The clinical effectiveness of a cognitive behavioural therapy intervention in a work setting: a 5-year retrospective analysis of outcomes

Abstract

Background

The National Institute for Health and Clinical Excellence (NICE) guidance recommends Cognitive Behavioural Therapy (CBT) as part of multidisciplinary occupational mental health interventions for people with long-term or recurrent short-term sickness absence from work (NICE, 2009). Despite this, there is a paucity of data for both randomised trials for CBT and literature that supports the transferability of CBT into occupational environments.

Aims

This service evaluation aimed to evaluate the clinical effectiveness of CBT by analysing data from a partnership scheme between a local authority and a local heath board using a routine employee population.

Methods

A clinical cohort of 81 employees referred through the partnership scheme completed CBT over a 5-year period via a CBT nurse therapist. A sample of 76 employees was included in the evaluation who completed pre-/post-measures to establish outcome. Of these, 30 were followed up at a 3-year point, completing the same measures.

Results

Each of the clinical measures yielded significant outcomes at 95% confidence intervals, and large effect sizes using Cohen’s d both at post-test and follow-up. No significant difference was shown between post-treatment and follow-up outcomes. CBT was demonstrated to be clinically effective within an occupational mental health setting.

Conclusions

In conclusion, partnership schemes with a focus on mental health between public sector agencies can have a positive outcome for the funding agency as well as individual employees.

Keywords

cognitive behavioural therapy (CBT)CBT nurse therapist occupational mental health partnership schemes workplace intervention

Introduction

Depression and anxiety are associated with work-related stress and are two of the most prevalent causes of work-related ill health and absenteeism (National Institute for Health and Clinical Excellence (NICE), 2009). In the five years following publication of the NICE guidelines for sickness/absence, the number of absences related to common mental health problems rose by 24% (Davies, 2014). In the UK, up to half of all long-term sickness absence is as a result of mental health problems, costing an estimated £26 bn to the economy per year (Houses of Parliament, 2012).

The consequences of depression and anxiety, both as a primary illness and in association with chronic illnesses, are well documented for sickness absence (Health and Safety Executive, 2014). These can include psychological distress leading to or resulting from a fear of returning to work, disturbed relations with colleagues and superiors, work stress caused by multiple factors or possible job loss. The evidence suggests that the presence of depressive symptoms plays a significant role in return-to-work outcomes for a range of illnesses (Mental Health Foundation, 2009). It can also prove difficult for those on sick leave due primarily to depression and anxiety to return to work. This is because the problems are intertwined with work-related stress, perceptions of bullying and harassment or poor working relations. Under these circumstances, many employees are reluctant to interact with their organisations, thus making return-to-work interventions difficult to apply.

It is recognised that public sector health providers, such as the National Health Service (NHS), are often designed to care for serious mental illness. They can struggle to provide timely assessments and intervention for individuals with depression and anxiety disorders. Even with the well-funded Improving Access to Psychological Therapies (IAPT) scheme in England, many are still not receiving the psychological input they require (Clark, 2011), although the advent of employment advisors might assist when work is a specific issue (Hogarth et al., 2013). Such barriers within the healthcare system can prevent an earlier return to work, as many employees have to wait for NHS procedures and interventions before they are considered fit enough to return to work (Mental Health Foundation, 2009).

NICE recommends cognitive behavioural therapy (CBT) as part of a multidisciplinary occupational health intervention for people with long-term absence or recurrent short-term sickness absence from work (NICE, 2009). This is despite a paucity of evidence regarding the value of CBT within occupational mental health for clinical populations (Dibben et al., 2012). For instance, the Cochrane Collaboration (2009), when reviewing interventions to improve occupational health in depressed people, incorporated no individualised CBT studies whatsoever. Furlan et al. (2012) completed a systematic review of interventions for depression in the workplace. Furlan et al. (2012) concluded that there is insufficient evidence to determine which interventions are effective, and concluded that no intervention could be recommended as a consequence. One of the main criticisms of the studies Furlan et al. (2012) analysed was that for the most part they used volunteer sample groups, as opposed to clinically referred individuals with identified mental health problems. This reduces the generalisability of the data found to occupational health populations. Seymour and Grove (2005), in their occupational health review of workplace interventions for people with common mental health problems, recognised the potential benefits of CBT interventions within this field, but acknowledged the lack of methodologically robust data to back this up, and in particular the shortfall of this data within the UK.

Some of this work has now begun outside of the UK. Blonk et al.’s (2006) controlled study compared CBT as usual with individual CBT treatment that had a return-to-work component delivered by labour experts, although the intervention did contain a Rational Emotive Behavioural Therapy (REBT) element to it, possibly making it different in many ways from most recognised CBT today. They established that positive outcomes for return to work were significant for their more specific workplace-based CBT/REBT programme, though they found no difference in this variable regarding the CBT as usual intervention. They recognise, though, that the focus on a graded return in the former group would have made a significant difference in itself. DeVente et al.’s (2008) study tested a CBT-based stress management programme both individually and within a group format. They found no significant findings related to improvement in symptoms or with absenteeism, and conclude that CBT alone is not enough to produce improvements in return-to-work variables.

Lagerveld et al. (2012) followed this up with significant findings in another trial. This dealt with some of the issues raised by their previous paper, including in its sample group individuals with common mental health problems, but still excluding certain diagnoses such as those with major depression and Post-Traumatic Stress Disorder (PTSD). Despite this, the most prevalent diagnosis remained adjustment disorder (67%). The specialised CBT version encouragingly resulted in enhanced return to work, thus providing further economic benefits. They measured this as a 20% reduction in costs for employers whose employees received the work-related CBT intervention, with individuals returning on average 65 days earlier than in the CBT as usual intervention. Once again though, they acknowledge the potential significance of encouraging the employee towards a phased return within their protocol. More recently our own occupational mental health partnership undertook a cost-benefit analysis (Hitt et al., 2016) and calculated that a £1 investment resulted in a £2.47 return to the exchequer. This included moneys gained through reduced sickness/absence, greater presenteeism and a reduction in usage of other health services such as GP appointments and medication. A focus on economic outcomes is considered an important outcome from work-based interventions (Seymour and Grove, 2005).

The Occupational Health Service of City of Cardiff Council identified that the local NHS services were unable to address the needs of employees affected by psychological health conditions as a result of complex work and personal issues. Services in Wales are not IAPT based and therefore waiting lists for evidenced psychological treatments can be lengthy. A dedicated pathway to seek timely assessment, treatment and rehabilitation using established specialist mental health services would overcome existing delays and potentially have a positive effect for employees. The resulting partnership agreement utilising the expertise within the Department of Liaison Psychiatry in Cardiff and Vale University Health Board was instigated in 2009, with an aim to pursue NICE recommended approaches in the form of CBT for the prevention and management of complex occupational mental health conditions. Mental health nurses trained as CBT therapists were employed to liaise with the largely nursing contingent of the council’s Occupational Health Service, aiding communication and contributing to the effectiveness of the service.

This paper, as a form of service evaluation, explores the outcomes of this occupational mental health service over a 5-year period. We aimed to test whether CBT delivered by nursing staff can achieve clinical effectiveness within a routine environment with this population. In doing so, we also evaluated the potential value of such a partnership.

Methodology

The partnership is staffed by two CBT nurse therapists who together make up a whole time equivalent. Both are British Association of Behavioural and Cognitive Psychotherapy (BABCP) accredited and are supervised by accredited BABCP practitioners as well as offering peer support to each other. Both remain registered nurses with the Nursing Midwifery Council. The CBT utilised was based on traditional CBT principles developed from Beckian theory (Beck, 1976), though taking into specific account the systemic nature of organisations. Beckian theory involves the development of a formulation (a model) regarding a client’s difficulties from which maintenance cycles can develop, potentially resulting in anxiety and depressed mood. Targeting what is maintaining a person’s problems can result in a reduction in the impact on mood/anxiety. The following elements were consistent themes within sessions: work-life balance issues, problems with colleagues/bullying, issues around responsibility, assertiveness and perfectionism. When specific disorders were targeted (e.g. PTSD), the up-to-date CBT protocol was applied. Specific to mental health nursing, other elements were also important, including medication dilemmas and issues with problem-solving, goal-setting and case management. Treatment was not manualised, in part due to the heterogeneous nature of the employees’ mental health problems. The service was not capped in terms of sessions. Provisional diagnoses were made by the CBT nurse therapists. Outcomes from this are available in Figure 1, but should be treated with caution as neither could provide formal diagnostic labels for the employees, and it is acknowledged that work-related stress and low self-esteem are not formal diagnoses according to standard diagnostic manuals. CBT practice was enhanced by the partnership working element, which included regular consented correspondence with Occupational Health staff and monthly meetings discussing client progress and business-related issues.

Figure 1.

Primary diagnoses of employees.

The majority of employees of City of Cardiff Council are referred to Occupational Health after 4 weeks’ absence from work as per policy. If deemed appropriate, a decision can be made at this point to refer to the Department of Liaison Psychiatry, where the employee receives an initial mental health assessment with a CBT nurse therapist. The assessment utilises the Hospital Anxiety and Depression Scale (HADS) (Zigmond and Snaith, 1983), nominated as the primary measure as it captures information about both depression and anxiety as well as having been validated previously for use with employees (Andrea et al., 2004; Bocerean and Dupret, 2014). The Patient Health Questionnaire 9 (PHQ-9) (Kroenke and Spitzer, 2001) is employed to measure mood, and the Work and Social Adjustment Scale (WSAS) (Mundt et al., 2002) is used to measure functioning, whilst a decision regarding suitability for CBT treatment is made using the Suitability for Short-term Cognitive Therapy Rating Scale (Gail et al., 2007; Safran et al., 1990). At this point a decision is made about whether to offer CBT treatment or signpost elsewhere. All employees assessed as suitable and treated with CBT completed further measures at a mid-stage and at the conclusion of the treatment process. The PHQ-9 and HADS are endorsed by NICE as validated assessment tools in aiding diagnostic interviews and also for monitoring progress (NICE, 2011).

The HADS measures both anxiety and depression on a scale of 0–21, where 0–7 is considered within a normal range, 8–11 is mild, 12–14 moderate and 15 plus is severe. ‘Caseness’ was established for the population in order to establish that a wide rage of severity was included. Caseness, as cited in other literature (Demyttenaere et al., 2009), is deemed to be a score greater than or equal to 11 on either the anxiety or depression sub-scales using the HADS, whilst a score of less than or equal to 8 would be deemed as in remission. The PHQ-9, a validated measure of mood (Arroll et al., 2010), contains items on sleep, appetite, concentration and interest in activities. Individuals scoring 5, 10, 15 and 20 are measured as having mild, moderate, moderately severe and severe depression respectively. Scores under 5 indicate an absence of depression. It is also suggested that high scores are associated with an increase in sickness/absence (Kroenke et al., 2001). The WSAS (Mundt et al., 2002) measures the impact of mental state on work alongside quality-of-life elements such as relationships/leisure time and family relationships. Scores above 20 indicate moderately severe or worse psychopathology, between 10 and 20 significant functional impairment, whilst a score under 10 is associated with subclinical populations.

An analysis of case notes from the 5 years of the Occupational Mental Health Liaison Psychiatry Service (between September 2009 and September 2014) was carried out with an aim to assess the outcomes of intervention with CBT. The outcomes were assessed by comparing the change in the scores of routinely conducted assessments: HADS, PHQ-9 and WSAS. The baseline scores at the initiation of the intervention were compared with end-of-treatment scores. The work measure on the WSAS was also tested separately to establish whether work perception was impacted by CBT treatment. Information regarding covariates including sickness/absence prior to referral and antidepressant prescription was incorporated to improve rigour, testing the hypothesis that these may have influenced the outcome in a positive direction.

At a 3-year point, as part of the audit process and as an attempt to assess relapse rates, all employees who had been through the CBT programme were contacted. The three clinical measures were sent out to 46 employees who had completed treatment up to that point, with the intention to assess whether their mental health had remained stable alongside their work and social functioning. Of these, 30 were returned.

Based on the categorisation from the Whitehall II study (Marmot et al.,1991) and after discussion with the Human Resources department of City of Cardiff Council, work roles were divided as in Table 1. Sickness/absence was measured using City of Cardiff Council database information, although some of this was missing. This was divided into four groups based on length or frequency of absence (Table 2).

Table 1.

Role division of employees.

Role	Number	Percentage
Manager	7	9.2
Qualified	32	42.1
Support	28	36.8
Manual labour	9	11.8
Total	76	100.0

Table 2.

Sickness/absence of treated employees.

	Frequency	Percent
0 weeks	12	15.8
1–4 weeks	7	9.2
5–12 weeks	17	22.4
13–24 weeks	9	11.8
25 weeks plus	2	2.6
Short-term absences	10	13.2
Other (e.g. suspended)	1	1.3
Not recorded	18	23.7
Total	76	100.0

All statistical analyses were conducted in R (V3.0.2) and SPSS version 23. Descriptive statistics of socio-demographic data are presented for index consultation and post-treatment. Analysis of covariance (ANCOVA) controlling for age, gender, sickness/absence, working diagnosis and medication prescription was used to quantify the change in outcome controlling for the baseline/index consultation score. Results were summarised using regression coefficients and 95% confidence intervals (CIs). Effect sizes were calculated following the Cohen’s d method (Cohen, 1988). Follow-up data also controlled for length of time elapsed since discharge, as this differed for each employee.

From the perspective of ethical approval, as this study was conducted as a service evaluation/audit of client data, collected as part of routine clinical management, no further ethical approval was required. Anonymity and confidentiality was protected throughout the evaluation process. This was clarified with the local Research and Development department within Cardiff and Vale University Health Board.

Results

A total of 263 referrals from the Occupational Health Service of the City of Cardiff Council were made to the Department of Liaison Psychiatry in the first 5 years. Out of those referred, 81 completed the CBT treatment, although 23 remained in treatment at the endpoint. A total of 76 patients completed the standardised clinical measures (see Figure 2). Two employees completed alternative measures, two were not asked to complete measures due to employment-related reasons and one set was absent. Analysis of the clinical records of these five individuals demonstrated that four made significant progress, which was not recorded, whilst one made less significant progress. We recorded 31 dropouts from treatment. We were unable to follow up these individuals so no further data could be collected. A high proportion of the dropouts were from the teaching profession (qualified and unqualified) (65%). Demographically, 61% (46) of patients were female and 39% (30) were male. Ages were split into four categories, 18–29 (n = 11), 30–39 (n = 22), 40–49 (n = 26) and 50–65 (n = 17). The mean age was 43. Neither gender nor the age of the employee at the outset of treatment was statistically significantly associated with outcomes.

Figure 2.

Flowchart showing pathways of clients referred to partnership over 5 years.

The employees had varied job titles and roles. The largest proportion of referrals were from schools (29%), followed by social services (13%). In terms of role categories (Table 1), the largest number was from a ‘Qualified’ group (42.1%). Results indicated a significant negative effect of being in the ‘management’ group (containing managers at various levels across council departments), which negatively correlated with outcome, though numbers in this group were low (9.2%). No other occupational group correlated with outcomes. There were no significant differences between outcomes for employees from different departments within the council.

The most common working diagnoses were depression (n = 11, 14.5%), mixed anxiety/depression (n = 12, 15.8%), PTSD (n = 10, 13.2%) and panic disorder (n = 8, 10.5%) (see Figure 1). There were no significant differences regarding outcome for these different diagnoses (compared with depression, mixed anxiety/depression: −1.97, 95% CI (−5.5, 1.6)).

Most of the referrals (80% where recorded, see Table 2) were for sickness absence as opposed to difficulty coping at work with mental health difficulties, in accordance with the intentions of the partnership arrangement. The period of sickness absence prior to referral was not linked to outcome. There was no pattern established in terms of whether employees who had longer sickness/absence were any less likely to improve. This was true using each of the clinical measures. Antidepressant prescription was not a significant covariate related to outcome against any of the measures.

All outcomes from clinical measures are described in Table 3 for our sample of 76 employees. The mean HADS scores were reduced from 24.64 (moderate severity) to 8.95 (within normal range) by end of treatment, with a mean difference of 15.69 (95% CI (13.72 to 17.66), p < 0.0001). The overall effect size was 2.59 (large). The mean scores for the PHQ-9 reduced from 15.50 (moderate severity) to 3.77 (within normal range), with a mean difference of 11.73 (95% CI (10.26 to 13.19), p < 0.0001). The effect size was 2.61 (large). The mean WSAS reduced from 23.92 (moderately severe) to 7.20 (subclinical), with a mean difference of 16.72 (95% CI (14.75 to 18.68), p < 0.0001). It achieved an effect size of 2.79 (large).

Table 3.

The clinical effectiveness of a CBT intervention in a work setting.

Clinical measure	Mean score (SD) (pre- treatment)	Mean score (SD) (post- treatment)	Degrees of freedom (post- treatment)	p (95% CI) (post- treatment)	Mean score (SD) (follow- up)	Degrees of freedom (follow-up)	p (95% CI) (follow- up)	Effect size (post- treatment)	Effect size (follow-up)
HADS	24.64 (6.78)	8.95 (5.45)	75	.000	11.16 (8.49)	29	.000	2.59	1.71
HADS (Anx.)	14.34 (3.81)	6.27 (3.14)	75	.000	7.66 (4.95)	29	.000	2.55	1.31
HADS (Dep.)	10.36 (4.39)	2.86 (2.88)	75	.000	3.50 (4.21)	29	.000	2.17	1.73
PHQ-9	15.50 (5.99)	3.77 (3.32)	75	.000	4.76 (4.46)	29	.000	2.61	2.04
WSAS	23.92 (7.71)	7.20 (6.20)	74	.000	10.50 (9.64)	29	.000	2.79	1.73
Work (WSAS)	5.44 (2.45)	2.09 (1.77)	60	.000	2.0 (2.60)	21	.000	1.92	1.36

As indicated in Table 3, the work component of the WSAS was completed by 61 employees. This was partly due to some being unable to complete this due to their work status, for example some employees were being dismissed or remained on sick leave at the end of treatment. Work perception mean scores reduced from 5.44 (where 6 indicates a ‘marked’ impact on work functioning) to 2.09 (where 2 indicates a ‘slight’ impact on work functioning). This achieved significance (95% CI (1.44 to 4.01), p < 0.0001) with an effect size of 1.92.

In terms of caseness, 70% of employees using the HADS scored between moderate and severe ranges (n = 66). Of those individuals who were above the cut-offs at the outset (n = 66), 53 employees (80%) moved from caseness to remission.

In terms of follow-up at a 3-year point in the partnership, 30 out of 46 responses were returned. The time interval between the completion of therapy and the completion of this follow-up was variable. On average, the follow-up period was 469 days (approximately 1 year and 3 months). This was controlled for and no association was found, so employees followed up after a longer period did as well on average as those whose follow-up period was shorter. Table 3 indicates that each of the clinical measures, including the measure of work perception, maintained significance at 95% CI and continued to demonstrate large effect sizes, though these had reduced slightly in all cases. Further analysis established that the increase in scores at follow-up was not clinically or statistically significant compared with post-treatment. There appeared to be a slight association of the HADS with female gender: female employees improved by an average of 13.5 points more than male employees. We also became aware of two employees who may have shown evidence of relapse at this stage (a return to caseness), and they were offered a further assessment. This was calculated as a 7% relapse rate.

Discussion

The results provide evidence that CBT in the workplace, delivered as a partnership between the NHS and a local authority and by mental health trained nurses could be associated with improvements in mental health outcomes. All the clinical measures resulted in significant outcomes at 95% CI and achieved large effect sizes. Significant improvements were demonstrated for measures of mood and anxiety alongside the WSAS, suggesting that CBT could also be associated with improvements in functioning within this population. The results compare favourably with other CBT trials with working populations, as many of the employees scored highly for caseness at outset, with a significant number in the moderate to severe category (n = 66). We are therefore able to suggest that CBT could work for a routinely referred population where exclusion based on severity does not exist. The results were not better explained by the addition of antidepressant medication or sickness/absence, as these were accounted for in the analysis. The ‘work’ outcome on the WSAS, whilst not validated for use in this way, provided some evidence that employees’ work perceptions changed directly as a consequence of CBT. This may lead to greater presenteeism (productivity in work), and, as suggested by evidence from the ‘economic outcomes’ data (Hitt et al., 2016), a reduction in sickness/absence. Employees on longer-term sick leave did as well as employees with minimal sickness/absence, which is an encouraging outcome.

Dropout from the CBT treatment group was high (n = 31), a significant proportion of which were from the teaching population (65%). This was caused in part by the difficulty experienced by teachers in being released from the classroom for treatment. This may be a consideration for the future development of service options for this particular population. One of the obvious issues of operating a work-based intervention where a percentage of the clients have returned or remain in work is the difficulty in assuring that they are supported to attend. Schools in particular might have found the pressure of releasing staff in school time quite difficult to support, or else the individuals themselves may have found it a difficult environment to leave part way through the day.

In regard to the differential impact dependent on employment type, we can offer a tentative reinforcement of previous data suggesting high stress levels within the school and social work populations (Health and Safety Executive, 2014). Consideration of the large size of these populations within the local authority is required. Despite the high levels of stress in specific populations, teachers and social workers attained similar outcomes to those of other populations of workers within the council, suggesting that CBT might be a useful intervention across different departments.

In terms of the differing employment groups, there is a suggestion that management may do less well in terms of outcome when compared with the other groups. Although the sample of managers was quite small, it may still reflect the extra stresses that managers may experience as part of their role. We may have expected a larger result for this group considering the overall significant changes, but this might reflect the systemic elements that are often more difficult to overcome within this population, for instance greater pressure on their roles as a consequence of austerity measures within their departments. There is evidence to suggest that managers do experience high levels of stress (Lu et al., 2000) that impact on their mental health, so protecting these vulnerable individuals from the weight of expectations of their role should be a consideration for employers.

Follow-up data at a 3-year point (n = 30/46) suggest that outcomes can be maintained over time with large effect sizes. Although these are less strong for each of the measures compared with post-treatment, the difference was not significant, and so any deterioration over time was minimal. Two participants from this sample had shown evidence of relapse with scores returning to a moderate to severe level (a return to caseness). This remains a low rate of relapse (7%), which was a positive finding. Females (n = 20) demonstrated a 13.5 point average decrease in the HADS score during follow-up compared with males (n = 10). Using our primary measure, this is an indication that the relapse rate differs and females were more likely to maintain gains made using our primary measure. It is difficult to conjecture why this might be the case, but it is an interesting observation. Obviously the numbers here are low and therefore the results need to be interpreted with caution.

In terms of the limitations of this study, clearly the sample was not randomised and there was no control group, and therefore we are not able to say definitively whether CBT made all of the difference to the employees’ outcomes. We have attempted to control for certain factors such as sickness/absence and antidepressant medication in order to exclude variables that could potentially improve outcomes. Due to the role of the CBT therapists (both coming from a nursing background) we have been unable to achieve a formal diagnosis that might allow us to benchmark against randomised controlled trials for the use of CBT with specific diagnoses. Nor have we used a manualised approach in regard to CBT, due in part to the heterogeneous population of individuals treated, though this could be considered a strength as well as a weakness. We also have some missing data, though have attempted to minimise and account for this where possible. We did have a group who dropped out of treatment (n = 31), but due to the lack of follow-up of these individuals our sample group represents a ‘treatment completer’ as opposed an ‘intention to treat’ sample. There is also a possibility that the partnership nature of our working relationship, with regular meetings and liaison with occupational health, may have resulted in enhanced outcomes. We cannot exclude this as a possible variable, but feel it was a useful addition to the service.

We might refer to this study as adding to the effectivess (i.e. a pragmatic trial in a real-world setting) as opposed to efficacy data provided by controlled studies, both for the use of CBT within a patient population generally, and more specifically within a work setting. Further studies that evaluate the use of CBT within a controlled employment setting would be recommended as the numbers of studies are limited, as well as other studies looking at novel partnership arrangements utilising CBT and their subsequent effectiveness within routine populations.

Conclusion

This study is intended to add to the literature regarding the use of CBT to improve mental health within an employee well-being environment, for which the number of studies is limited. As well as improving mood and anxiety levels, it has shown that CBT is associated with improvement for employees both in terms of reducing the impact of mental health on work perceptions and improving their general functioning. These outcomes were achieved within a routine environment where exclusion based on severity did not occur. The improvements also remained significant over time. This study also supports the clinical effectiveness of CBT, thereby adding to the literature regarding its potential transferability from clinical research to routine practice.

The use of CBT within employee well-being structures has been slow to develop due in part to the general acceptance of counselling services within many workforces being sufficient. This is despite NICE guidance. The outcomes from this study suggest that clinical effectiveness can be achieved for a broad range of mental health problems and a wide range of severity, using CBT as the tool for change. This study also supports the partnership approach among public sector organisations, utilising the skills within one to benefit the other, with no agenda other than to improve outcomes for the employee. Recommendations as part of the audit cycle for the development of future partnerships might include a particular focus on management as a population, as well as more flexible working patterns, in order to allow all staff groups to attend.

Integrated and collaborative working enables the provision of mental health care closer to the workplace, which is key to effective occupational rehabilitation. It is hoped for the future that further partnerships can be developed in this way, having a cost-benefit to the wider economy as well as providing positive outcomes for employees and the funding organisation.

Key points for policy, practice and/or research

CBT can be effective in a routine population of employees with a wide variety of mental health problems, and may also make a difference to the perception of their work.

The results gained may continue over a period after completion of treatment leading to greater presenteeism and reduced sickness/absence.

The partnership working with the organisation may contribute to the outcomes achieved, which may be enhanced by the employment of nurses both for the CBT component and also within Occupational Health departments.

Occupational health partnership schemes such as this should be a consideration within all public- and private-sector organisations.

Randomised controlled trials need to be undertaken to be assured the effects are solely related to the CBT intervention.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Ethics

This service evaluation did not require ethical approval. Anonymity and confidentiality were protected throughout the evaluation process.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

David Hitt is a cognitive behaviour therapy nurse therapist at Cardiff and Vale University Health Board, and programme tutor for CBT programmes at Cardiff University.

Tayyeb Tahir is a professor at Cardiff and Vale University Health Board.

Lyndon Davies is a specialist nurse practitioner at South Wales Fire and Rescue Service.

Jo Delahay is a cognitive behaviour therapy nurse therapist at Cardiff and Vale University Health Board.

Mark Kelson is a research fellow in statistics/senior statistician in the South East Wales Trials Unit, Cardiff University.

References

Andrea

Bültmann

Beurskens

AJHM

et al. (2004) Anxiety and depression in the working population using the HAD Scale. Social Psychiatry and Psychiatric Epidemiology 39(8): 637–646.

Arroll

Goodyear-Smith

Crengle

et al. (2010) Validation of PHQ-2 and PHQ-9 to screen for major depression in the primary care population. Annals of Family Medicine 8(4): 348–353.

Beck

(1976) Cognitive Therapy and the Emotional Disorders, New York: International Universities Press.

Blonk

RWB

Brenninkmeijar

Lagerveld

et al. (2006) Return to work: A comparison of two cognitive behavioural interventions in cases of work related psychological complaints among the self-employed. Work and Stress 20(2): 129–144.

Bocerean

Dupret

(2014) A validation study of the Hospital Anxiety and Depression Scale (HADS) in a large sample of French employees. BMC Psychiatry 14(354): 1–11.

Clark

(2011) Implementing NICE guidelines for the treatment of depression and anxiety disorders: The IAPT experience. International Review of Psychiatry 23: 318–327.

Cohen

(1988) Statistical Power Analysis for the Behavioural Sciences, 2nd edn. Hillsdale, NJ: Lawrence Erlbaum.

Davies

(2014) Chief Medical Officer (CMO) Annual Report: Public Mental Health, London: Department of Health.

Demyttenaere

Verhaeghen

Dantchev

et al. (2009) ‘Caseness’ for depression and anxiety in a depressed outpatient population: Symptomatic outcome as a function of baseline diagnostic categories. Primary Care Companion to the Journal of Clinical Psychiatry 11(6): 307–315.

10.

De Vente

Kamphuis

Emmelkamp

et al. (2008) Individual and group cognitive-behavioral treatment for work-related stress complaints and sickness absence: A randomized controlled trial. Journal of Occupational Health Psychology 13(3): 214–231.

11.

Dibben P, Wood G, Nicolson R, et al. (2012) Quantifying the effectiveness of interventions for people with common health conditions in enabling them to stay in or return to work. Department of Work and Pensions Report no. RR812, 1 October. London: Department for Work and Pensions.

12.

Furlan

Gnam

Carnide

et al. (2012) Systematic review of intervention practices for depression in the workplace. Journal of Occupational Rehabilitation 22(3): 312–321.

13.

Gail

Talbot

Annable

et al. (2007) Suitability for short term cognitive behavioural therapy. Journal of Cognitive Psychotherapy 21(4): 334–345.

14.

Health and Safety Executive (2014) Stress-related and Psychological Disorders in Great Britain 2014. London, UK: Health and Safety Executive.

15.

Hitt

Tahir

Davies

et al. (2016) Case study: An economic analysis of a cognitive behavioural therapy service in the workplace. Mental Health Practice 19(6): 25–31.

16.

Hogarth T, Hasluck C, Gambin L, et al. (2013) Evaluation of employment advisors in the Improving Access to Psychological Therapies programme. Department for Work and Pensions Research Report no. RR826. Sheffield: Department for Work and Pensions.

17.

Houses of Parliament (2012) Mental health and the workplace. Postnote no. 422, October. London: Parliamentary Office of Science and Technology.

18.

Kroenke

Spitzer

(2001) PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine 16(9): 606–613.

19.

Lagerveld

Blonk

RWB

Brenninkmeijer

et al. (2012) Work-focused treatment of common mental disorders and return to work: A comparative outcome study. Journal of Occupational Health Psychology 17(2): 220–234.

20.

Shu-Fang

Cooper

et al. (2000) Managerial stress, locus of control, and job strain in Taiwan and UK: A comparative study. International Journal of Stress Management 7(3): 209–226.

21.

Marmot

Stansfeld

Patel

et al. (1991) Health inequalities among British civil servants: The Whitehall II study. The Lancet 337(8754): 1387–1393.

22.

Mental Health Foundation (2009) Returning to Work: The Role of Depression, London: Mental Health Foundation.

23.

Mundt

Marks

Shear

et al. (2002) The Work and Social Adjustment Scale: A simple measure of impairment in functioning. The British Journal of Psychiatry 180(5): 461–464.

24.

National Institute for Health and Clinical Excellence (NICE) (2009) Managing long-term sickness absence and incapacity for work (CG19), London: National Institute for Health and Care Excellence.

25.

Safran

Segal

Shaw

et al. (1990) Patient selection for short-term cognitive therapy. In: Safran

Segal

(eds) Interpersonal Process in Cognitive Therapy, New York: Basic Books, pp. 257–267.

26.

Seymour L and Grove B (2005) Workplace interventions for people with common mental health problems. Evidence review and recommendations. Report for the British Occupational Health Research Foundation, September. London: British Occupational Health Research Foundation.

27.

Zigmond

Snaith

(1983) The Hospital Anxiety and Depression Scale. Acta Psychiatrica Scandinavica 67(6): 361–370.