Variation in referral and access to new psychological therapy services by age: an empirical quantitative study

Abstract

INTRODUCTION

In 2009, the Royal College of Psychiatrists suggested that the UK currently provides more mental health services for those <65 years of age than it does for older people.¹ Barriers to mental health treatment are known to differ by age² and concerns that discrimination against older adults may lead to reduced access to mental health treatments, including talking therapies, are expressed.³ Data available from the Improving Access to Psychological Therapies (IAPT) programme, which aimed to equitably improve access, allows exploration of the degree to which inequalities persist across different age groups.

The Adult Psychiatric Morbidity Survey (APMS) provides data on the prevalence of psychiatric disorder (treated and untreated) in the English adult population. ‘Common mental health problems’ (CMHPs) is an umbrella term describing difficulties in low mood and anxiety, and is a descendant of the term ‘common mental disorders’. The APMS describes common mental disorders as conditions comprising different types of depression and anxiety.⁴ Data from the APMS suggest prevalence of depression reduces with age from middle age onwards,⁴ a finding supported by others⁵ but not by Stordal and colleagues,⁶ who reported a linear increase of depression with increasing age after carefully adjusting for confounders for individuals in Norway. Whatever the truth about prevalence, it is clearly apparent that there is inequity of access. In the 1990s fewer than 3% of adults aged >65 years reported seeing a mental health professional.⁷ Using data from the APMS (2007), Cooper and colleagues⁸ found younger adults (16–34 years) to be 80% more likely than older adults (≥75 years) with the same severity of CMHPs to be receiving talking therapy. In contrast, older adults were more likely than younger adults to receive antidepressants or anxiolytics and hypnotics, suggesting that older adults are being prescribed medication rather than talking therapy and vice versa for younger adults here.

Inequities in access to talking therapies across age may be dependent on patient attitudes, practitioner attitudes, and/or system factors. Depressive symptoms are common in older adults, but psychological adjustment to ageing and chronic illness may mean symptoms are not acknowledged or revealed.⁹ It may be assumed by some healthcare professionals that older adults experience psychological distress as a natural and inevitable consequence of ageing.^9,10 Ross and Hardy¹¹ found GP decisions were influenced by patients’ help-seeking behaviours as well as their representations of mental health problems. GPs may believe older adults are less responsive to cognitive behavioural therapy than younger adults.^12,13 Patients’ belief that they can ‘manage themselves’ increases with age, making it less likely that they will disclose their mental health problems to a GP, less likely that they will be referred, and less likely that they will accept offered treatment.^14,15 It is possible that older adults attribute their symptoms to physical complaints, whereas younger adults have greater awareness of psychological problems.¹⁰ Older adults’ inability to express their psychological problems, and their greater self-stigma, reduce the likelihood that they will seek help,¹⁰ or be offered appropriate help by healthcare professionals,¹¹ creating further inequities in access to mental health treatment on the basis of age.

In 2008, the development of new psychological therapy services across England under the IAPT programme improved access to treatments for CMHPs,² encompassing individuals with obsessive-compulsive disorder (OCD) and post-traumatic stress disorder (PTSD), as well as the more prevalent anxiety and depression. By 2011, 142 primary care trusts (PCTs) had an IAPT service and £400 million was being invested up to 2015. Targets for rates of access to psychological therapy included services providing enough therapists to meet the needs of the whole PCT population, with half of those who completed the programme moving to recovery.¹⁶ IAPT services include routine collection of session-by-session outcome data for all individuals referred. The IAPT data enable exploration of differences in access to mental health treatments across age, allowing comparisons of referrals, access to the services, and responses to treatment.

Previous audits and monitoring suggest access to talking therapies is greater for younger individuals.⁸ This study aimed to accurately estimate differences in referral and access rates to the IAPT services and to compare the pathway through treatment across age bands, controlling for predicted prevalence of CMHPs.

METHOD

The overall design of the current study was to derive figures in each age band for total population, estimated prevalence and numbers referred, being seen, and patients who achieved the minimal clinical important difference (MCID) in symptoms; and then to calculate rates using different key denominators. The most recent APMS (2007)⁴ data were analysed with data collected from IAPT services in 13 primary care trusts (PCTs) from 2010–2011. The dataset was created for a service evaluation project of the IAPT services commissioned by the South West Strategic Health Authority in England.

Population and prevalence data were obtained from the APMS survey, which used a robust stratified, multi-stage probability sample of households, and which assessed and diagnosed psychiatric disorder according to diagnostic criteria where possible. Participants completed the revised Clinical Interview Schedule (CIS- R),¹⁷ which measures symptoms linked to the diagnosis of anxiety and depression, and provides an overall score for the presence of CMHPs (indicated by a score of ≥12). The APMS series is the largest, most detailed, and most recent (2007) data (at the time of collection) available for comparison of IAPT services use. The APMS provided numbers on estimated prevalence of CMHPs across sex (which the authors of the current study combined), as well as total population, in each of the 13 PCTs individually. Using this data, the numbers of people in each age group in the South West and the numbers of people in each age group in the South West who were estimated to have CMHPs were totalled, and these were used as denominators to calculate rates and proportions.

Numbers of people referred, obtaining access, and the associated clinical outcomes were derived from the South West IAPT evaluation database, which includes information from the IAPT service providers’ databases relating to 13 South West PCTs. Individuals aged 18–74 years of age were included in the study to reflect data taken from the APMS (which excludes those in care homes). The anonymised referral, access, and outcome data were generated from 82 513 individual episodes of care (76 734 patients). Patients’ demographic information (including age) and details of attendance and health outcomes were recorded at every clinical contact. The latter included the Patient Health Questionnaire (PHQ-9)¹⁸ as a measure of depression, and the Generalised Anxiety Disorder (GAD-7)¹⁹ questionnaire as a measure of anxiety.

The authors of this study estimated the proportions of the total populations having CMHPs; being referred to IAPT; obtaining access by attending their first session with an IAPT therapist; engaging in treatment by attending at least one further session; and those achieving MCID in both the PHQ-9 and the GAD-7. The proportions were calculated of those estimated to have CMHPs being referred to IAPT; obtaining access by attending their first session with an IAPT therapist; engaging in treatment by attending at least one further session; and those achieving MCID. The proportions were then calculated of referred patients obtaining access; engaging in treatment by attending at least one further session; and those achieving MCID. MCID was also shown as a proportion of those with two or more treatment sessions.

To calculate the MCID patients must have attended two or more (valid) clinical contacts and improvement was measured by comparing final therapy session score with baseline session score on the following outcome measures. The MCID value for PHQ-9 is a reduction of ≥5 points between first and last session, and for GAD-7 it is a reduction of ≥4 points between first and last session.²⁰ The proportion achieving the MCID in both PHQ-9 and GAD-7 by those with more than two therapy sessions in each age group was then calculated.

RESULTS

There were 82 513 treatment episodes recorded across the 13 IAPT services in the South West of England. The number of people in each age group in the study area is shown in Table 1 along with the estimated number of people with CMHPs and the numbers of people referred to IAPT services, attending at least one session with an IAPT practitioner; uptake of treatment by at least one further session; and achieving MCID. The numbers are also expressed as proportions of estimated prevalence, of referrals, and of those seen.

Estimated prevalence of CMHPs peaks in 45–49-year-olds (20.59%), with lowest estimated prevalence in 70–74-year-olds (9.47%). Referral rates as a proportion of CMHPs peak in 20–24-year-olds (22.95%) and then decrease from this point until 74 years of age (6.00%). Attendance rates as a proportion of referrals peak in 60–64-year-olds (76.97%) with lowest attendance rates in 20–24-year-olds (57.34%). Further detail can be found in Table 1.

The proportions of the population being referred, obtaining access, engaging with treatment, and achieving MCID each peak in 25–29-year-olds and decline thereafter, with lowest numbers for 70–74-year-olds. Figure 1 shows the contrast graphically. Figure 2 depicts a ‘vertically magnified’ version of Figure 1, showing the proportion of the population referred, obtaining access, continuing with treatment, and achieving MCID from the total population.

Open in a separate window

Figure 1.

Estimated proportion of CMHPs and number of those: referred, with access, with ≥2 sessions, achieving MCID as a proportion of total population across age in the South West of England. CMHP = common mental health problem. MCID = minimal clinical important difference.

Open in a separate window

Figure 2.

Proportion of those: referred, with access, with ≥2 sessions, achieving MCID of total population across age in the South West of England. MCID = minimal clinical important difference.

Access uptake, continued treatment, and achieving MCID, represented as a proportion of referrals, is depicted in Figure 3. Of those referred: proportions of people obtaining access increases until 64 years (76.97%); proportions of people with ≥2 treatment sessions increases until 59 years (35.31%); and proportions of people achieving MCID increases from 18 years until 69 years (15.86%).

Open in a separate window

Figure 3.

Number of those: with access, with ≥2 sessions, achieving MCID as a proportion of those referred across age in the South West of England. MCID = minimal clinical important difference.

The age group with the lowest proportion achieving MCID or reliable improvement is 70–74-year-olds (32.77%) but otherwise the percentage with MCID gradually increases from those aged 20–24 (37.08%) to those 65–69 years (46.17%), this age group having the highest proportion achieving MCID (Figure 4).

Open in a separate window

Figure 4.

Those achieving minimal clinical important difference as a proportion of those with ≥2 treatment sessions across age in the South West of England. MCID = minimal clinical important difference.

In summary, more referrals are made for younger adults with a peak age of 25–29 years as a proportion of the total population and a peak age of 20–24 years as a proportion of those with CMHPs, but, once referred, a higher proportion of older adults are engaging with and benefiting from treatment than are 20–29-year-olds.

DISCUSSION

Acknowledgments

Notes

REFERENCES