The impact of introducing Any Qualified Providers on hospital performance in England

The primary aim of the project is to examine the introduction of the Any Qualified Provider (AQP) scheme in England, on the behaviour of NHS and non-NHS providers with respect to volumes and quality of elective services which will be measured by patients' waiting time. Any qualified provider AQP) means that when patients are referred, usually by their GP, for a particular service, they should be able to choose from a list of qualified providers who meet NHS service quality requirements, prices and normal contractual obligations.

Senior members of the Health Foundation regularly meet with senior representatives from across government, where the Health Foundation’s views are sought on health policy and practice. The findings from this study will be communicated directly through this channel.

The outcomes from the study will provide Department of Health, NHS England and NHS Improvement with the evidence and hard facts on the effectiveness of the introduction of the AQP Programme on the performance of healthcare providers as well as the outcomes on patients.

The Health Foundation believes it has a legitimate interest for acquiring patient data for the purposes of this project.

Legitimate interest assessment is detailed below:

Purpose Test: The Health Foundation is a charity established to support the improvement of healthcare, including healthcare delivery. If analyses produced using patient data indicate how improvements can be made to how health care is delivered, then ultimately patients will receive better health care. The benefits of delivering better health care are essential for ensuring that patients receive good quality and timely diagnoses and treatment, this is critical for managing the health service. If The Health Foundation were not able to use the data requested, a robust evaluation of the AQP scheme could not be undertaken and therefore the public would not be able to benefit from wider implementation of this scheme. Therefore, the purpose of this project, which is to assess the effectiveness of the new AQP scheme aimed to improve NHS delivery, fits within this legitimate interest.

Necessity Test: The Health Foundation believes it is necessary to process data relating to individual patients (analysis) for the purposes of evaluating the effectiveness of the AQP scheme. The Health Foundation considers that without these data, it could not undertake a robust evaluation. The methods required to undertake the evaluation necessarily requires the patient data requested.

Balancing Test: Note that The Health Foundation will not have access to any identifiable patient data, and so staff at The Health Foundation will not know of the identity of any of the patients whose data are being analysed. This project will be listed on the NHS Digital Data Release Register, and also listed on The Health Foundation website so that the public can view information about how the data will be used. Given the remit of The Health Foundation to improve the quality of health care received, and that the organisation is an independent charity, and that The Health Foundation is committed to transparency, then it would be possible to explain the use of the data to patients. In addition, The Health Foundation has robust controls in place to ensure that any results of the analysis published cannot be used to identify patients, and that a number of other technical safeguards are in place to protect the data (as demonstrated by compliance with the Data Security and Protection Toolkit). Balancing the benefits of the project with the unlikely impact on any individual patient, The Health Foundation believes it will process the data requested fairly.

The Health Foundation believes that it is in the legitimate interests to process the data applied for (which include special categories of data, containing health records of individuals). The Health Foundation exists to help support improvements in public health. The AQP scheme has been identified as a programme that can deliver such improvements; The Health Foundation wishes to use necessary patient data to evaluate this scheme to determine if the scheme can deliver improvements to public health. In addition, The Health Foundation is confident it has the necessary safeguards in place to protect the confidentiality of the data (as demonstrated by compliance with the Data Security and Protection Toolkit), and therefore, any risk to the public by dissemination of the data to The Health Foundation are minimal and are mitigated.

Patient-level data are required, including diagnostic and surgery codes, and geographic information about where treatment was received by patients, throughout time (before and after the AQP programme was introduced).

Patient-level data from HES are required in order to undertake the following:

- Create measures of quality of treatment (for example, volumes of elective surgery such as knee and hip replacements, readmission rates for elective operations). The patient-level data will allow The Health Foundation to extract patient data for these types of treatment to be able to create the quality measures. Then, The Health Foundation will be able to use this quality measure and compare whether quality improved or fell when the AQP scheme was introduced

- In particular, the diagnostic and surgery codes are required to work out which patients receive elective surgery for e.g. knee and hip operations, so that the quality of treatment measures can be constructed

- Admission dates are required to sort out which patients received certain elective treatments (such as knee or hip operations) before or after AQPs were introduced, then it will be possible to determine whether the AQPs improved quality of treatment received over time. In addition, admission dates are required to calculate readmission, which is a potential measure of quality (for example, if a patient has had to have two or more hip operations in a short period, this may infer low quality treatment was provided to begin with).

- Geographies are required because the AQP was introduced only in certain areas, so these data are necessary to work out areas where AQP were introduced and where they were not, to be able to compare the effectiveness of AQP.

The English government introduced two policies to increase the non-NHS supply of care to NHS patients. The first was the new Treatment Centre programme including Independent Sector Treatment Centres (ISTCs). The introduction of ISTCs in 2003 was specifically aimed at expanding capacity in elective care for which there were long waiting lists in disease areas like hip, knees and cataracts. The second policy was the introduction of the Any Qualified Provider (AQP) programme in 2008 which allows licensed non-NHS providers to supply care to NHS commissioners at the same tariff as set for NHS providers. The AQP programme covered all elective hospital treatment in England. Capacity came on stream progressively and by 2011/12 19% of NHS funded elective hip and knee replacements in England were done in the private sector. These policy changes not only increased the capacity of hospital supply in elective services through the non-NHS providers but also gave an additional incentive for NHS providers to improve their quality of services.

The AQP programme was designed to introduce competition in a way that would improve the quality of treatment provided to patients. Reductions in waiting times for elective surgery such as hip and knee replacements are one way of measuring the effectiveness of AQP and this was already been undertaken. The Health Foundation would like to consider whether the AQP affected the quality of treatment provided in addition to changes in waiting times. For example, if a patient has to have many repeat hip replacements in a short spell of time, then it may infer poor quality treatment has been provided. The Health Foundation would like to determine how quality of treatment was affected by increased competition by health service providers when introduced by the AQP programme.

Patient-level data will be used for this project, so the data subjects will be patients who received hospital treatment for e.g. elective surgery in England before and after the time at which AQP was introduced.

AQP was only introduced in certain geographic areas. Therefore, the patients will be grouped into the following control groups and cohorts:

1 – patients who received treatment in hospitals before AQP was introduced

2 – patients who received treatment in hospitals in areas where AQP was introduced

3 – patients who received treatment in hospitals in areas where AQP was not introduced

The Health Foundation will be able to compare the differences in the outcomes in the quality of treatment received by patients in each of these groups so as able to determine how effective the AQP programme was.

As agreed in the original application, data is only being accessed by substantive Health Foundation’s employees and an award holder (on secondment to the Health Foundation) from the Office of Health Economics who has been specifically contracted for this analysis and has signed specific Health Foundation, IG and non-disclosure agreements.

To understand the impact of these two policies, it is important for policy makers (e.g. Department of Health, NHS England and NHS Improvement) to have empirical evidence on how NHS providers and non-NHS providers responded to the scope for increased non-NHS supply in NHS funded elective hospital services. However, such empirical evidence is very limited. Brown et al. (2008) used the Patients Reported Outcomes Measures (PROMs) data to compare the case-mix and patients' reported outcomes of surgery in ISTCs and in NHS providers. They find that the case-mix of patients treated in ISTCs reported slightly better outcome from that in NHS providers. However, there has not yet been any evaluation of the impact of introducing the AQP in England in 2008 on the hospital healthcare supply and quality of hospital services from private providers and its spill-over effect to the performance of NHS providers. This gap motivates the evaluation project. In particular this proposed study will empirically evaluate whether and to what extent the introduction of the AQP reduce patients’ hospital waiting time and increased the supply of hospital volumes. The Health Foundation will also explore how the private and NHS providers interact with each other under this policy context.

The Health Foundation will use the completed financial years of HES data from 2007/8 onward for the empirical analysis. AQP policy was introduced to England on the 1st April 2008. Hospital performance data in 2007/8 will be used as pre-AQP information. Data from 2008/9 onward will be used as post-AQP information. Provider information is recorded at hospital site level and patient activity and outcomes information is recorded at finished consultant episodes level. Data between 2003/4 and 2006/7 are used as Instrumental Variable (IV) for the econometric analysis. More details are reported in the method section. The data requested will be accessed by Health Foundation staff only (including a secondee from the Office of Health Economics (OHE), a research charity organisation. The secondee will be treated as a substantive employee of the Health Foundation). The Health Foundation takes full responsibility for the Information Governance and Data Security of the data held and processed. All outputs will contain only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide.

The Health Foundation will focus on the elective hip surgery data analysis. The Health Foundation will split the sample into four groups, i.e. pre-AQP data from Middle Super Output Areas (MSOA) that have no exposure to private providers, post-AQP data from MSOAs that have no exposure to private providers, pre-AQP data from MOSAs that have exposure to private providers and post-AQP data from MSOAs that have exposure to private providers. HES data will not be linked with any other data. The Health Foundation will take a Difference-in-Differences (DID) method to model two hospital outcomes measures, i.e. volumes and waiting time. The results will provide an estimate of the impact of introducing AQP on the performance of private providers and its spill-over effect on the NHS providers.

Private providers from Independent Sector Treatment Centres (ISTC) were introduced in 2003 while the Any Qualified Provider programme was implemented in 2008. It is then expected that different geographical areas (MSOA in this case) were more or less exposed to private providers from 2003 and 2007, and that this level of exposure was not random. For this reason, The Health Foundation will create a proxy measure, e.g. an "instrumental variable", which will flag the MSOA where private providers were already in place through ISTCs before the Any Qualified Provider programme started in 2008. In this way, The Health Foundation can control statistically for the previous exposure to ISTC and obtain robust estimates of the Difference in differences statistical model.

Both models that measure the hospital performance by volumes and waiting time will run at patients’ level as well as the MSOA level.

The research output will be reported to the key stakeholders, such as the English Department of Health, NHS England and NHS Improvement, through sharing the research report and academic publication, as well as presenting to key decision makers, e.g. Any Qualified Provider Team at the Department of Health and the NHS England.

The Health Foundation has confirmed that the HES data will not be linked with any other data.

All outputs will contain only data that is aggregated with small numbers suppressed in line with the HES Analysis Guide.

No additional years of data is being requested for this extension request to 01/02/2021.

All non-necessary variables (like maternity, psychiatry, etc.) have been dropped.

In order to provide empirical evidence, The Health Foundation will need to use real findings and not rely on estimates. Justification for the number of years of HES data requested is provided:

- HES data between 2003/4 and 2006/7 are used as Instrumental Variable for the econometric analysis.

- HES data from 2007/8 onward is used for the empirical analysis. The AQP Programme was introduced on 1st April 2008, 2007/8 data will be used as pre-AQP information. Data from 2008/9 onward will be used as post-AQP information.

The Health Foundation has confirmed that the HES data will not be linked with any other data.

The Office of Health Economics and the Health Foundation collaborate in the dissemination of the findings of the analysis. (The Office of Health Economics conducts research and provides consultancy services on health economics and related policy issues that affect health care and the life sciences industries).

Only one person from the Office of Health Economics (who has been contracted to the Health Foundation to complete this work) accesses the record level data to do analysis.

Any other employees from the Office of Health Economics would only see and deal with aggregated data with small number suppression applied when determining and collaborating in the dissemination of the findings of the analysis.

As well as publishing the work, the evaluation output will be widely disseminated to key stakeholders, for example the English Department of Health, NHS England and NHS Improvement and academic partner.

Data is only being accessed by substantive Health Foundation’s employees and an award holder from the Office of Health Economics who has been specifically contracted for this analysis and has signed specific Health Foundation, IG and non-disclosure agreements. The Office of Health Economics conducts research and provides consultancy services on health economics and related policy issues that affect health care and the life sciences industries. The Office of Health Economics and the Health Foundation collaborate in the dissemination of the findings of this analysis but it is the Health Foundation who ultimately make decisions about the data - thus they are sole data controller.

The Health Foundation originally expected the project to be completed by 31st August 2018. Due to staff changes the analysis has been extended until the first quarter of 2021.

In the second half of 2020/21, The Health Foundation will expect to publish a report to measure the impact of non-NHS provision on quality of NHS elective orthopaedics. The report will investigate whether exposure to greater competition from independent sector providers had any impact on the quality of elective surgery provided by NHS hospitals, as measured by revisions. The difference between private providers introduced first as Independent Sector Treatment Centres, and later as Any Qualified providers will be considered.

The Health Foundation results will be presented at key national and international conferences as well as the Office of Health Economics lunch time seminar (for some examples see https://www.ohe.org/events). Office of Health Economics seminars consist of 45 minutes presentations of the research findings, followed by discussion with audience for 45 minutes. Each of the participants gets a hard copy of the research paper/journal article. The aims of the seminar include (1) disseminating the research findings to policy makers and academic researchers; (2) providing opportunities to discuss with the key stakeholders. Attendance at the OHE seminars are by invitation only, these invitations are extended to policy makers at DH and NHS England and in the case of this study the specific policy makers who are already aware of the work will be invited to attend.

The Office of Health Economics (OHE) will seek to engage with the public and specifically patients by promoting the work to a number of patient advocacy groups. OHE will establish a webpage which will provide details on the specifics of the project (aims, methods, expected impact) and will be regularly updated with progress as findings and results emerge. The webpage will be written in plain English with simple messaging. Patient advocacy groups will be contacted and directed at the website. In addition, the findings of this project will be disseminated to the public and patients via regular blogs and via social media.

The Office of Health Economics aim to target the following patients group:

· Patient Participant Groups at GP practices and NHS Trusts

· Patient Liaison Group of the British Orthopaedic Association

The AQP policy allows qualified private providers in the NHS England to supply NHS paid hospital service. The policy provides direct incentives for the private providers to improve their performance as well as indirect incentives for the NHS providers through spill-over effect. It is important for the policy makers to understand (1) whether and how this policy might have an impact, (2) whether and how it has impact on NHS or/and private providers, (3) what the impacts are on patients’ outcomes. This information is crucial for understanding the effectiveness of the policy.

The Health Foundation aims to disseminate the findings through seminars and conferences (between 2019 and 2020) and peer- reviewed publications (it is expected from the second half of 2019). It is expected that the research outputs will generate wide interest for the policy/decision makers who are responsible for the healthcare resources allocation in England, providers/managers of the NHS and private hospitals, and academic researchers in the areas of health economics, health services research and health policy. Those colleagues will be the target audience of the seminars, conferences, report and peer reviewed publications. The Health Foundation will contact the key stakeholders directly to make sure that the research outputs are delivered to the target audience. For instance, the participants of the OHE seminar are through invitation only with the maximum spaces of thirty attendees. OHE seminars are held by the Office of Health Economics six times each year. The key stakeholders are regularly invited and attend the relevant seminars. The key stakeholders for the project include policy and decision makers at the DH and NHS England. They will be invited to attend the presentation. In the case of previous projects, the Health Foundation have used these seminars to present work on “Association between market concentration of hospitals and patient health gain following hip replacement surgery”, which was also published on the journal of health service research and policy in 2015 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4268175/). The presentation to the event generated lots of interest from a number of key stakeholders and the team was invited to present at Monitor (NHS Improvement) to discuss with economists there about the practical implications of the findings to the quality of hospital care and hospital market concentration in England.

In addition, the Health Foundation will rely on their strong links with key stakeholders (NHS teams, national policymakers, e.g., NHS England, and patient advocacy groups) to further disseminate the results and maximise the impact of the work. More specifically, the Health Foundation will actively engage with national policy makers, practitioners and researchers on top of the objective of scientific publication. The project will rely on a member of the Foundation’s Communications team to lead on dissemination of findings which will involve additional channels such as TV, radio interviews, articles on national, local and online media.

Example of engagement and key stakeholders:

• Senior members of Health Foundation staff regularly meet with senior representatives from across government, including the Treasury, Department of Health and Arms-Length Bodies (e.g. Monitor, CQC, NHS England, Health Education England).

• Joint projects or collaborations with NHS organisations. One example is the partnership with NHS England in evaluating new models of care outlined in the Five Year Forward view (http://www.health.org.uk/programmes/projects/improvement-analytics-unit).

• Regular engagement with policy makers at all levels on a range of topics where the Health Foundation has particular expertise: policy, data analytics, economics, patient safety and person-centred care. The Health Foundation’s views are regularly sought on health policy and practice, meaning that the findings from these HES-based analyses will be communicated directly with policymakers. One example is the recent engagement of the Economics team with NHS Wales (http://www.health.org.uk/programmes/projects/fiscal-sustainability-nhs-wales), leading to the following report (http://www.health.org.uk/publication/path-sustainability).

• The Health Foundation has a long history of funding programmes across the NHS which help to improve the quality of health care. For example, funding work on the relationship between patient flow, costs and outcomes in two NHS hospital trusts, which is related to the new project on understanding the drivers of A&E attendances.

• The Health Foundation have an active audience of professionals working in the NHS, many of whom are fellows sponsored by the Health Foundation, award-holders or part of their alumni.

In addition, The Health Foundation’s work has achieved widespread, extensive media coverage; examples of this are BBC News, BBC Radio 4, Financial Times, The Times, Daily Telegraph and The Independent.

Patients will be given the opportunity to discuss the findings and get further involved with the work. This will include their participation to external events, bespoke presentations and external seminars. The Health Foundation’s Communications team works with teams across The Health Foundation to maximise the impact of the work conducted and reach audiences effectively. The Communications team will support the Office of Health Economics through its blog and website, media activity and will promote the findings through its Twitter account which includes many patient groups. OHE will use the feedback from patients to further enrich the analysis and its impact on health care. The engagement aims to increase awareness amongst patients on the effect of Any Qualified Provider programmes in terms of patient outcomes. The work will provide further insight into the way this type of policy can affect patients' decisions and the quality of their care. Findings and results will be presented in aggregated form, and record level data will only be seen by the Health Foundation.

All outputs will contain only data that is aggregated with small numbers suppresses in line with the HES Analysis Guide.

The future of provider competition and independent sector involvement remains uncertain and controversial, and the evaluation will provide hard facts to inform a debate characterised more by opinion than evidence. To disseminate the findings through academic conferences will inform the policy makers and healthcare providers about the effectiveness of the AQP policy on the performance of healthcare providers as well as the outcomes on patients.

The evaluation outputs will be widely disseminated to key stakeholders, including policy makers and health/social care providers. The research outputs will provide empirical evidence to the policy makers about the impacts of the AQP policy on the supply of healthcare in the NHS as well as to what extent it improves the output measured by the reduction of waiting time. For instance, one of the aims for introducing the AQP policy was to reduce patients’ waiting time for hospital services. The policy aims to achieve it by allowing more providers to join the market to compete for hospital services contract. However, there is no empirical evidence about the effectiveness of the AQP incentives on different types of hospital providers (NHS/private providers) regarding the reduction of hospital waiting time. The research will provide empirical evidence for this. This evidence will also inform the policy makers about how to improve the design of the incentives in the future.

In particular, The Health Foundation expects that potentially, patients could benefit from this work if it is proved that the AQP programme had a positive effect due to increased competition between local health treatment providers, and was rolled out more widely by e.g. NHS England. The Health Foundation will look to see if the following will be impacted:

- Short waiting times for routine surgery

- Better quality of treatment (and care)

- More efficiently delivered treatment (and care)

- Fewer readmissions (particular for routine elective surgery such as hip and knee replacements)

- Reduced public subsidies for NHS funders due to increased efficiency gains (such as fewer readmissions)

- Reduced mortality

In addition, if the scheme is proven successful, it could potentially be extended to community services, and specialist treatments, which would benefit patients seeking other treatments. Such benefits would be measured and the magnitude of the benefits could be calculated in future work using HES data again.

The initial analysis of the data has focused on data management and descriptive statistics of the study sample. These findings had been discussed internally with the project team at The Health Foundation and not disseminated to external stakeholders. A summary of the sample selection process and descriptive statistics obtained is provided below.

A representative sample for analysis has been selected from the pool of HES data. This sample comprises HES episodes for elective hip and knee procedures in each year. The data used in this analysis include HES data between financial years 2005/6 and 2008/9. Two types of elective surgeries including hip replacement and knee replacement surgery have been included. Hospital providers have been categorised in either (1) private provider or (2) NHS provider and estimates have been derived to suggest if a MSOA (or PCT) has patients exposed to hip or knee replacement surgeries completed by private providers.

Descriptive statistics have been obtained to report:

- the number of MSOAs and PCTs in England that exposed to private providers for hip or knee replacement surgeries between 2005/6 and 2008/9. For both types of surgeries, the number of MSOAs/PCTs that exposed to private providers increased between the four year period.

- the number of episodes completed between 2005/6 and 2008/9 by provider type. There is an increasing number of episodes that were completed by private providers. The increasing trend is also observed by NHS providers but less dramatic compare to private providers.

- the number of providers by type. In general, the overall number of providers and the number of private providers increased over the four year period. In contrast, a decreasing trend is observed for the number of NHS providers.

- the waiting time for patients to get treatment in days by years, types of surgeries, types of providers with unit level at PCT/MSOA.

- whether MSOAs/PCTs that exposed to private providers up to 2005/06 present different volumes in number of episodes completed and waiting times. The exposure to private providers in an MSOA/PCT will become a benefit through increased competition, in turn reducing waiting times for NHS patients, and improving the quality of hip and knee replacements as observed by changes in revisions relative to primary surgery at hospital level.