Three areas of counselling service delivery are discussed below - counselling in primary care, student counselling, and workplace counselling. These settings have been selected because they represent major areas of counselling practice and can be taken as exemplars for research into counselling outcomes as a whole.
Counsellors have been employed in primary care settings in the UK since the 1980s, generally receiving referrals from GPs for patients reporting problems in such areas as relationship difficulties, moderate depression, anxiety and stress. Counselling is available in more than 50 % of GP practices in England (Mellor-Clark, 2000) and protocols have been developed which define the role and of counsellors in primary care (National Primary Care Research and Development Centre, 2001). A Scottish Office report conducted in the 1990s noted the potential value of counselling in primary care but called for further research to establish its effectiveness (National Medical Advisory Committee, 1998). A number of investigations of the effectiveness of counselling in primary care have been published, encompassing both controlled trials, and naturalistic studies.
Early randomised trials have compared the effectiveness of counselling with treatment as usual from the GP (Ashurst & Ward, 1983; Boot et al., 1994; Friedli et al., 1997; Harvey et al., 1998; Hemmings, 1997). More recently, randomised trials have compared counselling, cognitive-behaviour therapy (CBT) and treatment as usual (Ridsdale et al., 2001; Simpson et al., 2000, 2003; Ward et al., 2000), with patient preference conditions, and between counselling and drug treatment (Bedi et al., 2000). The patient groups taking part in these studies have included individuals assessed as suffering from moderate and chronic depression, anxiety and chronic fatigue. This set of studies has been systematically reviewed by Bower et al. (2002,2003). The key findings to emerge from these reviews have been:
• significantly greater clinical effectiveness in the counselling group compared with 'usual care' in the short term but not the long term;
• similar levels of clinical effectiveness when counselling is compared to CBT or to drug treatment;
• higher levels of user satisfaction with counselling compared with other treatments;
• similar total costs associated with counselling and usual care over the long term;
• some evidence that counselling can reduce antidepressant use and number of GP visits.
These studies have all comprised pragmatic randomised trials, which have involved compromise between methodological rigour and the realities of routine service delivery in a primary-care setting (Scott & Sensky, 2003). Overall, the results of these studies suggest that counselling represents an intervention that is equivalent in effectiveness to any of the other forms of primary care intervention currently available for patients reporting moderate levels of mental health difficulty. The only distinctive advantage of counselling, against other interventions, is that patients prefer it and are more satisfied at the end of treatment.
Evidence from a number of naturalistic or uncontrolled studies allows a more comprehensive understanding of the role of counselling in primary care to emerge. In a major study conducted by Mellor-Clark et al. (2001), more than 2 800 clients receiving primary care counselling were invited to complete the CORE evaluation questionnaire (Evans et al., 2000) at intake and end of counselling. Counsellors also completed questionnaires recording information about the type of therapy that was delivered. The findings of this study yield a unique picture of counselling in primary care practice. The average number of sessions received by clients was 4.3, with 32 % attending weekly, and 42 % fortnightly. The main approaches used by counsellors were person-centred, CBT and integrative. The mean age of client was 38 years, with women accounting for 70 % of the sample. The average waiting time was six weeks. At intake, 76 % of clients scored above clinical 'caseness' level on the CORE questionnaire, and 43 % were on medication to help with their psychological problems. Of those who completed questionnaires at termination, 75 % recorded clinically significant levels of change. The interpretation of these findings is problematic, however, due to the low (40 %) return rate for end-of-counselling client questionnaires in this study. Nettleton et al. (2000) collected evaluative data from a range of stakeholders involved with three GP practices in a rural area and reported that the counselling service was found to fill a gap by addressing the needs of a substantial group of patients for whom psychiatric care was seen to be inappropriate. Martinus et al. (2001) evaluated the effectiveness of a counselling service for alcohol problems, operated by a voluntary agency in a set of GP practices and found that 96 % of GPs regarded the service as very helpful and that the average alcohol use of clients had reduced substantially. Fletcher etal. (1995) examined the relationship between the provision of counselling and the prescribing of antidepressants, hypnotics and anxiolytics in general practice. They looked at differences between practices that directly employed a counsellor and those who had a visiting counsellor or who referred elsewhere. Highest levels of prescribing were in practices that employed a counsellor directly. Sibbald et al. (1996) analysed prescribing patterns for psychotropic drugs in practices with (n = 126) and without (n = 88) counsellors, drawn from a national survey. They found no appreciable differences for prescribing rates or drug costs.
In addition to studies that have evaluated the effectiveness of specialist counsellors in primary care, there has also been research into the use of counselling interventions by nursing staff in primary settings. Klerman et al. (1987) examined the efficacy of a brief psychosocial intervention for symptoms of stress and distress among patients in primary care delivered by nurse practitioners within minimal training (12 hours) in brief counselling. At follow-up, 83 % of counselling patients had returned to the non-clinical range of anxiety and depression scores, compared with 63 % of controls. There was some evidence that patients who had received counselling subsequently used more physician time. Holden, Sagovsky & Cox (1989) conducted a controlled study of health visitor intervention in the treatment of post-natal depression. Experienced health visitors trained for 6 hours in a Rogerian approach offered an average of nine weekly sessions of counselling to depressed women who had been randomly allocated to counselling or a treatment as usual condition. Patients who had received counselling showed significantly higher improvement on depression and psychiatric symptoms at follow-up (one month after the end of counselling).
In conclusion, the evidence as a whole suggests that counselling in primary care is highly acceptable to both patients and health professionals and produces short-term gains for patients compared to treatment as usual. Over the longer term, however, differential gains arising from counselling, as compared to treatment as usual, have not been observed. The effectiveness of counselling is equivalent to that achieved by other interventions that are available for primary-care treatment of moderate mental health problems, specifically CBT and drug treatment. The impact of counselling on health-care costs, such as drug bills and GP time, is unclear. It is possible that some patients who have benefited from counselling may require less assistance in future from the health-care system, whereas other may have been empowered to demand more care.
It seems clear that further research is needed in order to determine the effectiveness of counselling in primary care, in relation to a series of key factors. The quality of the collaborative relationship between GPs and counsellors employed in primary care practices would seem a crucial variable in terms of the appropriateness of referrals, and in containing difficult or complex cases. However, apart from a qualitative study by Hemmings (1997), this important contextual aspect of counselling in primary-care practice has not been examined. It would also be useful to evaluate the role of counselling from a stepped care perspective (Davison, 2000). In practice, as illustrated in the study by Mellor-Clark et al. (2001), many GPs seek to manage patients with moderate mental health problems within their own resources before referring them to counselling and may even regard counselling as a potential 'step' that in turn may lead to later referral to clinical psychology or a community mental health team. Finally, it is necessary to recognise that the outcome variables such as depression scores that have been used to evaluate the effectiveness of counselling in the majority of controlled studies are at best proxy measures for kind of learning and development, and capacity to use personal and social resources, that counsellors seek to facilitate in clients. It may be that a more sensitive evaluation of the longer term benefits of counselling in primary care will require the construction of appropriate methods for assessing these dimensions of change.
Although counsellors have been employed in virtually all universities and in many colleges since the 1960s, there exists only a very limited evidence base regarding the effectiveness of counselling in this sector. Currently, there are no systematic reviews of the effectiveness of counselling in university and college settings available in the literature. Keilson, Dworkin & Gelso (1983) found that open-ended and time-limited counselling (eight sessions) were both more effective than a control condition for students seeking help for personal and social difficulties. In a study carried out at a UK university, Rickinson (1997) found that 79 % of undergraduate student clients reported symptom levels consistent with psychiatric 'caseness' at intake. Following counselling, clients recorded substantial improvement on a range of mental health dimensions. A comparison group of students who did not receive counselling exhibited minimal change over the same period of time. Turner et al. (1996) found that time-limited counselling had a modest impact on adjustment in a sample of student clients. Vonk & Thyer (1999), in a study of the effectiveness of short-term counselling (average 10 sessions), found that, compared to students in a waiting-list condition, those who received counselling recorded substantial levels of clinical improvement. In this study, 73 % of clients met the criteria for psychiatric 'caseness' at intake. Research has also examined the impact of counselling on the retention of students in higher education. Turner & Berry (2000) analysed the records of 2 365 student users of a university counselling service, seen over a six-year period, in terms of graduation and retention rates. Compared with students who had not used the counselling service, former clients were more likely to re-enrol in the year following counselling (85 % versus 74 %), and were equally likely to graduate. Wilson et al. (1997) examined the academic records, after a two-year period, of students who had made use of counselling, and found that counselled students were 14 % more likely than their non-counselled counterparts to remain enrolled. Rickinson (1998) followed up final-year students receiving counselling. While 95 % of these students reported that their academic performance had been affected by their problems prior to entering counselling, all of them graduated successfully.
Research into university and college counselling indicates that short-term counselling interventions are effective in addressing personal and social problems experienced by students, and in enabling students to remain within the educational system. Further research is required to establish the generalisability of these findings.
Research into the outcomes of Employee Assistance Programme (EAP) and workplace counselling provision has largely focused on the impact of this type of intervention on three areas of psycho-social functioning:
• psychological, for example symptoms of stress and anxiety, and self-esteem;
• attitudes to work, for example job commitment and satisfaction
• work behaviour, for example sickness absence, job performance and accidents.
Several studies have also used client satisfaction as an indicator of the effectiveness of EAP/counselling programmes and some researchers have analysed the economic costs and benefits of workplace counselling provision. A scoping search conducted by McLeod (2001) identified 39 studies that examined the outcomes of workplace counselling. However, some of these studies were of poor methodological quality whereas others described interventions that comprised specific forms of psychotherapy, rather than counselling.
All of the studies that evaluated the effectiveness of specialist counsellors operating in workplace contexts used naturalistic designs, in which clients receiving counselling under routine circumstances were invited to complete questionnaires at the beginning of counselling, at the end, and at follow-up. Cheeseman (1996) carried out an investigation of two employee counselling services in the National Health Service. He reported that client levels of distress at the beginning of counselling were significantly higher than organisational norms and similar to psychotherapy out-patient clinical profiles. Substantial reductions in levels of distress and symptomatology were recorded between entering counselling and follow-up. Somewhat smaller, but also statistically significant, gains were found in relationship problems and use of positive coping strategies (particularly social support and the use of rational reflection). In an evaluation of a stress counselling service for Post Office workers, Allinson, Cooper & Reynolds (1989), Cooper et al. (1990) and Cooper & Sadri (1991) found that, at the beginning of counselling, clients had anxiety, depression and somatic scores that were significantly higher than those in a normative sample of UK workers. Post-counselling, clients showed significant improvement in anxiety, depression, somatic symptoms and self-esteem. There was no change in job satisfaction and a significant decrease in organisational commitment. Sickness absence events and days off both reduced significantly. Goss & Mearns (1997) evaluated the effectiveness of a local authority education department counselling service and found that 65 % of clients reported that counselling had improved their problems, and that sickness absence rates for a six-month period following counselling improved by 62 %. Guppy & Marsden (1997) analysed the outcomes of counselling provided for employees who had been referred to an alcohol misuse programme in a transportation company. Significant improvements were reported in mental health (35 % of clients showed clinically significant levels of gain), client and supervisor ratings of work performance and in absenteeism. There was no overall change in job commitment or satisfaction. Highley-Marchington & Cooper (1998) collected outcome data from nine different EAP/workplace counselling schemes, and found a high level of client satisfaction with services. Significant differences were found pre- and post-counselling on mental and physical health scales but no differences on job satisfaction measures. There was a significant improvement in sickness absence. Reynolds (1997) examined the effectiveness of counselling provided to local authority staff and found high levels of satisfaction with counselling, with 80 % of clients recording clinically significant change in depression. No effect on absenteeism or job satisfaction was found. Worrall (1999) evaluated the effectiveness of an external counselling scheme for public sector workers, and found high levels of client satisfaction and clinically significant levels of change in presenting problem ratings in 58 % of clients.
A limited amount of research into the economic costs and benefits of workplace counselling has been carried out. Blaze-Temple & Howat (1997) analysed the cost/benefit of an Australian EAP, and found that the service produced significant cost savings in absenteeism and turnover. Cost benefit ratio for counselling compared with no counselling was 1:1 (the EAP paid for itself). Bruhnsen (1989) conducted a cost-benefit analysis of university EAP and estimated a cost-benefit saving ratio of 1.5:1. The most significant savings were in alcoholism cases (21 % drop in sickness absence) and drug abuse cases (50 % drop). McClellan (1989) conducted a cost-benefit analysis of a local government EAP service in the US and concluded that the service did not produce a financial benefit to the employer because no plausible cost offsets could be identified in such areas as health insurance costs, paid sick leave, productivity improvements or employee turnover.
Although research into workplace counselling and EAP services consistently yields a generally positive view of the effectiveness of counselling in this setting, it is important to take account of the methodological limitations of the research base. While the studies that have been carried out on the impact of counselling on mental health, wellbeing and organisational commitment arrive at strong conclusions about the benefits of counselling, in almost all of these studies significant proportions of clients failed to complete end-of-counselling or follow-up questionnaires. Since it is likely that questionnaires are more often completed by satisfied service users, it is probable that the results of these somewhat overestimate the effectiveness of counselling by an amount that is impossible to quantify. The findings of economic studies of workplace counselling are equivocal and it is clear that further research of this type needs to be carried out in this area.
Was this article helpful?