SAS® Analytics helps leading research trust to inform the debate on UK health care
The Nuffield Trust is using SAS® to underpin essential research projects that are delivering new insights and helping drive the debate on UK healthcare.
Established in 1940, the Nuffield Trust is an authoritative and independent source of evidence-based research and policy analysis into health care in the UK. Its findings support healthcare providers, inform government decision-making and regularly feature in the media, fuelling the ongoing debate amongst policymakers, practitioners and the public on the future of health and social care in the UK. The Trust has used SAS® to provide essential analytics since 2007.
"We have a reputation for robust work - providing authoritative results people can rely on," says Adam Steventon, Senior Research Analyst. "We want to maintain that reputation, which is obviously linked to our use of SAS." His team focuses on quantitative analyses working with very large data volumes, with the results used by groups ranging from ministers and senior civil servants to health service commissioners and practitioners. "We're a new in-house team, established over the last few years, and some of us brought SAS experience from previous roles," he continues. "We considered other providers but decided SAS was a good (preferred) solution, especially given the data volumes involved. SAS can handle a lot of data, and does so pretty quickly. And since moving SAS to the server, it's astonishingly fast." A charitable trust, the Nuffield Trust licences SAS at a discounted rate.
Hundreds of millions of records
Steventon's team accesses and uses large administrative data sets from the NHS and other sources, including national Hospital Episode Statistics covering 12 million inpatient admissions each year and many more outpatient attendances. "We also link together with very novel data sets from GP practices, local authority and social care sources," he says, "which is exciting because we can build pictures of how individuals interact with the health and social care system over a number of years." This results in very large data volumes the Nuffield Trust is keen to explore to answer a range of research questions. "Some projects require hundreds of millions of records, from multiple sources and in different formats, from several hundred providers," Steventon adds. "We clean all that data, and link it together at the person level." SAS provides the flexibility to challenge, query and check analyses using a variety of methods. "Another thing we like about SAS is having the ability to write our own macros and commands. Most of the analytics are already available in SAS, but some more sophisticated ones aren't provided as pre-written functions, so we can write our own. We're a small team, and often collaborate on projects together. So we also need good audit capabilities, to produce data sets other people can understand and use, and keep records of how they were derived. SAS enables all that."
With analytical rigour and research quality taking precedence over speed, there are still opportunities to work rapidly: to deliver timely and so more relevant feedback for healthcare providers and other interested parties. "We have regular data feeds, fresh input monthly, and use that to update some analyses," Steventon says. "This is particularly important for our 'evaluation' work." For example, one part of the UK may be doing something innovative to prevent hospital admissions, which are undesirable to the individual and costly; it's highly useful for the Trust to evaluate outcomes quickly and efficiently "to give people on the ground as up-to-date a picture as possible."
Emergency admissions and telehealth
A project involving emergency admissions provides an example of how SAS is used. 35% of all admissions in the NHS England are classified emergencies, costing £11 billion a year, with Nuffield Trust analysis showing the number of such admissions had risen by 11.8% from 2004-9. That totals 1.35 million extra admissions costing over £1,000 each. "This research aimed to understand why this figure was rising," says Steventon. "We took a national approach, analysing data from throughout England. We used SAS for much the analysis, looking at different aspects across every single emergency admission." The output was a series of conclusions, including the fact that a rise in short-stay admissions (less severe cases) might suggest the clinical threshold for acute admissions had been lowered. Results were widely disseminated to policymakers and NHS providers and commissioners.
A more complex example involved using SAS in a large randomised control trial funded by the Department of Health. The Whole System Demonstrator (WSD) project looks at telehealth and telecare: using technology and a range of equipment to help people manage long-term conditions like diabetes and their care needs in their home, and aiming to maintain independence. "Pilot projects had suggested you could prevent admissions and save money while actually improving clinical outcomes, quality of care and the patient experience," says Steventon. In the current climate, being able to quantify the benefits delivered by such approaches are more important than ever before. SAS Analytics supports this research, in particular examining impacts on actual service use: "Whether it does reduce hospital or residential care admissions as well as impacts on GPs, community nursing, and so on," says Steventon. "Our approach is very novel, pulling together and using very large data sets from different NHS sources, local authorities, GPs and more, linking all this at a person level. We don't think anyone has done this before, and SAS is really good for this work – size is no object."
Having an impact
Steventon adds, "We track the success of our research, and our work has an impact on health policy development and commissioning. Our goal is to make sure health policy is informed by evidence as far as possible, and our work does lead to different decisions being taken. We are increasingly approached by organisations wanting to commission research. The big advantages of being able to use the administrative data sets is their national coverage, they're fairly standardised and they cover long periods, in some cases 10 years. We're exploring how we can use our analytics to help more and more NHS commissioners to evaluate their own pilot programmes, drawing on the data at our disposal. We're being approached to do a lot more evaluations because people trust us."
The results illustrated in this article are specific to the particular situations, business models, data input, and computing environments described herein. Each SAS customer’s experience is unique based on business and technical variables and all statements must be considered non-typical. Actual savings, results, and performance characteristics will vary depending on individual customer configurations and conditions. SAS does not guarantee or represent that every customer will achieve similar results. The only warranties for SAS products and services are those that are set forth in the express warranty statements in the written agreement for such products and services. Nothing herein should be construed as constituting an additional warranty. Customers have shared their successes with SAS as part of an agreed-upon contractual exchange or project success summarization following a successful implementation of SAS software. Brand and product names are trademarks of their respective companies.
Copyright © SAS Institute Inc. All Rights Reserved.
The Nuffield Trust
Dealing with very large data volumes (hundred of millions of records) and novel data sets; providing a wide range of robust analytical techniques plus the flexibility for bespoke development; delivering trusted research findings to maintain a world-class reputation
SAS® Analytics deployed first on the desktop then moved to a server-based solution, enabling wide-ranging quantitative analyses.
Processing very large and often novel data volumes in minutes; improved data flows, reliability and data security; enabling analyses to be challenged using various techniques; helping maintain and enhance the Trust's reputation; improved audit, governance and the ability to recreate and share analytical approaches.
“We considered other providers but decided SAS was a good solution. It can handle a lot of data, and pretty quickly. And since moving to the server, it's astonishingly fast. Computations that took two hours before take eight minutes.”
Senior Research Analyst