young adults (29.4%) showed clinically significant change (i.e., recovery).

Pre‐therapy CORE‐10 scores significantly predicted post‐therapy CORE‐10 scores (β = .48, 95% CI: 0.49, 0.63, p < .001), as did the number of sessions (β = −.15, 95% CI: −0.32, −0.13, p < .001) and client age (β = −.08, 95% CI: −0.45, −0.06, p = .011). These three predictors explained 26.2% of the variance, R 2 = .262, F(3, 772) = 92.57, p < .001. In summary, lower pre‐therapy CORE‐10 scores, more sessions, and older age were all predictors of lower post‐therapy CORE‐10 scores.

YP‐CORE

The mean pre‐therapy YP‐CORE score was 20.9 (SD = 7.6), and the mean post‐therapy YP‐CORE score was 15.8 (SD = 8.8), indicating a statistically significant reduction in psychological distress, t(451) = 13.6, p < .001, mean difference = 5.2 (SD = 8.1, 95% CI: 4.4, 5.90). Cohen's effect size (Cohen, 1988) was 0.64, which represents a medium effect.

Table 5 provides an overview of the amount of reliable and clinically significant change shown by young people on YP‐CORE. A total of 130 (37.2%) young people showed reliable improvement, a small proportion (n = 17, 4.9%) showed reliable deterioration, and 127 (36.4%) showed clinically significant change.

Table 5

Reliable and clinical change – YP‐CORE

	Male (11–13 years, n = 25, %)	Male (14–16 years, n = 71, %)	Female (11–13 years, n = 61, %)	Female (14–16 years, n = 192, %)
Reliable improvement	10 (40.0)	30 (42.3)	22 (36.1)	68 (35.4)
Clinically significant changea	10 (40.0)	32 (45.1)	21 (34.4)	64 (33.3)
Reliable improvement and clinically significant change	7 (28.0)	21 (29.6)	18 (29.5)	49 (25.5)
No change	14 (56.0)	38 (53.5)	36 (59.0)	114 (59.4)
Reliable deterioration	1 (4.0)	3 (4.2)	3 (4.9)	10 (5.2)

Defined as moving from ‘caseness’ at first session to below caseness at last session (i.e., recovery).

Overall, males were more likely to show clinically significant change on the YP‐CORE than females (43.8% and 33.6%, respectively). Similarly, males were more likely to show reliable improvement than females (41.7% and 35.6%, respectively).

Pre‐therapy YP‐CORE scores significantly predicted post‐therapy YP‐CORE scores (β = .50, 95% CI: 0.48, 0.67, p < .001), as did client age (β = .10, 95% CI: 0.09, 0.82, p = .015) and client gender (β = 1.00, 95% CI: 0.28, 3.44, p = .021). These predictors explained 28.6% of the variance, R 2 = .286, F(3, 431) = 58.53, p < .001. In summary, lower pre‐therapy YP‐CORE scores, younger age, and being male were all predictors of lower post‐therapy YP‐CORE scores.

CHI‐ESQ

Table 6 summarizes the number of clients who responded ‘not true’, ‘partly true’, and ‘certainly true’ to each item on the ESQ expressed as a percentage of the total number of clients who responded to that item.

Table 6

Responses to Experience of Service Questionnaire

	Not true (%)	Partly true (%)	Certainly true (%)
I feel that the people who saw me listened to me	0.7	2.9	96.5
It was easy to talk to the people who saw me	1.3	14.3	84.4
I was treated well by the people who saw me	0.5	1.7	97.8
My views and worries were taken seriously	0.8	3.1	96.1
I feel the people here know how to help me	0.9	15.4	83.7
I have been given enough explanation about the help available here	0.9	12.1	87.0
I feel that the people who have seen me are working together to help me	1.1	9.1	89.8
The facilities here are comfortable (e.g., waiting area)	2.8	17.9	79.3
My appointments are usually at a convenient time (e.g., don't interfere with school, clubs, college, work)	3.0	14.9	82.0
It is quite easy to get to the place where I have my appointments	2.2	16.8	81.0
If a friend needed this sort of help, I would suggest to them to come here	1.1	7.4	91.5
Overall, the help I have received here is good	0.9	2.9	96.2

For clients who responded to all nine of the ‘satisfaction with care’ items of the ESQ (n = 1,207, 87.8%), mean scores were calculated. The mean ‘satisfaction with care’ score, where 18 would indicate the highest possible level of satisfaction, was 17.26 (SD = 1.76, range = 0–18).

Discussion

The present study established the largest dataset yet for counselling in the VCS, with over 2,000 clients seen across nine services in England between 2014 and 2016. Paired outcome data were available from just over two‐thirds of clients, with the majority of these being identified as ‘closed cases’. This collection of routine outcome data is consistent with developments in CAMHS, where the Children and Young People's Improving Access to Psychological Therapies (CYP IAPT) programme, introduced in 2011, has encouraged practitioners to routinely collect session‐by‐session data (Wolpert et al., 2016). However, data completeness seen in the CAMHS data completeness is low, estimated to be only around 21% (Wolpert et al., 2016).

There was a difference in the demographic profile of young people seen in VCS services compared to those seen in CAMHS and school‐based counselling services. In the former, we found a higher representation of ‘older’ clients, females, and those from BME backgrounds. The mean age of clients in the present study was 20 years, compared to a mean age of 12 years in CAMHS (Wolpert et al., 2016) and 13.86 years in school‐based counselling (Cooper, 2009). As young people in the United Kingdom typically transition from CAMHS to adult mental health services at 16 or 18 (NHS, 2016), which is also the approximate age that compulsory education or training ends (HM Government, 2008), these findings are broadly in line with expectations. However, the comparatively higher average age of those accessing VCS services suggests that these services may be filling a gap in statutory and educational service provision.

The proportion of clients from BME background in the present study was 31.2%, compared to 18% in CAMHS (Wolpert et al., 2016), and 3.0% in school‐based counselling (Cooper, 2009). This suggests that VCS services may play an important role in meeting the needs of minority client groups, who are often under‐represented in talking therapies (Mind, 2013). Importantly, this suggests that such services may be more accessible and be perceived as less stigmatizing than CAMHS and school‐based counselling services by minority groups. However, it is also important to note that most of the services who provided data for the current study were in inner‐city areas where populations tend to be more ethnically diverse.

The most common presenting issues in young people seen as part of the present study are broadly in line with those identified in CAMHS and school‐based counselling, namely: family relationship difficulties, depression/low mood, generalized anxiety, and peer relationship difficulties. Importantly, however, the proportion of young people presenting with generalized anxiety disorder and depression/low mood was markedly higher in the present study (67.8% and 60.5%, respectively) compared to CAMHS (49% and 50%, respectively) and school‐based counselling (9.3% and 9.6%, respectively; Wolpert et al., 2016; Cooper, 2009). This is in line with findings from school‐based counselling services where it is noted that clinical presentations are less common (Cooper, 2009). Reasons for the differences between VCS services and CAMHS are unclear, although it may be related to the differences in the demographic profile of clients seen in these settings. Despite this, it is important to note the high proportion and complexity of issues being presented with at VCS services, which has been documented previously (Youth Access, 2016).

With respect to complexity factors, around a fifth of clients accessing both VCS and CAMH services experience parental health issues, and approximately one in 8–9 have experienced abuse or neglect (Wolpert et al., 2016). A much higher proportion of young people in the present study had been in contact with the Youth Justice System compared to those seen in CAMHS (9.6% and 2%, respectively). Again, this may be due to differences in the demographic profile of clients seen in these settings, with vulnerable and disengaged groups being at a higher risk of involvement with the Youth Justice System (McAra & McVie, 2010). Conversely, young people seen in CAMHS were more likely to experience issues at home (59%), and attendance (31%) and attainment difficulties (42%) in education, employment, and training than those seen in VCS services (45.5%, 22.8%, and 23.3%, respectively).

The proportion of young people in the present study showing levels of reliable improvement and clinically significant change (or ‘recovery’) is comparable to that in CAMHS (Wolpert et al., 2016); namely, just over half show reliable improvement and approximately one‐third ‘recover’. In contrast, ‘recovery’ rates of 45% have been reported in school‐based counselling services (Cooper, 2013), indicating that young people seen in school‐based settings may have improved outcomes compared to other types of services for young people. However, it is also important to note the distinct differences in the presenting issues of young people seen in the present study and CAMHS, and those who present to school‐based counselling, which may account for some of these inconsistencies. One explanation for the recovery rate in the present sample is the high average level of psychological distress at intake, which was around 20 on both CORE‐10 and YP‐CORE; meaning that young people often would need to make improvements of approximately 10 points or more to achieve ‘recovery’ – far beyond the mean improvement of 5 or 6 points in the present study. Levels of ‘reliable deterioration’ in young people seen in VCS services are roughly half (around 5%) of that reported in CAMHS and school‐based counselling services (around 10%; Wolpert et al., 2016; Cooper, 2009).

In our VCS sample, we found that clients’ ages significantly predicted outcomes, with those at the more extreme ends of the age spectrum showing greater improvements than those in the middle of the age range. A possible explanation for this is that those in the mid‐range of ages (16–20 years) are more likely to be experiencing significant life transitions, such as leaving school and moving from child to adult services, which can be difficult to manage (Paul, Street, Wheeler, & Singh, 2014).

Importantly, high levels of service satisfaction were reported by young people in the present study, greater than that reported in CAMHS (Wolpert et al., 2016) and consistent with school‐based counselling services (Cooper, 2013). For example, 96.5% of clients in the present study felt it was ‘certainly true’ that they ‘were listened to’, compared to 84.6% of those seen in CAMHS (Wolpert et al., 2016). Similarly, 82% and 81% of clients in the present study felt it was ‘certainly true’ that their appointments were at convenient times and a convenient location, respectively, compared to 62% and 66.4% of those seen in CAMHS (Wolpert et al., 2016). The latter finding suggests that VCS services may be experienced as more accessible to young people and young adults than CAMHS.

Limitations

The limitations of the present study are primarily attributable to the naturalistic – or non‐experimental – design of the study. First, the lack of a control group limits the extent to which it can be concluded that VCS services provide effective interventions for reducing psychological distress. However, it should be noted that the same is true for data routinely collected in statutory services. In addition, the naturalistic study design enables an accurate reflection of what happens in everyday practice, rather than the controlled – and sometimes unnatural – environment of an experimental study design. Furthermore, the present study does not enable the longer‐term follow‐up of young people to understand more about the lasting effects that counselling may have on psychological distress.

The ‘cascade’ training model for counsellors that was employed for the present study is also a limitation in that the authors cannot be sure that a consistent and quality approach was taken. However, due to the limited resources on which this study operated, this was the most appropriate and financially viable option available. Furthermore, no measure of quality or fidelity was used to assess the counselling provided, making it difficult to draw conclusions around which particular model of counselling is associated with these changes.

Finally, whilst data completion rates for the present study were better than those seen in many CAMH services, they were still only around 60–65%, which should be taken into consideration when drawing conclusions from the data.

Conclusions

The present study is the first to provide a systematic, and substantive, dataset on the outcomes of counselling in VCS services for young people and young adults. It indicates that the outcomes of these services are comparable to those seen in CAMHS. Clients seen in VCS services expressed particularly high levels of satisfaction with the services they received.

Commissioners should use the findings from this study to inform their thinking around the services that they commission. Specifically, they should recognize the distinct opportunity that VCS services may provide by being able to bridge the gap in provision as young people transition from child to adult services. Findings from this research also suggest that these services may be able to reach marginalized groups who may not be accessing other services.