Background
Welfare rights advice has been provided within general practices in the United Kingdom for over 20 years, mostly through Citizen's Advice Bureaux (CAB). The service has been advocated as contributing to reducing health inequalities and improving the health of those in poverty, through financial gain. A systematic review of these services published in 2006 found considerable positive evidence for these services in terms of increased financial income for patients. However, there was little robust evidence that they also impact on health (Adams et al., Reference Adams, White, Moffatt, Howell and Mackintosh2006).
A number of studies have shown that welfare rights service provision benefits people with mental health issues through reduction in anxiety, stress and depression (Adams et al., Reference Adams, White, Moffatt, Howell and Mackintosh2006). Most studies of such services, however, suggest they are used most frequently by older people, those with disabilities and ethnic minorities, (Coppel et al., Reference Coppel, Packham and Varnam1999; Wiggan and Talbot, Reference Wiggan and Talbot2006; Moffatt and Mackintosh, Reference Moffatt and Mackintosh2009) and not by those with mental health problems (Abbott and Hobby, Reference Abbott and Hobby2003). Conversely, over 70% of general practitioner respondents to a postal questionnaire believed there was a mental health component to welfare rights consultations (Harding et al., Reference Harding, Sherr, Sherr, Moorhead and Singh2003).
Relatively, little work has explored the impact of welfare rights advice provision on health services and we have identified no studies that quantified this. A small number of published studies have sought the views of service providers and practice staff on the perceived impact of the service. One such study involving focus groups with health care staff in Bradford identified a perceived reduction in reception staff time helping patients to complete forms, which was confirmed by a questionnaire to practice managers who believed that the workloads of reception staff, nurses and GPs were also reduced (Greasley and Small, Reference Greasley and Small2005). Another exploring the impact of a Welsh national programme of welfare advice identified a belief that the adviser had reduced the workload of the primary health workers (Borland and Owens, Reference Borland and Owens2004).
This latter study also found that both primary care staff and service providers considered the services impacted on the quality of advice available to patients, helped remove any stigma associated with accessing such services, thus improving access and also raised confidence among patients in seeking advice and among health workers in providing it (Borland and Owens, Reference Borland and Owens2004). Other work has also shown that the perceptions of staff in practices where such services are available are that these benefit patients, however, some practices have difficulties in providing services, due to lack of funding and space (Harding et al., Reference Harding, Sherr, Singh, Sherr and Moorhead2002).
Sefton Primary Care Trust (PCT) in Merseyside, England has a diverse population with extremes of deprivation, but CAB Health Outreach (CABHO) services are currently only provided to nine practices, all located in the most deprived areas. The service is well established and routinely gathers data on numbers referred per practice and patient perspectives of the service (Citizens Advice Sefton – Internal report, 2010). No evaluation of the service from the perspective of the National Health Service (NHS) has been conducted. A total of 250 patients were referred to CAB staff during the period April–September 2009. Anecdotally, CAB staff perceived the service to be particularly used by patients with mental health problems (Citizens Advice Sefton – Internal report, 2010). The proportions of people with low mental well-being scores are known to be high in areas of high deprivation, where the service is provided (Deacon et al., Reference Deacon, Carlin, Spalding, Giles, Stansfield, Hughes, Perkins and Bellis2009). Therefore, an evaluation led by the mental health improvement team of the PCT set out to investigate whether there was any evidence that patients referred to the CABHO service did have mental health issues and that the service reduced workload.
The objectives of the study were:
• To determine CAB and general practice staff perceptions on the impact of a CABHO service on staff workload.
• To quantify the frequency of mental health issues among patients referred to the CABHO service.
• To measure any impact of the CABHO service on appointments, referrals and prescribing for mental health.
Methods
As a service evaluation, NHS ethics approval was not required, however, approval to conduct the study was obtained from the PCT Care Quality and External Assessment Group. The evaluation was conducted in two parts.
Views of practice staff and CABHO service providers
Three slightly different structured interview schedules were used to elicit relevant information from practice managers, GPs and CAB staff. These consisted of both closed and open questions designed to obtain factual details of the CABHO service provision and also perceptions of the service's impact on practice staff time and activities, on prescribing and on referrals to mental health services. Information leaflets were sent, along with letters of invitation to participate in the evaluation, to the practice managers of all nine practices receiving the service. The CAB Team Manager was approached to identify CAB staff providing Health Outreach services and to identify the GPs from each of the nine practices who referred clients to the CABHO service.
All nine practice managers, all 15 referring GPs and all five CABHO staff, plus the CAB team manager, also a service provider, were invited to participate in an interview. Interviews with staff who agreed were conducted either face-to-face or over the telephone by one researcher, however, some GPs elected to receive a written version of the interview schedule instead and responded by post. Actual responses to closed questions, brief notes of the response to open questions relating to practical issues and verbatim quotations for responses to the open questions requesting opinions were recorded on paper at the time of the interview. Data from interviews and written responses were combined for analysis by J.K., an experienced health services researcher. Responses to closed questions were simply quantified and the responses to each separate open question were analysed using thematic analysis. The data from GPs were linked to those from practice managers from the same practices and information provided by CAB staff was also linked to the practices where they provided services.
Use of health services
All practices were approached in writing for permission to extract data from the medical records of patients referred to the service during April–September 2009. In those which agreed, data were extracted for all patients referred, over the six months before first appointment with CABHO and six months after, on the following:
• GP and nurse/other appointments, whether these were acute or for monitoring purposes and whether they related to mental health issues, derived from free-text notes.
• Referrals to mental health services and the reasons for these.
• Number of prescriptions for antidepressants and hypnotics/anxiolytics and the number of days treatment with these drugs. Amitriptyline use for other indications was excluded from the analysis.
Data were gathered by two experienced PCT staff, familiar with audit of primary care medical records, checked for completeness and consistency by a third team member and entered into Excel, then transferred to SPSS version 17.0 for analysis. Frequency data were calculated for all measures. Differences between measures before and after appointments with the CABHO service were compared using Wilcoxon signed-ranks test.
Mental health issues were defined as a record of at least one appointment with a GP or nurse in which a mental health issue was recorded, a referral to mental health services or issue of a prescription for an antidepressant/hypnotic/anxiolytic. This definition was used to estimate the proportion of referred patients with any mental health issue. Read codes were not used in view of the known practice of not assigning such codes unless a clear diagnosis was available. Prescriptions for antipsychotic drugs were excluded, as these were more likely to be prescribed for serious long-term disorders.
Results
Practice demography
Within the nine practices receiving the CABHO service there were 32 GPs, ranging from single-handed to 13 GPs per practice (27.5 full time equivalents), but only 15 were identified as referring to the CABHO service. The practice list sizes ranged from 1728 to 16 558, average 6269, covering a total of 56 419 patients. All practices were located in areas within the lowest quintile of deprivation as assessed by Index of Multiple Deprivation 2007.
Interview data
All six CABHO staff providing services, all nine practice managers and four GPs were interviewed and two further GPs returned questionnaires. Collectively, the GP respondents represented five of the nine practices.
All practice managers felt the overall impact of the CABHO service on reception staff time was minimal. Furthermore, none felt that it had any adverse impact on any other services provided by the practices. All nine considered that the service was beneficial to patients, particularly the facility for patients to see someone at the practice rather than go elsewhere. However, none had gathered any data about specific benefits of the service for patients and most (six) felt that it had had no impact on GP appointments, referrals to the mental health services (eight) or on prescribing (nine). Two were of the view that the service may have had an impact on GP appointments, but regarded this as difficult to quantify:
May have done – must have saved GP appointments.
PM5
All GPs considered the CABHO service was beneficial for their patients. Perceived benefits ranged from helping GPs in an area where they are not expert to seeing patients who do not need a doctor:
One benefit to me is removing the burden of knowing nothing about this.
GP2
There was a perception that the service was particularly of benefit to patients with mental health issues:
It would be difficult to quantify, but patients often report reduced stress and anxiety.
GP6
Only one considered that GP appointments, referrals to mental health and prescribing had all definitely reduced. Two others felt that the CABHO service was possibly of benefit in reducing prescribing or had the potential to do so, whereas most felt that it had no impact on appointments and mental health referrals or were unsure of any impact. Only one indicated there were any problems for the practice in hosting the CABHO service relating to availability of space. Conversely, five felt that it reduced their workload, the sixth indicating it was possibly reduced. One GP estimated reduced consultation rates, whereas another estimated gaining time savings of 2 h per week.
Interviews were conducted with the five CAB workers who provided services to all nine practices and the Team Manager. There was agreement that the clients seen at the health outreach service differed from those seen in regular CAB services because of their high frequency of mental health problems and the impact of financial worries on health:
Having worked at a Drop-in Service, the main difference is a bigger proportion of clients at Surgery suffer from Mental Health Problems. These people would fall through the net if we didn't have the service we provide.
CAB 1
Five of those interviewed felt that the service also had a positive impact on practice staff, by saving GP time, allowing them to concentrate on medical issues and reducing pressure on doctors. One believed that medicines use may be reduced and one indicated the service did not impact negatively on practice staff time.
Use of health services
Patient characteristics
Six practices gave permission for the medical records of patients referred to the CABHO service to be accessed. The average list sizes and number of GPs staffing the six practices that were included in this part of the study were greater than in those that did not consent (average list size 7700 versus 3400; average four GPs versus two). However, two of the three single-handed practices among the nine did allow examination of records. All used fully computerised records. A total of 148 records were examined in these practices, ranging from 8 to 36 per practice, a total of 59% of all patients referred by the nine practices. More females (96; 65% of total) than males (52) had been referred to the CABHO service. The ages of the patients are shown in Table 1. Almost all patients had received GP appointments during the study, with 90% having used an acute appointment (Table 2). A total of 68 (46%) of all clients referred to the service may have had a mental health issue, as defined in the section ‘Methods’. All of these had a GP appointment related to mental health, 14 (9%) a nurse appointment related to mental health, 50 (30%) received a prescription for an antidepressant and 26 (18%) for a hypnotic or anxiolytic, while 30 (20%) had a referral to mental health services.
Age distribution of patients using the CABHO service in six practices
CABHO = Citizen's Advice Bureaux Health Outreach
Changes in use of health services among 148 patients using the CABHO service
CABHO = Citizen's Advice Bureaux Health Outreach.
*P < 0.05 Wilcoxon signed-ranks test.
Changes in health service utilisation
The overall changes in use of health services comparing six months before and after referral to CABHO are shown in Table 2. The greatest change seen was in the number of GP appointments, which reduced from an average of 4.90 appointments per patient to 4.26 per patient, a total of 93 fewer appointments for the 148 patients. This reduction was statistically significant (P = 0.017). The average number of nurse appointments was reduced only slightly from 1.50 to 1.35 per patient. Overall appointments that were related to mental health did not change, whereas referrals to mental health services showed a slight increase.
There was a reduction in the number of prescriptions issued for both antidepressants from an average of 1.20 to 0.96 (22% reduction) and hypnotics/anxiolytics from 0.38 to 0.22 per patient (42% reduction), with the latter reaching statistical significance (P = 0.016). Overall, the total number of patients who were taking an antidepressant changed little between the two periods (3%), but there were ten patients who stopped use of a hypnotic/anxiolytic (44%), seven fewer on either drug type and four fewer who made use of any of the mental health services (see Table 3).
Numbers of patients using mental health services before and after referral to the CABHO service
CABHO = Citizen's Advice Bureaux Health Outreach.
*Any practice appointment related to mental health, referral or issue of prescription.
Discussion
This study has demonstrated that almost half the patients referred to the CABHO service in this PCT may have had a mental health issue, which is in line with the perceptions of the CAB staff providing the service. In terms of practice staff workload, there was a perception that the CABHO service had no detrimental effect on staff time. Conversely, in fact, the study found that overall patients referred to the service used fewer GP appointments after referral than in the equivalent time period before. There was also a small, but significant reduction in the prescribing of hypnotics/anxiolytics in patients after referral to the service compared with the period before referral. There were, however, small, non-significant increases in primary care appointments relating to mental health and referrals to mental health services. One possible explanation for the contrasting findings could be that patients had fewer but more productive discussions with GPs about mental health issues following consultations with a CABHO adviser about their welfare problems.
Strengths and limitations
Staff perceptions were obtained from managers in all nine practices and all CAB staff providing the service to these practices, but only from GPs in five practices, which may have resulted in bias. Furthermore, difficulties in reaching GPs resulted in the need for different methods to obtain their views, even though the same structured questionnaire was used, has the potential to result in differing findings; however, none were noted. The small number of GPs in the nine practices who referred patients to the CABHO service and were thus eligible for interview is a further limitation. Although the locality provides a wide range of ‘social prescribing programmes’, such as Active Sefton, Relax and Revive, Active Reading, Creative Alternatives, Wellbeing Sefton, all aimed at improving general well-being through non-medical interventions, the limited number of referrals received from GPs has been acknowledged as an important issue, limiting potential benefits. Other work has determined reasons for this which has been used to amend practice (O'Keefe and Mackridge, Reference O'Keefe and Mackridge2012 unpublished data). Only two-thirds of the practices, involving 59% of all patients referred, permitted medical records to be accessed, thus the data on health service utilisation may have some selection bias. No demographic data were available for the patients referred from the three practices that did not permit data collection. We used a very broad definition of mental health issues, in order to ensure that all possible patients were included, although we recognise that our data do not therefore equate to studies that have used more specific definitions or Read codes.
We recognise that this was an exploratory study and included no control group, which would have enabled comparison of health service utilisation in patients not referred to CABHO. Such a study was not possible because we were setting out to identify if indeed patients referred to the CABHO service did appear to have a high prevalence of mental health issues, as was suspected. The before and after design employed allowed patients to effectively act as their own control, although the six-month periods used for comparison do not account for any seasonal effects and may also be too short to show real differences (Mackintosh et al., Reference Mackintosh, White, Howel, Chadwick, Moffatt, Deverill and Sandell2006). Related to this latter limitation is the finding that improvements in welfare benefits may not be achieved within a six-month period and furthermore, we have no confirmation that the apparent changes we found were related to any actual assistance received from the CABHO service. We also recognise that this design did not permit any other factors, which may have influenced prescribing or referral to be taken into account; however, no specific initiatives relating to such prescribing of the drugs studied or mental health referral processes took place during the study period. The analysis excluded patients prescribed amitriptyline for non-psychiatric indications, as far as it was possible to elicit this from records. No detailed examination of records took place to assess the potential impact of either referral to the CABHO service or other concurrent issues on individuals. Therefore, caution must be exercised in drawing any conclusions from this exploratory quantitative assessment of appointments and prescribing.
Implications for policy and practice
In contrast to other studies, which have shown that few patients with mental health problems use primary care welfare rights services (Coppel et al., Reference Coppel, Packham and Varnam1999; Borland and Owens, Reference Borland and Owens2004; Wiggan and Talbot, Reference Wiggan and Talbot2006; Moffatt and Mackintosh, Reference Moffatt and Mackintosh2009), we found a high proportion of people who appeared to have a mental health issue. We chose to examine this specifically as there was a perception among CAB staff and GPs anecdotally that the service was of particular benefit to patients with mild-to-moderate mental health issues. GPs elsewhere have expressed a similar perception (Harding et al., Reference Harding, Sherr, Sherr, Moorhead and Singh2003). A number of studies have suggested that welfare advice reduces problems such as anxiety and worry and improves mental health (Abbott and Hobby, Reference Abbott and Hobby2003; Adams et al., Reference Adams, White, Moffatt, Howell and Mackintosh2006), and may reduce overall use of health services and medicines (Abbott and Hobby, Reference Abbott and Hobby2003). We are not aware of any other studies that have specifically examined prescribing or appointments to quantify changes following welfare rights advice; however, reductions in both have been found following appointments with support workers for patients presenting with mixed health and social care needs (Abbott and Davidson, Reference Abbott and Davidson2000). Our work suggests that it may be worth conducting larger studies to examine this further.
Mental well-being is increasingly being assessed among the general population. Below average mental well-being, as assessed by the Warwick-Edinburgh Mental Wellbeing Scale (WEMWBS; Tennant et al., Reference Tennant, Hiller, Fishwick, Platt, Joseph, Weich, Parkinson, Secker and Stewart-Brown2007) is high in areas of deprivation within North West England, where this study took place (Deacon et al., Reference Deacon, Carlin, Spalding, Giles, Stansfield, Hughes, Perkins and Bellis2009). Thus, the high proportion of people who were referred to the CABHO service with possible mental health issues reflects this demographic. As a result of this work, staff providing the CABHO service have been administering the WEMWBS to clients on referral and after six months. To date 82 (74%) of 113 people who completed this measure on referral had scores indicating low well-being, which may confirm our finding that many of their clientele at the Health Outreach service may have mental health issues. This very high proportion considerably exceeds the 8.4% of people with low well-being in Sefton PCT as a whole (Harrison et al., Reference Harrison, Deacon, Spalding, Carlin, Giles, Stansfield, Perkins and Bellis2009) and the figure of 16.8% in the whole of North West England (Deacon et al., Reference Deacon, Carlin, Spalding, Giles, Stansfield, Hughes, Perkins and Bellis2009). Initial findings suggest that a high proportion of clients show increases in WEMWBS at six-month follow-up. As welfare advisory services are frequently targeted to more deprived communities, it seems likely that similar services elsewhere may also have considerable numbers of patients with low mental well-being. Regular use of such tools may provide a valuable additional measure of patient benefit.
Although our study found no reduction in appointments for mental health problems or referrals, our data do indicate a reduction in the overall number of GP appointments and prescribing for anxiolytics and hypnotics. These are relatively small changes, which require to be reproduced in other studies; however, they do confirm beliefs of benefits to practice workloads, expressed by practice staff both in our study and elsewhere (Harding et al., Reference Harding, Sherr, Sherr, Moorhead and Singh2003; Borland and Owens, Reference Borland and Owens2004; Greasley and Small, Reference Greasley and Small2005). Although other studies have shown that GPs believed patients were less likely to approach their practitioner for advice on welfare issues (Harding et al., Reference Harding, Sherr, Sherr, Moorhead and Singh2003), practice managers felt that the service reduced staff time spent on helping patients to complete forms (Greasley and Small, Reference Greasley and Small2005).
Expansion of the service has been found to be desirable in surveys of both GPs (Harding et al., Reference Harding, Sherr, Sherr, Moorhead and Singh2003) and practice managers (Harding et al., Reference Harding, Sherr, Singh, Sherr and Moorhead2002) and indeed this issue was raised in our interviews. Lack of funding and space were the principal reasons cited by practice managers for not having in-house advice (Harding et al., Reference Harding, Sherr, Singh, Sherr and Moorhead2002) and one practice in our study also highlighted this space as a problem limiting expansion. With the move to commissioning by primary care consortia, it is important that data are available to support decisions about continuation or expansion of services such as the CABHO. There is existing evidence of the benefits to patients of welfare rights advisory services from the literature, but our preliminary study now also suggests there may also be some reduction in NHS utilisation. Given the limitations of our study, these results should be interpreted with caution and further larger, controlled studies carried out.
Acknowledgements
This work was supported by Sefton Primary Care Trust and Liverpool John Moores University. The authors would like to thank all staff who gave their time to participate in the interviews and for sharing their views. They are also grateful to practice staff for providing access to medical records. The authors also acknowledge the work of Katherine Mather in supporting the data collection from medical records.
