Introduction
Over the last decade, cognitive behavioural therapies (CBT) have been widely recognized as effective for many mental health problems and are broadly recommended as a treatment of choice to healthcare services in England and Wales. Considerable investment has been made to increase the availability of CBT therapies for those who might benefit (Department of Health, 2011). The Improving Access to Psychological Therapies (IAPT) programme is a large-scale initiative to extend the reach of evidence-based psychological therapies such as CBT via investment in the training of thousands of new therapists in England, alongside a fundamental restructuring of therapy services and the kinds of interventions they offer. To date, 3600 new CBT therapists have been trained by the IAPT programme and many new high-volume services targeting anxiety and depression have been established, which, in their first 3 years, have seen one million patients (Clark, Reference Clark2011; Department of Health, 2012). The most recent data available indicates that IAPT services are treating an average of 9.7% of the prevalence of common mental health problems in their local communities (Department of Health, 2012). This represents a considerable improvement on earlier estimates of access rates for CBT (Brugha et al., Reference Brugha, Bebbington, Singleton, Melzer, Jenkins and Lewis2004).
In addition to a shortage of CBT practitioners, the availability of CBT to the population has, historically, been further limited by the uneven distribution of therapists across the country. Shapiro et al. Reference Shapiro, Cavanagh and Lomas(2003) found a 20-fold discrepancy in the availability of accredited CBT practitioners between the best and least well-served population deciles; findings were highly indicative of “postcode availability” of the best-qualified CBT practitioners. It is unclear what impact a decade of initiatives aimed at improving access to CBT has had on the equity of availability. The present analysis replicates and extends our earlier report (Shapiro et al., Reference Shapiro, Cavanagh and Lomas2003) to investigate the current availability of CBT in England and Wales, and its comparative geographic distribution 10 years on.
Method
An absolute and comparative evaluation of BABCP membership by postcode area was conducted. To evaluate systematically the extent of inequity in the distribution of CBT practitioners, the 105 postal code areas of England and Wales were cross-tabulated by four counts of CBT practitioners (all members, accredited therapists, all members who are nurses, all members who are clinical psychologists) derived from BABCP membership lists as of May 2012 with the population for each area from 2001 Census data provided by the Office of National Statistics.
BABCP membership records were used to obtain the number of CBT practitioners per 100,000 of population, for each postal code area. Post-code areas (N = 105) were ranked from most to least well-provided with CBT practitioners according to each definition. Postal code areas were then grouped into deciles (10 groups, each containing about one-tenth of the total population of 52.04 million) and ordered according to the number of therapists per 100,000 of population with respect to each criterion in turn. The exact population size of each decile varied somewhat around the average value of 5.20 million, as the entire population of areas falling at the boundaries between deciles were assigned to a single decile. Such assignments were done in such a way as to minimize the variation in population size between deciles.
The cumulative percentage of BABCP members in 2012 per population decile was compared with 2002 data using paired sample t-tests anchored by decile. For this analysis, 2002 data was extracted from our previous study of the geographic distribution of BABCP membership (Shapiro et al., Reference Shapiro, Cavanagh and Lomas2003). BABCP member data were also compared to a model distribution of equitable availability where each population decile would house 10% of the total membership.
Results
In mid-2012, the total number of BABCP members was 9397, of whom 7576 lived in England and 274 in Wales. The total number of accredited members was 3164, of whom 2663 lived in England and 42 in Wales. The total number of BABCP members in England and Wales has doubled during the past decade, and the number of accredited BABCP members in England and Wales has multiplied by 4.5 times.
Nurses (1761; 22%) and clinical psychologists (1858; 24%) remain the largest professional member groups. Clinical psychologist member numbers have increased by 36% and nurse member numbers have increased by 59% over the past decade. Overall, other member groups have increased almost three-fold over the last 10 years (2002: 1500; 2012: 4231). Table 1 shows the geographic distribution of therapists according to each definition (member; accredited member; clinical psychologist member; nurse member) from the most to the least well-provided decile of the population of England and Wales in 2012.
Table 1. Distribution of BABCP members, accredited members, clinical psychologist members and nurse members across 105 postal code areas in England and Wales in 2012
The distribution of BABCP members and accredited members remains highly inequitable; one in five are based in postal code areas occupied by the best-served 10% of the population. There is a five-fold discrepancy in availability of accredited members between the best and least well-served population deciles. The distribution of BABCP clinical psychologist members and nurse members is also inequitable. The best-provided 10% of the population has 15 times more clinical psychologist members and 6 times more nurse members available within its postcode area per 100,000 of population than the worst-provided 10%.
Comparative analysis found that the cumulative percentage of all members in 2012 is significantly closer to equity than the 2002 membership distribution (t(9) = 4.39, p<.01), but still differs significantly from equity (t(9) = 5.68, p<.001). Similarly, the cumulative percentage of accredited therapists in 2012 is significantly closer to equity than the 2002 membership distribution (t(9) = 5.76, p<.001), but still differs significantly from equity (t(9) = 5.67, p<.001). The cumulative percentage of clinical psychologist members in 2012 maps directly onto the 2002 membership distribution – has not significantly improved (t(9) = 1.81, p = 0.1) – and still differs significantly from equity (t(9) = 5.51, p<.001). The cumulative percentage of nurse members in 2012 is marginally more equitable than the 2002 membership distribution (t(9) = 2.54, p = 0.03), but still differs significantly from equity (t(9) = 5.45, p<0.001).
Discussion
This comparative analysis notes a substantial increase in the number of BABCP members during the last decade, and a measurable improvement in equity of the geographical distribution of BABCP members to serve population need. These data suggest that more people for whom CBT is an indicated treatment may now benefit from these therapies, and that in comparison to one decade ago the availability of these widely recommended, evidence-based interventions is now more equitable throughout the population of England and Wales. However, neither the absolute numbers of BABCP members, nor the geographic distribution of these members, indicates that timely and equitable access to CBT therapies is likely to be available for all of those who could benefit.
Continuing geographic inequity in the availability of CBT is most starkly illustrated by findings that the best-provided 10% of the population has 5 times more accredited CBT therapists within its postcode areas than does the worst-provided 10%. Overall, these findings remain indicative of “postcode availability” of CBT.
Further workforce and service development for CBT should be considered an ongoing healthcare priority (Shafran et al., Reference Shafran, Clark, Fairburn, Arntz, Barlow and Ehlers2009). More, and more equitably distributed, CBT therapists will be needed to meet the demands of the IAPT programme, which aims to offer universal equitable access to evidence-based interventions, to treat 15% of all adults with depression and anxiety with CBT, and to improve access to evidence-based psychological therapies, including CBT, in other patient groups such as children and young people, people with severe mental illness, people with long-term physical conditions and people with medically unexplained symptoms (Department of Health, 2011, 2012).
Previously, we discussed hypothetical solutions to the challenge of improving equitable availability of CBT. These include the up-scaling of training efforts and redistribution of therapists to areas with higher unmet need, the potential of tele-therapy to extend the reach of CBT interventions beyond well-served areas, and the potential for increased volume of delivery of treatments based on the principles of CBT via guided self-help interventions (Shapiro et al., Reference Shapiro, Cavanagh and Lomas2003). Each of these aims remains relevant. Of course, such inequitable availability of health care is not confined to CBT nor to England and Wales, and is reflected in continuing policy discussion of variation in healthcare.
Limitations
The replication elements of this report have similar limitations to the original study: On the population (denominator) side, our analysis concerned only geographical equity at postal code area level, and alternative units of geography should be considered. In terms of the numerator (therapists), BABCP membership is an imperfect indicator of the availability of CBT, and no doubt an underestimate of the absolute number of people (members and non-members) providing CBT services in England and Wales. Future research should complement the current study methodology by drawing on additional resources, including IAPT service data, and NHS training and workforce data, in an attempt to provide a more accurate estimate of the availability of CBT.
Acknowledgements
Thanks to Caroline Monroe of the British Association for Behavioural and Cognitive Therapies, Bury, Lancashire, for providing the membership data, and to Emma Scutt of the Office of National Statistics, Titchfield, Hampshire, for providing census data. Thanks also to David Shapiro and Howard Lomas, for their contribution as co-authors to the original work described here, and their comments, along with those of Shirley Reynolds, and three anonymous reviewers, on an earlier version of this manuscript.
Comments
No Comments have been published for this article.