Cost of Improving Access to Psychological Therapies (iapt) programme: an analysis of cost of session, treatment and recovery in selected pcts in East of England region Muralikrishnan Radhakrishnan, PhD




Дата канвертавання21.04.2016
Памер128.91 Kb.
Cost of Improving Access to Psychological Therapies (IAPT) programme: an analysis of cost of session, treatment and recovery in selected PCTs in East of England region

Muralikrishnan Radhakrishnan, PhD

Health Economist

Cambridge and Peterborough CLAHRC

Institute of Public Health,

University of Cambridge,

Forvie Site, Robinson Way,

Cambridge CB2 0SR

E-Mail: mrk38@medschl.cam.ac.uk



Geoffrey Hammond, PhD

IAPT Evaluation Research Associate

Cambridge and Peterborough CLAHRC

18 Trumpington Road

Cambridge CB2 8AH

E-Mail: Geoffrey.C.Hammond@gmail.com



Louise Lafortune,PhD

Senior Research Associate

Cambridge and Peterborough CLAHRC

Institute of Public Health,

University of Cambridge,

Forvie Site, Robinson Way,

Cambridge CB2 0SR

E-Mail: ll394@medschl.cam.ac.uk



Public Heath Theme

NIHR CLAHRC for Cambridgeshire and Peterborough

Cambridge
JUNE 2011
Acknowledgements

Thanks are due to Prof.Carol Brayne, Caroline Yvonne Lee and Dr.Christine Hill, Cambridge and Peterborough CLAHRC, Institute of Public Health, University of Cambridge for their critical comments on the study. Muralikrishnan Radhakrishnan’s and Louise Lafortune’s research was supported by the Cambridge and Peterborough CLAHRC. Geoffrey Hammond’s research was supported by the Strategic Health Authority of the East of England. The views expressed in this report are those of the authors and may not reflect those of the funder.



Competing Interest

None declared



Ethical approval

The Cambridge and Peterborough CLAHRC waived the need for consent based on independent evaluation which stated that the data were free of patient identifiable information and that the research project proposed represented no deviation from normal clinical care and should be deemed service evaluation.



Disclaimer

Papers published from the Cambridge and Peterborough CLAHRC are intended as a contribution to



current research. Work and ideas reported may not always represent the final position and as such may sometimes need to be treated as work in progress. The material and views expressed in this report are solely those of the authors and should not be interpreted as representing the collective views of CLAHRC staff and their research funders.


CONTENTS










Abstract………………………………………………………………………………………………………………………

iv


Introduction……………………………………………………………………………………………………………….


1


Methods…………………………………………………………………………………………………………………….


3

Cost per Session……………………………………………………………………………………………………………..

5

Cost of a Treatment Course……………………………………………………………………………………………

6

Cost per Recovered Patient…………………………………………………………………………………………….

7

Sensitivity Analysis………………………………………………………………………………………………………….

7


Results………………………………………………………………………………………………………………………..


8

Cost per Session………………………………………………………………………………………………………….....

10

Cost of a Treatment Course……………………………………………………………………………………………

11

Cost per Recovered Patient…………………………………………………………………………………………….

12

Sensitivity Analysis………………………………………………………………………………………………………….

13


Discussion……………………………………………………………………………………………………………………


15

Key Findings…….………………………………………………………………………………………………………….....

15

Implications……..………………………………………………………………………………………………………….....

15

Limitations……….………………………………………………………………………………………………………….....

17

Conclusion……….………………………………………………………………………………………………………….....

17


References………………………………………………………………………………………………………………….


18


Appendix…………………………………………………………………………………………………………………….


20


ABSTRACT
Background

The Improving Access to Psychological Therapies (IAPT) programme is an important innovation in the provision of mental health services in the UK. It was originally designed to increase access to psychological therapies, thereby alleviating the burden and distress associated with anxiety and depression. Recent literature has reported on improvements in clinical outcomes, changes in employment status as a result of the programme, but not on the costs of the programme.


Aims

To estimate costs associated with a single session, completed course of treatment and recovery for four treatment types in IAPT services in selected Primary Care Trusts in the East of England region.



Methods

The analysis used clinical outcomes for patients from 5 PCTs who had completed/ended treatment between 2009 and 2010. The main measure of clinical outcome used was recovery at the end point of treatment. The outcome of treatment were assessed in terms of changes in symptomatic measures (severity of depression and anxiety symptoms) which were recorded using brief questionnaires(PHQ-9 and GAD -7) required by the IAPT routine clinical dataset. Data regarding financial & treatment activity were combined with data assessing clinical outcomes. Generalising a few costing assumptions, the cost of session, treatment course and recovery were estimated.


Results

Of the 10789 patients who ended treatment between 2009 and 2010, 8464 patients attended 2 or more sessions. Out of them, 4854 patients (57.3%) received only low intensity treatment, 2230 patients (26.2%) received only high intensity treatment. In addition, 252 patients (3.0%) had stepped down (from high to low intensity) treatment and 1128(13.3%) had stepped up (from low to high) intensity treatment. Overall, 3371 patients (39.8%) achieved recovery according to changes in symptomatic measures as recorded by the PHQ-9 and GAD-7. Across all PCTs, the average cost of a high intensity session was £176.97 and the average cost for a low intensity session was £98.59. The average cost of treatment was £493(low intensity), £1,416(high intensity), £699 (stepped down) and £1514 (stepped up). The cost per recovered patient was £1,043(low intensity), £ 2,895(high intensity), £1,653(stepped down) and £2,914(stepped up). Sensitivity analysis reveals that the cost per session, treatment and recovery is sensitive to the major assumption on cost ratio, indicating that inaccurate ratios are likely to influence the overall estimates.



Conclusions

Results indicate that costs currently exceed previous estimates. Replication and additional analyses along with evidence based discussion surrounding alternative, cost-effective methods of intervention is recommended. It is likely that improvements in current IAPT practice cannot occur until current practice is scrutinised and treatment approaches that are both effective and financially viable are identified, studied, and highlighted.



Introduction

Large, multinational epidemiological studies indicate that approximately 16% of the population experience depression and anxiety over a lifetime (Kessler et al., 2003; Singleton et al., 2001). In the UK, a 2006 report from Centre for Economic Performance (CEP) stated that “crippling depression and chronic anxiety are the biggest causes of misery in Britain today” (CEP,2006). The financial cost of depression in the UK was recently estimated at approximately 150 billion pounds in 2009/2010, of which 30 billion is thought to be work related (Sainsbury Centre,2010). For the UK, it has been estimated that a cross subsidy of £7-10 billion on social security benefits payments are made to cover the unemployment costs of people with high prevalence mental health problems (CEP,2006).


In order to alleviate the distress and costs associated with anxiety and depression, the UK Department of Health announced an unprecedented increase in funding for the provision of psychological therapies in the National Health Services (DoH, 2007). With the possible exception of the closure of asylums and an accompanying increase in community based mental health treatment in the 1960s, the Improving Access to Psychological Therapies (IAPT) programme represents a major policy directive in UK mental health services in the past 50 years. Developed on economic arguments and clinical evidence base, IAPT was developed to promote access to National Institute for Health and Clinical Excellence (NICE) - approved, Cognitive Behavioural Therapy (CBT) based talking therapies treatments. IAPT represents a public health approach to the treatment of mild to moderate depression and anxiety, common conditions judged to cause a preventable enormous burden of morbidity and disability in the UK and worldwide (Layard,2006; Richards & Suckling R, 2009). IAPT services have been commissioned throughout England, with more than 300 new therapists recruited for training in the East of England (EoE) alone between 2008 and 2011. By substantially increasing the number of therapists, IAPT is intended to facilitate increased referrals and reduced waiting times, with the potential to increase patient reported satisfaction and reduce self-reported depression and anxiety (CEP,2006; Layard,2006). The funding for the IAPT programme was contingent on the successful pilot implementation of a new treatment programme in two demonstration sites: Doncaster in Northern England and Newham in East London. The outcomes of these implementations would be used to argue that the increased funding from the Government could deliver increased clinical outcomes in terms of magnitude and volume to justify the investment (Richards & Suckling, 2009). A recent evaluation of both sites indicated that at least 55% of patients who attended at least two sessions (including an assessment interview) recovered and 5% transitioned from unemployment into part or full-time employment (Clark et al, 2009). This study generally agreed to demonstrate that the talking therapies implementation model can be effective in the treatment of depression.
CBT is recommended by the NICE as an appropriate evidence-based psychological intervention for depression and anxiety disorders (Clark et al, 2009). NICE also recommends a stepped-care approach to the delivery of psychological therapies in mild or moderate depression and anxiety. The basic fact of the stepped approach dictates that patients should receive treatment that is simultaneously the least restrictive treatment option and the most likely to provide significant health gains. In the talking therapies model, a two tier implementation is used: a high and a low intensity intervention is available. Clark et al (2009) states that in the light of evidence, some individual patients respond well to ‘low-intensity’ interventions (such as guided self-help and computerised CBT). In the case of severe depression and some anxiety disorders such as post-traumatic stress, NICE guidelines recommend that patients be provided ‘high-intensity’ face to face psychological therapy. Broadly, these two types of therapies are provided by two types of therapists: ‘low-intensity’ and ‘high-intensity’.

The rationale for widespread implementation of the IAPT programme, besides greatly decreasing the problem of mental illness in UK, was that economic gains associated with increased productivity and reemployment of those individuals treated would dwarf all costs associated with the programme (Layard et al 2007). Layard et al (2007) state in their paper on economic costs that providing psychological therapy to people not now in treatment would result in “the cost to the government to be fully covered by the savings in incapacity benefits and extra taxes that result from more people being able to work”. Layard et al (2007) also argued that these stepped care, expanded psychological therapies programmes would cover the cost to the government by the extra output in GDP produced by the treated person, savings to the exchequer in incapacity benefits and extra taxes as a result of more people being able to work. The extra GDP produced by treating one person was estimated at £1,100, the estimated extra earnings from 1.1 additional months of work over a two year period. The savings to the exchequer in terms of incapacity benefits and extra taxes generated is estimated to be around £ 9,000 per year or £18,000 for a two year period. Further, they also estimated a conservative amount of £300 as a savings to the NHS per treated person over a two year period. This gives a total benefit of £19,400 to society over a two year period for a person treated. Adjusting for inflation rates from 2007 figures to 2010 year figures, the total estimated benefit to the society per treated person is £20,989 or approximately £21,000 for two years (£10,500 per year).


The cost benefit calculation described in the paper assumed that the cost of providing a standard course of roughly ten meetings of CBT is £750 or £75 per session. These unit costs were obtained from a recognised national source (Curtis & Netten, 2006). Curtin & Netten (2006) provides a session cost of £66, and NICE Guidelines for Post-traumatic stress disorder (PTSD) provide a session cost of £82. Layard et al (2007) averaged these estimates to arrive at £75 per session (Layard et al 2007). However, this estimate has yet to be tested empirically. Given the large disparity between the estimates, there is a pressing need to estimate the costs using data from psychological therapies programmes to assess what the cost really is.
The recent evaluation of the two pilot sites by Clark et al (2009) reported on clinical outcomes and improvements in employment status but did not report on the associated costs of the programme. The cost of the programme is an important consideration for psychological therapy implementation. Given that these sessions are provided by high and low intensity therapists who differ significantly in their training and salary costs, there could be a significant differences in the costs of the sessions delivered by each group.
Unpublished data from the Bedfordshire Primary Care Trust (PCT) in the East of England region reveals that ‘high intensity activity’ costs 1.8 times more than ‘low intensity activity’. Salaries constituted at least 3/4 of the overall expenditure for high intensity and low intensity services, indicating that IAPT sessions are human resource intensive. Given differences in the salaries between high and low intensity employees, it becomes important to estimate specifically the cost per single high intensity and low intensity session. The cost of a typical course of completed treatment is also an important facet of cost to consider. Given that Clark et al (2009) reported the percentage of patients who recovered, it is key to estimate the cost of a recovered patient through the IAPT programme and its activities.
This study is an attempt to estimate the cost of session, completed treatment and recovered patient using financial data from 5 PCTs in the East of England region. The study uses clinical outcomes, financial information, and a few explicit assumptions to arrive at the costs. The methods, results, limitations and implications for policy are discussed in the following sections.

Methods

This analysis is an attempt to cost low intensity and high intensity session, treatment course and recovery provided by the IAPT programme in the East of England region. Broadly, two types of therapies are provided (low and high intensity) by the IAPT programme. However, preliminary data analysis revealed that a number of patients also ‘stepped down’ (i.e. transitioned from high to low intensity) and ‘stepped up’ (, i.e. transitioned from low to high intensity) between the low and high intensity sessions. Costs associated with these four types of treatment course and transitions were estimated.

Costing was done from the IAPT programme perspective. The costs calculated were based solely on costs associated with service budgets of IAPT programmes in the EoE; they did not include costs incurred from a wider perspective (patient and other societal costs). The population of interest was adult patients who were above 18 years and received treatment through the IAPT programme provided by five PCTs in the region. For the analysis, financial data was combined with clinical outcomes for 8,464 patients who had attended more than 2 treatment sessions in the programme between 1 April 2009 to 31 March 2010. These patients include those who completed treatment, deceased during the treatment, declined to continue with treatment, dropped out of treatment, or were found unsuitable for treatment by IAPT therapists. Data on patient’s clinical progress and outcome during the financial year specified above were collected on a sessional basis by high and low intensity clinicians as part of routine clinical assessment. This information was stored on the clinical record keeping software, PC-MIS, for all of the areas assessed. The primary clinical outcome assessed was recovery at the end point of treatment. Secondary clinical outcomes were assessed based on changes in symptom severity (severity of depression and anxiety symptoms) as recorded using two brief questionnaires required for completion of the IAPT routine clinical dataset (DoH,2008). These were the Patient Health Questionnaire Depression scale (PHQ-9) and the Generalised Anxiety Disorder scale (GAD-7) (Appendix 1).

The Patient Health Questionnaire Depression scale (PHQ-9) has undergone validation in both the UK and the US (Kroenke et al., 2001; Cameron et al., 2008). Nine questions enquire about a symptom (or set of symptoms) of depression and are scored from 0 (“Not at all bothered by the problem”) to 3 (“Bothered nearly every day”). Sum scores range from 0 to 27. A score of 10 or above has been identified as a potential clinical cut point for diagnosis of depression (Lowe et al., 2004); severity bands are defined as follows: 0-4 not depressed, 5-9 mild depression, 10-14 moderate depression, 15-19 moderate/severe depression,and 20-27 severe.


The Generalised Anxiety Disorder scale (GAD-7) is a seven item measure of the severity of anxiety symptoms (Spitzer et al., 2006). The measure uses the same response options and item scores as the PHQ-9. Sum scores range from 0 to 21. A score of 8 or higher on the GAD-7 has been suggested as a threshold for determining those with a probable clinical diagnosis of anxiety (Spitzer et al., 2006). Severity bands are defined as follows: 0-4 not anxious, 5-9 mild anxiety, 10-14 moderate anxiety, and 15-21 severe anxiety.
For the current analysis, those patients who fulfilled criteria for “caseness” at their initial session and transitioned into “non caseness” at their final session were considered recovered, irrespective of whether they completed treatment or had any other treatment outcome (e.g. declined treatment, found unsuitable for further treatment) (Figure 1). To be classified as cases at baseline, these individuals required to score at an initial assessment session, a clinical cut-off for the PHQ–9 of 10 points or more and the GAD-7 of 8 points or more. Individuals were classified as being cases if their PHQ-9 or GAD-7 scores were above the stated thresholds and were required to be below this clinical cut-off on both measures at the final treatment session to be classified as recovered. Recovery rates were computed for those individuals who have ended IAPT treatment by dividing the number of recovered cases by the number of cases treated.
Figure 1 - Categorization of patients as recovered using PHQ9 and GAD7 Scores


flowchart: process 4

straight arrow connector 17

flowchart: process 6

straight arrow connector 23

flowchart: decision 5

flowchart: alternate process 30
NO


straight arrow connector 25

straight arrow connector 1030
YES


flowchart: process 39

straight arrow connector 1032


flowchart: process 43

straight arrow connector 45

flowchart: decision 47

flowchart: alternate process 49

NO

straight arrow connector 53



straight arrow connector 48

YES



autoshape 22

Financial information on the total spend on IAPT programmes from five PCTs in the East of England region were sourced for the 2009-2010 financial year (1 April 2009 to 31 March 2010) from reports prepared by Mental Health Strategies for the Department of Health (Mental Health Strategies 2010)1. The PCTs included in the study were Bedfordshire, Cambridgeshire, North & East Hertfordshire, Suffolk and West Hertfordshire. PCTs selected for inclusion had complete data on IAPT sessions and clinical outcomes for the entire period assessed along with available information on their service budget over the same period. Except for North & East Hertfordshire PCT (which started IAPT operations in April 2009), all PCTs started their IAPT operations in September 2008.


Initially, the cost per session (high and low) was established, after which the cost of a completed treatment course and cost per recovered patient for the four types of treatment course/transition was estimated. The following subsections explain the methods and assumptions used to estimate the different costs.

Cost Per Session

Financial information for each of the PCTs provided the overall total spend for the financial year: the proportion of the total amount spent on low and high intensity sessions was not specified. Therefore, a framework was developed in order to allocate the expenditure to either high or low intensity activity (Figure 2).


The first step involved apportioning the total IAPT expenditure in a given year to the volume of high or low intensity interventions delivered over the same period. Unpublished data from Bedfordshire PCT revealed that ‘high intensity activity’ costs 1.8 times more than ‘low intensity activity’. A local tariff developed for IAPT services in the EoE region also confirmed that the cost ratio was closer to this estimate (Nolan, 2009). This cost ratio was stable even if the cost items, salaries and overheads, were considered individually. Hence, it was assumed that high intensity sessions would cost 1.8 times more than the low intensity sessions in all of the reviewed PCTs. In order to arrive at the total cost of high and low intensity sessions, the total session ratio was multiplied by the assumed cost ratios to arrive at the percentage allocation of IAPT spend. The total spend was apportioned according to the percentage allocation to arrive at the total costs for ‘high intensity’ and ‘low intensity’ sessions. In order to estimate the per session costs, the total costs were then divided by the total number of high and low intensity sessions delivered during the financial year specified above.
Average duration for high intensity and low intensity sessions varied across PCTs. The cost per session was also estimated using mean duration of sessions, to confirm that it did not vary from the original estimates of cost per session. The mean duration was multiplied with the number of sessions to arrive at the total duration for high and low intensity sessions. The respective total cost was then divided by this total duration to arrive at the cost per minute for high and low intensity sessions. The cost per minute was then multiplied with the respective mean duration to arrive at the cost per session.

Figure 2 – Framework for estimating cost of session, treatment and recovery




Cost of a Treatment Course

Using the per session cost estimate, the cost for a single course of treatment delivered by different IAPT providers was estimated (Figure 2). For this, all patients from the 5 PCTs who finished treatment regardless of the reason for ending treatment (i.e. those who completed treatment, deceased, declined, dropped out or were found not suitable for IAPT ) between 1 April 2009 to 31 March 2010 were included in the analysis. Each patient’s entire treatment course was traced back to their initial session to ascertain the number of sessions they received, including the initial assessment session. The main group of interest were those who completed treatment, since their cost represents the cost of successfully completed course of prescribed treatment.


The median number of sessions received during the course of treatment was assessed separately for four groups of patients: those who remained in high intensity, those who remained in low intensity, those who transitioned from low to high intensity (stepped up) and those who transitioned from high to low intensity (stepped down). The median rather than the mean was used in further calculations due to the fact that sessional data was not normally distributed. The median number of sessions was then multiplied by the session cost associated with the corresponding PCT to produce the cost of a single course of treatment.
The cost of a single course of treatment was calculated primarily for those patients who successfully completed the prescribed treatment sessions. The median number of sessions along with their interquartile range (25th and 75th percentile) was also estimated to provide a range for the costs. The cost of treating patients in the other categories like deceased, declined, dropped out and not suitable were calculated using the method outlined above. In addition, the cost of providing the single session (usually assessment) to those who attended only one session was also estimated by multiplying the respective session cost with the number of patients. These patients were assessed either by a low intensity or high intensity therapist. In order to apportion these costs between the four treatment types, they were allocated on the basis of total treated patient proportions. These costs were calculated in order to estimate the total costs of treatment and to arrive at the cost per recovered patient, as described in the following section.
Cost per Recovered Patient

To estimate the cost per recovered patient, two estimates were required: the total cost of treating all the patients and the number of patients who recovered, in all categories (completed deceased, declined, dropped out and not suitable). The total cost of treatment was calculated by multiplying the cost of treatment course with the number of patients who underwent treatment. The total cost was then divided by the number of patients who recovered to arrive at the cost per recovered patient for each PCT for the four different treatment courses (Figure 2).


Sensitivity Analysis

For a base case estimate, it is assumed that that high intensity sessions would cost 1.8 times more than the low intensity sessions. To check for robustness of results, the cost ratio was varied, using low to high ratio estimates of 1.6 and 2 respectively. The main intention of the sensitivity analysis is to report the extent to which the cost per session, treatment and recovery changes if the cost ratios are varied.



Results
In total, 10,789 patients completed or ended IAPT treatment for various reasons (Figure 3) between 1 April 2009 - 31 March 2010, of whom 21.2% attended only 1 session. Among those who attended 2 or more sessions, 4844 (44.9%) successfully completed the allotted treatment. It is to be noted that a significant number of patients dropped out of treatment (n=1961; 18.2%) and 861 (8%) were found unsuitable for IAPT treatment after attending 2 or more sessions.
Figure 3 – Distribution of Patients who ended IAPT treatment (2009-2010)

The baseline demographic and clinical characteristics of the 8464 patients who attended 2 or more sessions are presented in Table 1. 4325 patients (51.1%) belonged to the age group 26 -45 years; the majority of patients (5576, 65.9%) were women. In full time or part time employment at baseline assessment were 5155 patients (60.9%). Referrals to the IAPT programme were through GPs for 7511 patients (88.7%). Patients from Suffolk PCT constituted a majority of the sample (31.9%), followed by Cambridgeshire PCT (20.4%), Bedfordshire PCT (19.3%), West Hertfordshire PCT (17.6%) and North & East Hertfordshire PCT (10.8%). Based on baseline PHQ-9 scores, 2146 patients (25.4%) were classified as having moderate depressive symptoms, 1987 patients (23.5%) with moderate-severe depressive symptoms and 1787 patients (21.1%) with severe depressive symptoms. On the GAD 7 scores at baseline, 2365 patients (27.9%) were classified with moderate anxietal symptoms and 3348 patients (39.6%) with severe anxietal symptoms.

Table 1 - Baseline Demographics and Clinical Characteristics of Patients

Table 2 presents the number of patients who were treated and who recovered using the clinical definition of recovery in each PCT (PHQ9 Score < 9 and GAD7 score < 8 at the final/last session). The corresponding recovery rate is also computed. Of the 8464 patients treated, 4854 patients (57.3%) received only low intensity treatment, 2230 patients (26.2%) received only high intensity treatment, 252 patients (3.0%) were stepped down in their treatment and 1128 (13.3%) were stepped up. Of the 3371 patients (39.8%) who achieved recovery, 1992 patients (59.1%) received low intensity treatment exclusively, 884 patients (26.2%) received high intensity treatment exclusively, 101 patients (3.0%) were stepped down and 394 patients (11.7%) were stepped up. Across all PCTs, a recovery rate of 55.7% was observed by individuals completing treatment after receiving exclusively low intensity therapy. The recovery rate was 54.4% among those who completed high intensity treatment, 54.9% among those who completed treatment but stepped down and 52.9% among those who completed treatment and stepped up. Recovery rate were in excess of around 20% among those who dropped treatment among all the therapy types. For those who declined treatment, the recovery rate was in the range of 14% to 25% among the four treatment types. For those found not suitable for treatment, the recovery rate was less than 10% for the low intensity, high intensity and stepped up treatments and 17.2% for the stepped down treatment.


Table 2 – Distribution of Patients Treated, Recovered and Recovery Rates
Cost per Session

Table 3 provides the total IAPT spend in each of the five PCTs assessed, the total session ratio (proportion of low and high intensity sessions), and the proportion of the total cost allocated to high and low intensity activity for the 2009-2010 financial year. All the PCTs spent in excess of £ 2 million on the IAPT programme, with Suffolk PCT spending the most (approximately £ 4 million).


In terms of the session ratio, Cambridgeshire PCT provided the highest proportion of low intensity sessions (59.3%) whereas North & East Hertfordshire PCT provided the lowest proportion (46.9%). For the high intensity sessions ratio, Suffolk PCT had the highest (54.9%) and Cambridgeshire PCT had the lowest (40.7%). In terms of the actual number of sessions, Suffolk PCT delivered the most, with a total of 35,936 sessions and North & East Hertfordshire PCT delivered the fewest (11,780 sessions). The low number of sessions at North & East Hertfordshire PCT is explained by the fact that the provider started providing services in April 2009, whereas all the other PCTs started their operations in September 2008.
High intensity sessions account for more than 60% of total costs for all of the PCTs studies except Cambridgeshire, where it accounts for 55%. As expected, this translates into significantly higher total spent on high intensity activity. The significantly higher total spent for Suffolk is because they performed the highest number of individual sessions.
Across all PCTs, the average cost for a low intensity session was £98.59 and it ranged from £78.31 (Suffolk PCT) to £150.17 (North & East Hertfordshire PCT). The average cost of a high intensity session was £176.97, and ranged from £140 (Suffolk PCT) to £270.41 (North & East Hertfordshire PCT) across PCTs. The overall cost per session was £ 137.73 for all PCTs, and ranged from £112.70(Suffolk PCT) to £ 214.01 (North and East Hertfordshire PCT).

Table 3 – Total Spend, Sessions, Cost and Cost per Session of High and Low Intensity Activity (2009 -2010)




The cost per session for high intensity and low intensity was also estimated using the mean durations (Table 4) and the results show that the estimated cost per session is similar to previous estimates.
Table 4 – Estimated Cost per Session using Mean Session Duration

Total Duration = No.of Sessions (Table 3) X Mean Duration

Cost per Minute = Total Costs (Table 3) / Total Duration

Cost per Session = Cost per Minute X Mean Duration
Cost of a Course of Treatment

The median number of sessions in a given course of treatment for each treatment type (high or low, stepped up or down) is given in Table 4. For all the PCTs combined, the median number of sessions received was 5 for individuals remaining on low intensity treatment and 8 for individuals remaining on high intensity treatment. People who stepped down from high to low intensity sessions received a median of 3.5 sessions of low intensity interventions and 2 sessions high intensity sessions. For the groups who initiated treatment with low intensity intervention and were stepped up, they received approximately 1 low intensity session and 8 high intensity sessions, suggesting that most individuals were stepped up after initial assessment. However, substantial variation in the median number of sessions across PCTs was observed. For instance, Bedfordshire PCT and Cambridgeshire PCT provided 10 sessions for high intensity treatment compared to other PCTs who provided in the range of 6-8 sessions.


Table 4 presents the cost of treatment for the four different treatment types across PCTs. For all PCTs, the average estimated cost for a course of exclusively low intensity treatment was £493 and fluctuated between £313 (Suffolk PCT) and £901 (North & East Hertfordshire PCT). The estimated cost of a single course of high intensity treatment was £1,416 and ranged from £987 (Suffolk PCT) to £1,793 (West Hertfordshire PCT). The average cost of a treatment course for individuals who were stepped down was £699 and ranged from £579 (Suffolk PCT) to £946 (West Hertfordshire PCT). The average cost of a treatment course for individuals who were stepped up was £1514 and ranged from £1206 (Suffolk PCT) to £1891 (Bedfordshire PCT). Stepping up and down treatments cost more than low intensity and high intensity treatments.

Table 4 – Median Number of sessions and Cost of Treatment




The cost of treatment for those in the deceased, declined, dropped out and not suitable categories were calculated using similar methods and are presented in Appendix 1.
Cost per Recovered Patient

The total cost of treating those who completed treatment are calculated by multiplying the number of patients treated (Table 2) with the respective cost of a single course of treatment (Table 4). The cost of treating patients from the deceased, declined, dropped, not suitable categories and assessment (those who had only one session) was also estimated using a similar method. The overall total cost of treating all categories of patients under different therapy types are given in Table 5. Irrespective of therapy received, those who completed treatment constituted the majority of the total IAPT expenditure (over 70%). However, fluctuation was observed across PCTs. At £2,559,111, high intensity therapy is the highest single total cost for across all treatment types for all PCTs. This pattern of high intensity treatment being associated with the highest cost is observed for all individual PCTs except Cambridgeshire PCT, where low intensity therapy ranks highest among the total costs. This is expected since Cambridgeshire PCT provided the highest proportion of low intensity sessions (59.3%) In terms of individual PCT cost, West Hertfordshire PCT has the highest cost of £504,533 for low intensity therapy, and £674,970 for high intensity therapy. North & East Hertfordshire PCT has the highest cost for stepped down treatment (£55,115) while Suffolk PCT has the highest cost for stepped up treatment (£456,425).


From the number of patients who recovered (Table 2) and the total costs (Table 5), the cost per recovered patient is calculated and is presented in Table 6. For the low intensity therapy, the estimated cost per recovered patient is £1,043 for all PCTs (ranging from £687 (Suffolk PCT) to £1,952 (North & East Hertfordshire PCT)). For the high intensity therapy, the cost per recovered patient is £ 2,895 and varied significantly across PCTs (from £1,849 (Suffolk PCT) to £4,066 (West Hertfordshire PCT)). For the stepped down therapy, the estimated cost per recovered patient is £1,653 and ranges from £1,012 (Suffolk PCT) to £2,111 (West Hertfordshire PCT). For the stepped up therapy, the cost per recovered patient is £2,914 and ranged from £2,341 (Suffolk PCT) to £ 3,629 (Cambridgeshire PCT). Stepping up therapy yields the highest cost per recovered patient, followed by remaining in high intensity, stepping down and remaining in low intensity treatment. Overall, including all therapy types, the cost per recovered patient is £1,766 and varies from £1,301 (Suffolk PCT) to £2,773 (North & East Hertfordshire PCT).

Table 5 – PCT wise Total Cost of Treatment




Table 6 – Cost per Recovered Patient


Sensitivity Analysis

An important assumption used to estimate the costs was that high intensity sessions cost 1.8 times more than the low intensity sessions. To check for the sensitivity of the cost ratio, it was varied between a lower costing ratio of 1.6 and higher one 2.0. The results are presented in Table 7. As the cost ratio is reduced to 1.6 from the base case of 1.8 the cost per session, treatment and recovery for low intensity session increases between 7.1 - 7.6% and decreases for the high intensity session by 4.1 - 4.3%. As the cost ratio is increased to 2.0 times from the base case, the reverse is observed - the cost per session, treatment and recovery for a high intensity session increases by 3.7% and decreases for a low intensity session by around 6.5%. A similar pattern is observed with stepping patterns, although the fluctuation in costing as a result of changes in the ratio represents 3.5% or less of the original estimated cost in any scenario. The sensitivity analysis reveals that the cost per session, treatment and recovered patients is sensitive to this assumption, indicating that inaccurate ratios are likely to influence the overall estimates of session, and treatment course costs.


Table 7 - Sensitivity Analysis of Cost Ratio (All PCTs)





Discussion
Key Findings

This analysis is, to the knowledge of the authors, one of the first costing estimates associated with implementation of psychologically based talking therapies in the UK. Reported information concerning public sector spending on psychologically based talk therapies were combined with clinician reported estimates of session level activity to arrive at an estimate of session costs. These session costs were then expanded to assess costs associated with a treatment course under a number of different treatment conditions (remaining in low or high intensity treatment, stepping up or down) and treatment outcomes. Although the IAPT service has been found to be highly effective in reducing symptoms of anxiety and depression, the extent to which it is cost-effective in roll out of a whole programme has not been demonstrated or discussed in the current evidence base. Costing up treatment under these conditions is important step to assessing the cost-effectiveness of the implementations along with an important consideration in future service design and planning.


Layard et al (2007) estimated that the cost of providing a standard course of roughly ten meetings of CBT is £750 or £75 per session. However, these costing were based on an average session cost of treatment in community mental health teams and treatment in a specialist PTSD clinic, neither of which implement the talking therapies organisation model (i.e. stepped care and two tier workforce). Additional government estimates of cost based on a capacity model of activity (assuming a maximum number of treatment interventions based on number of hours, days in a working year) have estimated the high and low intensity session costs to be approximately £87 and £27, respectively, with an average cost of 58 pounds.
Given the results above (low intensity cost of approximately £100; high intensity session cost approximately £175), 75 or 58 pounds per session would seem to be a significant underestimation of the cost of treatment as currently provided. Although the costing estimates provided here likely incorporate significant start-up costs and are also influenced by the fact that the majority of low intensity trainees are relatively inexperienced, results would seem to suggest that a 2-3 fold reduction in cost/increase in activity is likely required in order to bring psychologically based talking therapies into the financial envelope originally proposed.
Among those who completed treatment, there was a recovery rate of more than 50% in all types of therapy, which is in line with what Clark et al (2009) report. This rate is also similar to what was expected when the IAPT programme rolled out (Layard et al 2007). However the natural rate of recovery is 30%, making the recovery rate that can be directly attributable to treatment only 20%. Although the effectiveness of the interventions are likely to increase as workers become more experienced, even with the cost figures estimated in this study, the IAPT programme will be cost beneficial based on an estimate of a benefit to the society per treated person of £10500 per year (as estimated by Layard et al (2007)).
Implications

Additional analyses performed by the authors would indicate that there are a number of ways to reduce current costs with psychologically based talk therapies clinics. Preliminary analysis of IAPT activity on June 2010 would suggest that the average number of patients seen by a high intensity therapist was approximately three while for a low intensity therapist, it was approximately five individuals. Significant fluctuation in the level of daily activity, and median number of sessions provided was also observed across PCTs. Interestingly, longer median sessions were not necessarily associated with better treatment outcomes. As such, an increase in daily activity would be one of the most readily observable methods to decrease sessional costs, although this would have to be done without a corresponding drop in treatment effectiveness. In addition, standardised treatment packages of a set duration agreed upon by the clinician and patient at the outset of treatment may be an effective method to standardise treatment duration and outcomes.


Increases in treatment volume may also be realised by changes in the way treatments are delivered. The use of technology mediated treatment, in particular telephone mediated interventions, might be instrumental to bring down the per session cost and thereby the cost of treatment and recovery. Psychological interventions delivered by telephone are more convenient and, in a growing number of situations, have been found to be as or more effective at reducing symptoms in patients treated with CBT based interventions. In a small scale study of telephone treatment for patients with obsessive compulsive disorder, a 40% reduction in clinician treatment time for telephone therapy is being reported (Lovell et al 2006, Robinson et al 1990). Recent analysis of current IAPT data in this cohort has also found equivalent reduction in anxiety, depression resulting from face-to-face and telephone mediated treatment in propensity score matched patient cohorts (in press). If suitable and accepted by patients and clinicians, telephone mediated therapy may allow for an increase in productivity while simultaneously reducing the need for additional treatment spaces and travel, two of the largest non-salaried costs reported in IAPT. With telephone therapy, service organisations and clinicians may be no longer constrained by conventional working hours and treatment facilities, conferring some flexibility in working hours. Literature also suggests that telephone-based delivery may also have the potential to remove patient perceived barriers to initiating treatment (Mohr et al 2006).
Other modalities of therapy also have the potential to increase the volume of patients treated, although additional research is required as to their efficacy. Online CBT delivered by a therapist in real time was found to be cost-effective compared with usual care in the treatment of depression in primary care in UK if society is willing to pay at least £20 000 per Quality Adjusted Life Year and could be a useful alternative to face-to-face CBT (Hollinghurst et al 2010). Group psycho-educational workshop interventions have been mentioned in NICE guidelines within the context of low intensity treatments and may represent a valid way to reduce cost for low intensity therapists. However, more research is required to investigate if group based interventions are effective and cost effective compared to face to face therapy.
Stepping up and stepping down was associated with increased costs relative to remaining in high or low intensity therapies, indicating a substantial penalty in patient misallocation to the wrong treatment. Further investigation into predictors of response in high and low intensity and into predictors of stepping is required in order to inform clinicians concerning likely patient response to treatment. Similarly, identification of distinct patient typologies (known as a patient casemix) and their observed response to treatment should provide clinicians with more information with which to correctly allocate individuals into high or low intensity treatments.
Among the total costs for treatment, a significant amount of at least 22% of the costs goes for the treatment of those who decline, drop out or are found not suitable for treatment. Some of these costs could be reduced through reviewing the current system of working. Following up of those who decline and drop out could be facilitated to convert them to complete treatment. Similarly, it takes a median of more than 2 sessions to identify those patients who are found not suitable for treatment. If they could be identified during the initial assessment session, this has the potential to reduce overall operating costs.
The cost per recovered patients in this study should not be used as a basis for an economic evaluation, since the type, distribution, and baseline severity of patients is likely to vary greatly across regions and cannot be directly compared. If a proper economic evaluation is to be performed, one of the first steps will be to derive a nationally representative casemix sample which comprehensively characterises the types of patients presenting to talking therapies services. Reduction in anxiety and depression (PHQ-9 and GAD-7) in each casemix can then be directly compared along with the resulting clinician time, and provide the ability to identify standardised, cost-effective examples of effective, good clinical practice.
Limitations

One limitation of this study was that a pivotal assumption in the costing approach adopted for all of the data was based on observed data in a single PCT. Specifically, the observed cost ratios ratio of 1.8 to 1 for high to low intensity costs was observed in itemised costing for the Bedfordshire budget and applied to the costing data for all other PCTs. The sensitivity analysis reveals that the cost per session, treatment and recovered patients is sensitive to this assumption, indicating that inaccurate ratios are likely to influence the overall estimates of session, treatment course costs. However, additional information provided from other PCTs would indicate that this ratio is representative of the difference in staffing costs. Average salary information submitted from four PCTs (unpublished data) showed that the ratio of low to high intensity average salary was between 1.6 to 1.8, similar to the costing ratio used in the analysis. Given that the IAPT therapies are highly human resources intensive and that the vast majority of the workers are salaried (and that these salaries form the majority of the total spend on high and low intensity services), it is expected that the ratio used is a reliable approximation of the true difference in overall costs. A similar cost ratio was observed in the overhead costs of high intensity and low intensity activity in the data from Bedfordshire PCT, indicating that overhead cost ratios were unlikely to influence the overall ratios.


The development of a more accurate micro costing approach is advocated in order to confirm or improve upon the estimates used in this analysis. This remains an area of important future research. In addition, the costs assessed here are based on the overheads of treatment. A broader perspective incorporating the overall societal costs may also be prudent considering potentially significant further costs for patients in terms of medication, travel for therapy and income loss due to the treatment course.
Conclusion

This is one of the first assessments of costs of talking based psychological therapies in the UK. Results indicate that costs currently exceed previous estimates, although a number of caveats exist. We invite replication and additional analyses along with evidence based discussion surrounding alternative, cost-effective methods of intervention. It is likely that improvements in current IAPT practice cannot occur until current practice is scrutinised and treatment approaches that are both effective and financially viable are identified, studied, and highlighted.




References:
Cameron IM, Crawford JR, Lawton K, Reid IC,(2008), Psychometric comparison of PHQ-9 and HADS for measuring depression severity in primary care. British Journal of General Practise; 58: 32-6.
Centre for Economic Performance (2006), The depression report: A new deal for depression and anxiety disorders. London: London School of Economics and Political Science Centre for Economic Performance
Clark D.M., Layard R, Smithies R, Richards D.A, Suckling R, Wright B, (2009), Improving access to psychological therapy: initial evaluation of two UK demonstration sites, Behaviour Research and Therapy; 47: 910–920.
Curtis L, & Netten A, (2006), Unit Costs of Health and Social Care. University of Kent, Canterbury: Personal Social Services Research Unit.
Department of Health, (2007),Press release: Johnson announces £170 million boost to mental health therapies. Wednesday 10 October 2007 12:51. GNN ref 152603P.
Department of Health (2008), Improving Access to Psychological Therapies (IAPT) Commissioning Toolkit, Department of Health : Crown
Hollinghurst S, Peters TJ, Kaur S, Wiles N, Glyn Lewis, Kessler D, (2010), Cost-effectiveness of therapist-delivered online cognitive–behavioural therapy for depression: randomised controlled trial. The British Journal of Psychiatry; 197 : 297-304
Kessler RC, Berglund P, Demler O, Jin R, Koretz D, Merikangas KR, (2003), The epidemiology of major depressive disorder: Results from the National comorbidity survey replication (NCS-R), Journal of the American Medical Association; 289:3095–3105.
Kroenke K, Spitzer RL, and Williams JB,(2001), The PHQ-9: validity of a brief depression severity measure. Journal of general Internal Medicine; 16:606-13.
Layard R, (2006), The case for psychological treatment centres. British Medical Journal; 7548: 1030-2.
Layard R, Clark D, Knapp M, Mayraz G, (2007), Cost benefit analysis of psychological therapy. National Institute Economic Review; 202: 90 -8
Lovell K, Cox D, Haddock G, Jones C, Raines D, Garvey R, Roberts C, Hadley S,(2006), Telephone administered cognitive behaviour therapy for treatment of obsessive compulsive disorder: randomised controlled non-inferiority trial. British Medical Journal; 333(7574): 883-5
Lowe B, Kroenke K, Herzog W, Grafe K,(2004), Measuring depression outcome with a brief self-report instrument: sensitivity to change of the Patient Health Questionnaire (PHQ-9). Journal of Affective Disorders; 81: 61-66.
Mental Health Strategies (2010) 2009/10 national survey of investment in adult mental health services. Report prepared for the Department of Health, available at http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_117488
Mohr DC, Hart SL, Howard I, Julian L, Vella L, Catledge C, Feldman MD,(2006), Barriers to psychotherapy among depressed and nondepressed primary care patients. Annals of Behavioural Medicine; 32(3): 254-58.
Nolan G, (2009), Developing a Local Tariff for IAPT Services in NHS East of England. Cambridge: NHS East of England
Richards D, & Suckling R,(2009), Improving access to psychological therapies: Phase IV prospective cohort study. British Journal of Clinical Psychology; 48: 377-396.
Robinson LA, Berman JS, Neimeyer RA,(1990), Psychotherapy for the treatment of depression: a comprehensive review of controlled outcome research. Psychological Bulletin; 108: 30-49.
Sainsbury Centre, (2010), The Economic and Social Costs of Mental Health Problems in 2009/10. London: Sainsbury Centre for Mental Health.
Singleton N, Bumpstead R, O'Brien M, Lee A, Meltzer H,(2001), Psychiatric morbidity among adults living in private households, 2000. London: The Stationary Office.
Spitzer RL, Kroenke K, Williams JB, Lowe B, (2006), A brief measure for assessing generalized anxiety disorder: the GAD-7. Archives of Internal Medicine; 166: 1092-7.

APPENDIX 1 - Patient Health Questionnaire Depression scale (PHQ-9) and the Generalised Anxiety Disorder scale (GAD-7)





APPENDIX 2 - Median number of sessions and cost of treatment of deceased, declined, dropped and not suitable.





1 According to Mental Health Strategies (2010) report, the IAPT spend for the year 2009 -2010 includes both the SHA contribution and matching funds by the PCT, except in the case of Cambridgeshire PCT, where only the SHA contribution is provided. Personal communications with IAPT managers at Cambridgeshire PCT revealed that an additional £810,000 has been spent as matched funds for IAPT.



База данных защищена авторским правом ©shkola.of.by 2016
звярнуцца да адміністрацыі

    Галоўная старонка