Introduction
Research has shown that worklessness is bad for health, both at the individual and socioeconomic level. Moreover, remaining in, or a quick return to work after a period of sickness absence has many positive benefits (Waddell and Burton, Reference Waddell and Burton2006; Black, Reference Black2008; Black and Frost, Reference Black and Frost2011). Data from the UK Health and Safety Executive showed that in 2010/2011, around 1.1 million people were suffering from ill-health that was caused, or made worse, by their occupation. In addition, around 27 million working days were lost as a result of work-related ill-health (WRIH) or injury, at an estimated cost to the UK economy of £13.4 billion (Health and Safety Executive, 2011/2012).
It has also been widely recognised that general practitioners (GPs) may struggle with their contractually obliged role in the sickness certification system. Citing a lack of WRIH training, and a concern to maintain the doctor–patient relationship, GPs may feel ill-equipped to offer advice to patients about return to work issues (Hiscock and Ritchie, Reference Hiscock and Ritchie2001; Hussey et al., Reference Hussey, Hoddinott, Wilson, Dowell and Barbour2004; Wynne-Jones and Mallen, Reference Wynne-Jones and Mallen2009; Money et al., Reference Money, Hussey, Thorley, Turner and Agius2010). This may be unsurprising as only an estimated 4% of GPs have undertaken specialised postgraduate training in occupational medicine (OM) (Howie, Reference Howie2005; Health and Social Care Information Centre, 2009).
In an effort to reduce the burden of sickness absence in the United Kingdom, recent government initiatives have included the introduction of a Statement of Fitness for Work, or fit note, in April 2010 (Department for Work and Pensions, 2010). Instead of declaring a person to be simply fit or unfit for work, the fit note allows GPs to indicate that a person may be fit for some types of work, and to suggest approaches to facilitate a return to work including a graded return, altered work hours, amended duties and workplace adaptations. The fit note has attempted to shift focus to what a person can do, in terms of the clinical condition and its effect on function, and consequent fitness for work.
Initial fit note evaluations are mixed. From the GPs’ perspective evaluations are largely positive (Welsh et al., Reference Welsh, Mallen, Wynne-Jones and Jinks2012; Chenery, Reference Chenery2013; Shiels et al., Reference Shiels, Hillage, Pollard and Gabbay2013). However, the reaction from industry and employers is less so (Thomson et al., Reference Thomson and Hampton2012; EEF, 2012; 2013). A fundamental change to the well-established certification system might produce difficulties for GPs who, as reported, can find this aspect of their job both complex and challenging (Tellnes, Reference Tellnes1989; Hiscock and Ritchie, Reference Hiscock and Ritchie2001; Aylward, Reference Aylward2004; Hussey et al., Reference Hussey, Hoddinott, Wilson, Dowell and Barbour2004; Roope et al., Reference Roope, Parker and Turner2009; Wynne-Jones and Mallen, Reference Wynne-Jones and Mallen2009; Money et al., Reference Money, Hussey, Thorley, Turner and Agius2010). Given that the focus of the fit note is now very much on what a person can do, it is essential that GPs (via appropriate training) have an understanding of patients’ work tasks, and how their day-to-day employment might cause or aggravate ill-health (Coole et al., Reference Coole, Watson and Thomson2013). GPs need to be able to make recommendations about the best way to manage a condition while keeping the employee in work, or for getting patients back to work as quickly as possible after a period of absence.
To start to address the training issue, workshops were set up by the Department for Work and Pensions (DWP) in order to provide essential training in health and work so that GPs could complete fit notes effectively. However, a national survey in 2012 found that only a small proportion of GPs (1 in 10 in England and 2 in 10 in Scotland and Wales) had received any WRIH training in the previous 12-month period (Hann and Sibbald, Reference Hann and Sibbald2013). The effects of training GPs in OM on their attitudes and capacity to provide work adjustment advice are unknown. Given the focus on return to work issues and keeping employees in work, there is a presumption that GPs trained in OM might behave differently (ie, more positively) to GPs without OM training. Therefore, the broad aim of the study reported in this paper is to determine whether GPs with training in OM have different perceptions and attitudes regarding the fit note, compared to non-OM trained GPs.
Method
The Centre for Primary Care at the University of Manchester was commissioned by the DWP to establish a baseline measure of GPs’ knowledge, attitudes and behaviour towards health and work in GB in 2010, against which future work could be benchmarked (Hann and Sibbald, Reference Hann and Sibbald2011). Questions relating to GPs’ views on work and health; their role, training and confidence in promoting the health benefits of work; early experience of fit notes; and the availability of back to work services, were included in the 6th National General Practitioner Worklife Survey. A total of 30 questions were developed via a number of strategies. These included a review of policy documents and published research; discussion with policy customers in the Health Work and Well-Being Strategy Unit (HWWB) and also with DWP officials. These questions were subsequently tested via GP focus group and cognitive testing of candidate items in a number of GP interviews. Results were discussed with the HWWB and the questions refined or discarded as appropriate. This resulted in the development of the final 19 questions for use in the survey (See Hann and Sibbald, Reference Hann and Sibbald2011 for a fuller overview).
Questions 1–18 (Q1–Q18) were organised into themes – see Table 1 – and GPs were asked to indicate their answer by ticking one of the four options – completely disagree/somewhat disagree/somewhat agree/completely agree. For the majority of questions, it was anticipated that GPs would agree with the statement: only for Q6, Q7, Q15 and Q16 was there an ‘expectation’ that GPs would disagree with the statement. The phrase ‘favourable response’ is used to describe the anticipated opinions in this paper. For Q17 and Q18 an additional (fifth) option was added, that is, a ‘don’t know’ response. Question 19 required a simple yes/no response about health and work training in the previous 12 months.
Table 1 National GP worklife survey themes and questions
In 2012, the 7th National GP Worklife Survey (Hann and Sibbald, 2011), included a cross-sectional element incorporated into the original survey design, which allowed comparison between GPs with prior training in OM and non-OM trained GPs. GPs with training had been trained to Diploma level in Occupational Medicine of the Faculty of Occupational Medicine and were reporters to The Health and Occupation Research network in General Practice (THOR-GP), a project which aims to determine the incidence of occupational disease and sickness absence burden in the United Kingdom (Hussey et al., Reference Hussey, Turner, Thorley, McNamee and Agius2008).
The 7th National GP Worklife Survey was posted to a random selection of 4179 GPs; 2995 were drawn from the NHS Prescription Services data set for England, 750 from the NHS Shared Services Partnership data set for Wales and 434 Scottish GPs were drawn from the Information Services Division data set. This sample covers ~10% of all GPs from England, Scotland and Wales. To this sample, an additional 182 THOR-GP reporters who had not been selected as part of the ‘main’ sample – were added, giving a total of 4361. Questionnaire packs (covering letter, survey questionnaire and pre-paid return envelope) were sent between September and November 2012. Non-responders were mailed up to twice more at three weeks intervals. Confidentiality was maintained by identifying GPs using a unique serial number known only to the research team. Ethical approval was given by the North West Multi-centre Research Ethics Committee (MREC 11/NW/0832).
Analysis of survey responses
Survey responses were analysed using Stata (StataCorp, 2011). The representativeness of the respondents was assessed and probability weights (see below) were subsequently derived. The weighted ‘raw’ data were then summarised for each of the 19 questions, for THOR-GPs and non-THOR-GPs separately.
The analysis was undertaken on three GP groups: (1) GPs trained to Diploma level in OM – THOR-GPs (n=107). Around 10% of the non-THOR-GPs stated that they had undertaken some form of training in work and health (but of undefined duration, ie, a GP road show or equivalent) in the previous 12 months (ie, answered ‘yes’ to Q19), therefore the non-THOR-GPs became two further groups, (2) those GPs who had undertaken some work and health training, (n=173) and (3) GPs with no work and health training (n=1483).
GPs’ responding to the ‘completely disagree’ and/or ‘completely agree’ options were few in number (for Q1–Q10 the response to either end of the spectrum was <5%; for Q11–Q16, the response was <10%) therefore the four response categories were collapsed into two, that is, agree/disagree, for analytical purposes. Binary logistic regression was used to determine whether there were significant differences in responses between the three groups, adjusting for age group, gender, contract type (GP provider versus other contract) and country. Odds ratios (OR) and their 95% confidence intervals (95% C.I.) are presented to illustrate the strength of these associations.
Results
Excluding duplicated GPs and undelivered questionnaires (<1.5% of mailout), the overall response rate to this 7th National survey was 40.4% (40.2% England; 40.4% Wales; 41.5% Scotland). The response rate of the THOR-GP group was 48.6% (48.3% England; 38.0% Wales; 60.0% Scotland).
In general, the greatest response biases within the national random samples were the same across all three countries. There was an over-representation among GPs aged 50–59 years, and a corresponding underrepresentation of the youngest (under 35 years), and oldest GPs (60 years and over). In England and Wales, GPs with non-provider contracts were underrepresented. In addition, female GPs were underrepresented in Wales.
Within the THOR-GP sample, female GPs were underrepresented across all three countries. Moreover, nearly half the THOR-GPs (47%) are employed in training practices, with larger than average practice list numbers. In order to mitigate these biases, and to ensure that the respondent sample more closely reflected the population it was designed to represent, we derived country-specific ‘probability’ weights for the national data, and THOR-specific weights for the THOR-GPs (all countries combined). These weights were the inverse of the probability of being sampled (not applicable for the THOR sample, as all THOR-GPs were surveyed) – by age group, gender and contract type – in each country, adjusted for non-response.
Analysis of weighted raw responses
Table 2 shows that the overwhelming majority of GPs agreed that work was beneficial for health (Q1: 98.6%) and that worklessness was detrimental to people’s health (Q2: 93.0%). The majority of GPs also agreed that helping patients to stay in or return to work was an important part of a GP’s role (Q3: 89.6%) and an indicator of success in the clinical management of working age patients (Q4: 75.9%). Around two-thirds of GPs agreed that they had a responsibility to facilitate a patient’s return to work (Q5: 66.6%). Most GPs disagreed that patients had to be fully recovered from their condition before they recommended a return to work (Q6: only 19.9 agreed). However, over three-quarters of GPs agreed that they felt obliged to issue sickness certificates for reasons not strictly medical (Q7: 78.3%). The majority of GPs agreed that they felt confident in dealing with ‘return to work’ issues (Q8: 62.1%). Self-reported current knowledge of sickness certification was good (Q9: 80.8%), but current knowledge of the benefits system was poor (Q10: 27.8%). The majority of GPs reported positive impacts of the fit note on the quality of consultations (Q11: 65.6%, Q12: 64.1%) and on the outcomes for patients of fit note consultations in their making a return to work (Q13: 60.1%, Q14: 84.5%). GPs were roughly evenly split about whether fit notes had lengthened consultation times or not (Q15: 52.1%): while only a minority reported that it had made no change to their practice (Q16: 28.9%). Only a small proportion of GPs perceived that good services were available locally where they could refer patients for advice (Q17: 19.2%) or support (Q18: 18.0%).
Table 2 Proportion of agreementFootnote a with survey questions by all GPs and by GP group
THOR-GP=The Health and Occupation Research network in General Practice.
a Results expressed as percentages relate to proportion of GP groups ‘agreeing’ with the question whereby agreement is the ‘favoured’ response unless indicated by a * in which case the ‘favoured’ response was to disagree.
THOR-GPs were generally most likely to give the favourable response, that is, to agree to (Q1–Q5, Q8–Q10) or disagree with (Q6–Q7) questions relating to ‘relationships between work and health’, a ‘GPs role in helping return patients to work’, ‘return to work management’ and ‘knowledge of the certification system’. Absolute differences between THOR-GPs/trained non-THOR-GPs and non-OM trained GPs, in excess of 10% (which could be viewed as clinically significant, but this is simply a threshold chosen by the authors, it could equally be set at 5%, 15%, 20%) were observed in response to Q8, Q9 and Q10. Moreover, in response to Q8, THOR-GPs and trained non-THORs also differed by >10%. This was also the case for Q6. However, on questions relating to the ‘impact of the fit note’ and ‘service support availability’, GPs with some training in work and health gave the favourable response more frequently to Q11–Q16. Collectively, their opinions of the fit note were very similar to THOR-GPs. The responses of other GPs (with some training in OM) to Q11, Q13 and Q15 differed by >10 absolute percentage points compared with non-OM trained GPs. In fact, overall, the responses of GPs with no training appeared to be notably different from the other two groups. We have tested these differences formally, controlling for GP characteristics. The results by ‘theme’ are discussed in more detail below.
Logistic regression analyses
The regression modelling used ‘degree of training’ as the key independent variable. Of note, we have only reported OR for significance at the 5% level. Other pairwise, between-group comparisons can, therefore, be assumed to be non-significant at this level.
Relationship between work and health (Q1–Q2)
Given that nearly all (98.6%) of the GPs in the study agreed with the statement ‘work is generally beneficial for people’s health’, we did not conduct a regression analysis on the responses to this question, as there was insufficient variation. There was a statistically significant difference [OR=2.74; 95% C.I.=(1.25, 6.01); P=0.012] between THOR-GPs and GPs with no-OM training in response to the statement ‘worklessness is generally detrimental to people’s health’, with the former group being more likely to agree with the statement. However, as very few GPs disagreed with this statement, the findings should be viewed with caution.
GPs role in helping return patients to work (Q3–Q5)
There was a statistically significant difference between THOR-GPs and GPs with some training [OR=2.84; 95% C.I.=(1.17, 6.86); P=0.021] and between THOR-GPs and GPs with no-OM training [OR=3.07; 95% C.I.=(1.50, 6.31); P=0.002] in their agreement with the statement ‘helping patients to stay in or return to work is an important part of a GP’s role’: THOR-GPs were more likely to agree than either of the other GP groups. THOR-GPs were also more likely to agree with the statement ‘staying in or returning to work is an important indicator of success in the clinical management of people of working age’, compared with GPs with no-OM training [OR=1.71; 95% C.I.=(1.11, 2.65); P=0.016]. For the statement, ‘GPs have a responsibility to society to facilitate return to work’, there was a significant difference between GPs with no-OM training and THOR-GPs [OR=1.62; 95% C.I.=(1.06, 2.46); P=0.026], and a difference between GPs with some training and GPs with no-OM training [OR=1.42; 95% C.I.=(0.99, 2.03); P=0.056]: the two GP groups with training were more likely to agree with the statement.
Return to work management (Q6–Q8)
THOR-GPs were significantly more likely to disagree with the statements that ‘patients had to be fully recovered before recommending a return to work’ [OR=1.87; 95% C.I.=(1.12, 3.14); P=0.017) and that GPs ‘felt obliged to give sickness certificates for reasons that are not strictly medical’ [OR=1.68; 95% C.I.=(1.10, 2.57); P=0.017] than GPs with no-OM training. In terms of GPs’ confidence in dealing with return to work issues, THOR-GPs [OR=2.85; 95% C.I.=(1.76, 4.62); P<0.001] and GPs with some training [OR=1.70; 95% C.I.=(1.17, 2.48); P=0.005] were significantly more likely to agree with the statement than GPs with no-OM training.
Knowledge of certification system (Q9–Q10)
THOR-GPs [OR for Q9=4.24; 95% C.I.=(2.33, 7.73): OR for Q10=2.21; 95% C.I.=(1.55, 3.15)] and GPs with some training in work and health [OR for Q9=3.22; 95% C.I.=(1.79, 5.82): OR for Q10=2.68; 95% C.I.=(1.91, 3.74)] were significantly more likely to agree with both statements than GPs with no-OM training (P<0.001 for all pairwise comparisons). There were no significant differences between THOR-GPs and GPs with some training.
Impact of fit note (Q11–Q16)
Overall, GPs’ responses to the fit note statements were positive and are summarised in Table 3. THOR-GPs and GPs with some training were significantly more likely to agree that the fit note had improved the quality of their discussions with patients (Q11) and advice given (Q12), and increased the frequency with which they recommended a return to work (Q13) than GPs with no-OM training. There were no significant differences between THOR-GPs and GPs with some training. GPs with some training were statistically significantly less likely to disagree that the fit note had increased consultation length than GPs with no-OM training (Q15).
Table 3 Responses to fit note section of Worklife Survey
THOR-GP=The Health and Occupation Research network in General Practice; OR=odds ratios; C.I.=confidence interval.
Support service availability
Few GPs agreed with both statements regarding good local services to which they could refer patients for ‘advice’ and ‘support’ in returning to work. GPs with some training were more likely to agree to both statements than GPs with no-OM training [OR for Q17=2.25; 95% C.I.=(1.56, 3.25): OR for Q18=2.28; 95% C.I.=(1.57, 3.30): P<0.001 in both cases] and with THOR-GPs [Q18 only: OR=1.84; 95% C.I.=(1.07, 3.14); P=0.026].
Aggregated response analysis
Primarily, we have focussed on GPs’ responses to individual questions. However, we were also interested in their ‘aggregated’ responses within themes (excluding support service availability). The mean scores and the percentage of GPs who gave the favoured response to all questions within the theme – including by ‘degree of training’ – are shown in Table 4. For the themes ‘Relationship between work and health’, ‘GPs’ role in helping patients return to work’ and ‘Return to work management’ the mean scores and percentages of GPs giving all favourable responses increased with increasing OM training. However, this was not the case for the themes ‘Knowledge of certification system’ and ‘Impact of fit note’, where the most favourable responses were seen in GPs with some training in OM. Arguably of most interest was the ‘Impact of the fit note’ theme, where only one in six GPs gave the favoured response to all six questions. This figure varied very little between GPs with different levels of training. On average, overall, GPs gave the favoured response to four out of the six questions.
Table 4 Mean scores by theme for all GPs and GP groups
THOR-GP=The Health and Occupation Research network in General Practice.
The vast majority of GPs (>90%) gave the favoured response to both questions enquiring about ‘relationships between work and health’, while just over half (56%) did so for the three questions on ‘GPs’ roles in helping return patients to work’. GPs with some training were more likely to give the favoured response to all three questions in this theme. Only one in seven GPs gave the favoured response to all three questions about ‘return to work management’: this was likely to have been predominantly influenced by responses to Q7, on which the majority of GPs agreed they issued sickness certificates for reasons not strictly medical. Only two in seven GPs gave the favoured response to both questions relating to ‘knowledge of the certification system’: this can be attributed to poor knowledge of the benefits system (Q10). Again, GPs with some training were more likely to give favourable responses to both questions on this theme.
Discussion
The underlying hypothesis of the study is that GPs with training in OM, when compared with a randomly selected sample of GPs, will have a more positive attitude towards the fit note. The analyses undertaken show that after controlling for age group, gender and contract status, THOR-GPs and GPs with some training were, in general, the groups of GPs most likely to indicate the favoured response with more positive attitudes to return to work and fit note issues than GPs with no-OM training. Research undertaken before the introduction of the fit note with GPs in THOR had shown that many of these GPs felt that their training was indeed helpful when it came to certifying absence (Money et al., Reference Money, Hussey, Thorley, Turner and Agius2010). Work in Scandinavia has also shown that GPs with postgraduate training in occupational health tend to issue fewer sickness certificates, and those GPs who also worked as part-time industry medical officers, certified shorter periods of absence (Tellnes, Reference Tellnes1989). The study reported here found that female GPs and older GPs (aged 55+) consistently indicated the favoured response more often than male GPs and younger GPs. Previous work has found that a GP’s gender and age are likely to have an impact on doctor–patient consultations and outcomes, in particular an increased prevalence of male GPs certifying ‘intermediate’ (6–28 week) sickness absence for male patients (Shiels and Gabbay, Reference Shiels and Gabbay2006).
GPs who had undertaken some training in work and health in the previous 12 months were more likely to agree, than GPs with no-OM training, that the introduction of the fit note had increased their consultation times, and that there were good services locally to refer patients for return to work advice and support. These findings may be a result of the recent training in work and health issues that the GPs reported to have received. Training GPs to understand patients’ work roles and how adjustments may help keep that person in work, or return to work after sickness absence provides a challenge, but as reaction from employers has shown, this is crucial for realising the potential of the fit note system (EEF, 2012; 2013). Some evidence indicates, however, that fit notes are failing to return employees back to work early, and crucially, the fit note advice received by employers from GPs has yet to improve (Coole et al., Reference Coole, Watson and Thomson2013; Confederation of British Industry, 2013). Indeed although revised guidance for GPs means they no longer have to give fit note advice specific to the individuals’ job, rather the patients’ general fitness for work (Department for Work and Pensions, Health and Well-Being Directorate, 2013), some employers feel that GPs continue to lack effective training to fully engage with the fit note process (EEF, 2013). One evaluation of 58,700 fit notes issued between 2011 and 2013 found that in 7% of cases, the GP had not provided any additional advice for the employer and/or patient (Shiels et al., Reference Shiels, Hillage, Pollard and Gabbay2013). A recent CBI report (2013) found that only 10% of employers were confident that GPs were sufficiently trained to use fit notes differently to sick notes, and only 5% believed GPs to have a sufficient understanding of the workplace to be able to use fit notes to employers’ potential benefit. In addition, over two-thirds (68%) of the employers surveyed stated that they were not confident in the training GPs had received for the fit note system.
THOR-GPs and GPs with some training were significantly more likely to agree that they have a responsibility to society to facilitate return to work. The pressure to certify sickness absence within the traditional doctor–patient relationship has long been acknowledged as a barrier in the sickness certification system, with GPs prioritising their role as patients’ advocate above their role as gatekeeper (Hiscock and Ritchie, Reference Hiscock and Ritchie2001; Hussey et al., Reference Hussey, Hoddinott, Wilson, Dowell and Barbour2004; Wynne-Jones and Mallen, Reference Wynne-Jones and Mallen2009; Money et al., Reference Money, Hussey, Thorley, Turner and Agius2010). It is interesting to note that THOR-GPs were more likely than the other two GP groups to disagree with the statement that there was an obligation to give sickness certificates for non-medical reasons. Money et al. (Reference Money, Hussey, Thorley, Turner and Agius2010) found that many GPs trained in OM expressed resistance to the ‘traditional’ doctor–patient relationship and stated that their training in OM helped them during sickness absence negotiations with patients, and often facilitated a proactive approach to certification via the inclusion of employers and occupational health services. The wider recognition of the potential impact of the doctor–patient relationship has been addressed through the DWP and Department for Health funded Fit for Work Service (FFWS) pilot schemes (which were one of the recommendations put forward in the review of the working age population; Black, Reference Black2008). Some of the evaluations (Hillage, Reference Hillage2012; LeicesterFit4Work, 2012; Thomson et al., Reference Thomson and Hampton2012; Allied Health Professions Federation, 2013) identified the importance of using an ‘impartial’ GP (ie, not the patients’ advocate) for undertaking fit for work consultations and signing of fit notes. It is believed that the use of an impartial, occupationally trained GP/other health professional to offer advice could help improve the management of patients at risk of having long-term absence from work. Following this pilot work, and in response to the difficulties some GPs appear to have completing fit notes, the DWP is implementing (late 2014) an independent FFWS. This service will be made available (via referral) to patients entering, or expected to enter, four weeks of sickness absence. The patient will undergo an assessment by an occupational health professional who will explore the issues preventing a return to work. A plan will be drawn up outlining actions and recommendations for a return to work, and this will, in effect, replace the employers’ need for a fit note from the GP with respect to evidence for Statutory Sick Pay (Department for Work and Pensions, 2014). The Fit for Work scheme may end up alleviating some of the ‘return to work’ burden that GPs face, however, GPs will still require an understanding of the issues surrounding health and work in order to recognise and deal with absences of one to four weeks which may require adjustments, phased returns, meetings with management/occupational health and so on but which will not be referred to the scheme
Strengths and weaknesses
The study originally had two main objectives: to investigate whether GPs with training in OM have different perceptions and attitudes regarding the fit note, compared with non-OM trained GPs (as reported here) and second, to investigate whether GPs with training in OM may differ from other GPs in their management of patients with health and work problems. This second objective was to be explored in a parallel project which aimed to determine the recognition of ill-health caused or aggravated by work in GPs without OM training. However, we were hampered by an extremely low response rate: out of 1975 randomly selected GPs (from the main Worklife sample) presumed not to have OM training, only 15 (1%) agreed to participate, therefore further research to fulfil this second objective was not pursued. This lack of engagement from GPs is not uncommon in primary care research (Hillage, Reference Hillage2012; Rannard et al., Reference Rannard, Gabbay, Sen, Riley and Britt2013), particularly with respect to long-term worklessness, sickness absence certification and rehabilitation issues. Rannard et al. note that, in their experience, workplace health appears to be an area of practice in which GPs report a lack of knowledge, confidence and training, and therefore tend to avoid research precisely because of the challenging issues associated with these areas.
It could be argued that a weakness of the study was the response rate to the first objective (non-OM trained GPs 40%; THOR-GPs 49%). However, this is much improved when compared with the response to the 6th National GP Worklife Survey in 2010 (34%) and is in keeping with the response rates to the first five National GP Worklife Surveys (response rate range of 40–50%) (Hann and Sibbald, Reference Hann and Sibbald2011).
The GPs in the National GP Worklife Survey were selected from nationally representative samples and the response biases observed, particularly in terms of an underrepresentation of younger GPs and female GPs’ were corrected and adjusted via ‘probability’ weights to ensure the sample more closely reflected the population it was designed to represent. The geographical profile of the group of THOR-GPs taking part is also highly comparable to the coverage of GP practices in the United Kingdom with preliminary analysis of the population captured by THOR-GP showing it to be nationally representative (Hussey et al., Reference Hussey, Turner, Thorley, McNamee and Agius2008)
The eligibility criteria for GPs reporting to THOR-GP is that they are trained to Diploma level in OM, and as noted, it is estimated that only around 4% GPs have undergone this specialist training (Howie, Reference Howie2005; Health and Social Care Information Centre, 2009). It could be argued, therefore, that these GPs (given their training) already function to a higher standard in terms of having an appreciation of the issues surrounding health and work, notably rehabilitation back to work and corresponding workplace adjustments. Therefore, a more favourable perspective among these GPs might have been expected?
GPs still had limited awareness of local services to which they could refer their patients for advice and/or support about a return to work, which is an area of potential concern. It is unclear though whether GPs thought that such services did not exist in their local area or that they were simply unaware of them. This would depend on the GPs’ local knowledge and/or experience, neither of which we have an indication of in this context. It is, therefore, possible that respondents indicated that they disagreed rather than they did not know. Nevertheless, our findings indicate that the responses from GPs who have undertaken training to Diploma level in OM, and GPs who had received some form of work and health training in the previous 12 months (the nature and length of which is unknown), are associated with significantly more positive attitudes to the fit note and a GPs role in helping patients return to work. Significance needs to be considered in the context of absolute differences: many of the differences between groups – in percentages in agreement with the statements – are small in absolute terms. Their statistical significance is due to the large sample sizes. Some differences might be significant purely by chance; one would expect 1 in 20 tests to be so. That said, the THOR-GP group and the GP group with some training in health and work do appear more supportive and knowledgeable about patients’ health, work and well-being. This study reveals a difference between trained and non-OM trained GPs in their attitudes to the fit note and to work and health generally; a further study to evaluate the effect of OM on the recognition and management of ill-health is recommended.
Acknowledgements
The National General Practitioners Worklife Survey was funded by the Department of Health Policy Research Programme. The Health and Well-being Directorate at the DWPs funded the questions in the survey relating to GPs’ attitudes to work and health. This project (no: 138) was funded by the National Institute for Health Research (NIHR) School for Primary Care Research (SPCR). THOR-GP is part of THOR network and programme of research, and is partly funded by the Health and Safety Executive (HSE).
Conflicts of Interest
None.
Disclaimer
This article/paper/report presents independent research funded by the National Institute for Health Research (NIHR). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health.
