Lessons learnt from the delivery of a successful…

This article was first published in The Journal of mHealth

“Without data, you’re just another person with an opinion.” - W. Edwards Deming.

Operating theatres are critical to the delivery of acute healthcare, they are also one of the most expensive resources within a hospital and receive the most attention when trusts are under financial pressure. In addition, they are having to deal with ever increasing demand.

Over four million people are currently waiting for surgery and operating waiting lists have increased by another 5% in the last year. Reports also suggest that 291,000 additional surgical procedures ‘could’ have been undertaken last year, if hospital trusts improved the scheduling of their operating lists.

These figures are well known but what is less-well publicised is the cumulative costs of these inefficiencies. Waiting list initiatives, overtime payments, private sector activity, as well as locum, agency and bank expenditure ensure trusts continue to haemorrhage cash as they try to meet patients’ needs and their contractual obligations. That means many trusts are looking for support to transform the operation of their theatres.

The key questions I always face when I am working on these transformation projects is: “what additional revenue can we expect to receive, and what cost savings will you deliver?” My response is always frustrating to my questioner, “to give you a reliable answer, I’d really need to see your data”.

Garbage in, garbage out

With a certain degree of confidence, it is possible to predict the levels of increased activity and revenues, or the potential cost savings which could be reliably gained from a focused and structured theatre transformation programme. There are tried and tested approaches and blueprints available as well as track records and case studies to draw on, but a truly effective transformation requires reliable data and the trouble is that data is not always available.

When activity and cost data is requested, a common response is, “But I don’t believe our data is accurate.” Unfortunately that assessment is frequently correct and if the data quality is less than perfect, then it becomes difficult to build confidence in the improvement potential.

In my experience, if your people lack confidence in the underpinning data – believe them. Conversely, there is a need to be cautious when they tell you their data is strong. This is not to suggest that they are untrustworthy but it is vital that the data used to inform any key decision making must be scrutinised and its reliability and validity challenged.

Our theatre improvement work at The Leeds Teaching Hospitals NHS Trust in 2018 and 2019 supports this assertion. For instance, although key timestamps were recorded at every stage during an operating procedure, the scheduling of theatre lists continued to be based on procedure durations that had been estimated by the surgeon. Only when we compared predicted case length versus actual durations were we able to convincingly show the potential to schedule additional patients on lists. Consequently a scheduling process was introduced based on median case times for specific procedures, and particular surgeons and anaesthetists. This allowed an additional 488 cases to be delivered within the trust’s existing resourcing.

The critical lesson of this experience is that poor data quality should not be a reason for inaction. Forging ahead when you have little faith in your own data is uncomfortable but maintaining the status quo isn’t working. Static operating theatre scheduling templates which have remained unchanged for several years are unsustainable and action is needed now. The challenge is to find ways to engage those involved so that they either use existing data better or seek to improve its quality and usefulness.

The virtuous circle

If you were to ask me what the key success factors in Leeds were, they would include an innovative scheduling solution, bulletproof data quality, and some easy wins. However, the fundamental driver, that was more important than all of these, was the effort made to truly engage all staff groups with the underpinning data and empower them to own the issues, challenges and solutions.

People engage with data when it is provided to them in a format which is useful, meaningful or even provocative. They take ownership when they feel the data is being used to inform decision making, and their insight plays a valuable part in enriching it. By presenting the data effectively you will see improvements in data quality, because the people who can influence the data will always act to make it reflect the difficult and often impossible tasks they face every single day.

In Leeds one of the key issues was lack of workforce engagement and fatigue. Anecdotally we heard that theatre teams were facing regular and significant overruns that affected morale and work-life balance. When scrutinising this data, it was apparent that overruns were actually less frequent than the team thought. However when they did occur, late finishes could be significant. Engaging all staff groups around this data was helpful both in allaying their dissatisfaction and showing how scheduling decisions were impacting others. The resulting clearer communication and forecasting accuracy led to a 30 hour per month reduction in overruns overall.

Previously some of this data had been inaccurate or non-existent, making the idea of basing decisions on it seem counter-intuitive. However the engagement with those using it, or who had a vested interest in its effective use, meant that almost overnight, it was possible to transform teams and departments into a much more data-driven operation, almost by stealth. What is important in these situations is to visualise what information you have, irrespective of your confidence in it. Engage people with it and, where challenge exists, interrogate the data and underlying process to make it right and improve confidence in it.

Making it stick

However, creating sustainable improvement involves more than simply implementing some tried and tested process, tool or blueprint. Fundamental to embedding new processes is the need to shape thinking and in Leeds a key to achieving this was to make the data ‘sticky’. Thanks to the operational and clinical teams, we were able to embed the use of data into daily decision making, whether that was to provide more accurate scheduling, improve list utilisation forecasting or more robust patient waiting list management.

We quickly found that once the team cared about the data, they will behave in ways to improve it, which in itself opens up further opportunities to use analytical outputs to drive transformation further down the line. After all, data should drive insight, which should drive action. In Leeds, this created a legacy of fact-based, data-driven decision making which will empower, enable and encourage the teams to make informed improvement decisions and generate new ideas.

The experience showed clearly that transforming mindsets will always trump overhauling processes. So, even if you don’t trust the data, get it out there and let your people own it. Because, as Deming said, “without data, you’re just another person with an opinion”.