How good is your data literacy? Why it matters for health sector workers
As we enter a new data-driven frontier in healthcare, we know that the data literacy of health sector workers will materially impact our ability to make timely evidenced-based decisions and leverage new technologies.
If you feel your data skills could be better - don’t worry, you’re in good company. In a recent survey by Qlik, only 12% of healthcare professionals across Australia were fully confident in their ability to read, work, analyse and argue with data.
What is data literacy?
Bowen & Bartley (2013) define data literacy as “understanding what good data and data analysis is so that you can make stronger arguments and better evaluate the arguments of others”.
Data literacy empowers clinicians and managers to comprehend data and then use it to make clinical and business decisions with confidence. Yet we see significant variances in data literacy in individuals and organisations. Often there is a significant amount of access to good quality analytics and data storytelling at senior leadership levels in health, but that doesn’t necessarily trickle-down to middle management and clinical teams on the floor - they don’t always comprehend the data. The daily and weekly reports flow into their inboxes, and while they might check a few numbers, they don’t always know what to do with the information presented to them (*read* -> *archive*, repeat).
Why data literacy matters
The ability to understand and communicate clearly about data is an increasingly important skill for the 21st-century health sector worker, for a number of reasons.
Data literacy enables the effective use of analytics - supporting decisions at pace - which in turn improves clinical, operational and financial outcomes at every level of our health system,
When you’re comfortable analysing and interpreting your own information, it significantly reduces the lag in getting access to information. It also lets your analysts focus on other value-adding work,
We’re running close to the curve in the law of diminishing returns on improvement work, and clinical teams are burning out from change fatigue. We need data to better guide us towards where we should concentrate our improvement efforts; and
Even if you yourself don’t work directly with data, being data-literate will allow you to ask the right questions and meaningfully engage in the conversation at work.
Stages of Data Literacy
We see four key stages of data literacy for the typical health sector worker:
Awareness: At this stage we’re aware of the importance of data to our role. We have key measures that we look at in regular reports - we don’t do much with these other than note the numbers in the report; we often make decisions without data - but we know how measures are calculated and how they’re captured.
Comprehension: At this stage, we’re clear on the information we need to perform our role and how the results impact our organisation's strategy. We don’t totally comprehend what we’re looking at, but when people provide insights to us, we use them in our decision-making processes.
Interpretation: At this stage we’re able to do some of our own self-service data analysis - or to take existing reports we get and adjust them slightly to get better information. We’re comfortable with taking raw data and making decisions regularly.
Interrogation: This is where we have the confidence to do ‘self-service’ data analysis. We’re known by others as “being good with data” and are comfortable turning insights into engaging data stories which give people on our team the confidence to make decisions.
Not everybody needs to be a data guru, but as noted earlier, being data-literate will allow you to ask the right questions and meaningfully engage in evidence-based decision-making.
Hallmarks of data-literate organisations
Over the last five years, I’ve worked with more than ten health districts across Australia and New Zealand. Throughout that time I’ve seen varying levels of data maturity. Organisations with highly matured data literacy have the following things in common:
Teams regularly say “what does the data say”. Teams are able to comfortably read reports, comprehend what they’re saying and then make decisions. This leaves information and decision-making units with more time to focus on value-add analysis.
They know what to measure, and while they cover a broad-balanced range, they focus on impact measures. Highly data-matured healthcare organisations follow a balanced measurement strategy that features four components: experience, process, impact and balancing measures. They focus on impact measures (a few KPIs) because they measure objective output, but they know it can take a while “for the wheels to turn”, so, they use process measures (many metrics) because they’re leading indicators and quickly give them a nudge on how their systems are performing. Experience measures ensure they’re delivering better services to patients and staff. Keeping a watchful eye on balancing metrics ensures there’s no unintended consequences or negative outcomes present.
They understand and challenge each other when they’re using the wrong measurement lens for the situation. In health we often see three different approaches to measurement:
measurement for research,
performance management; and
measurement for improvement.
All of these lenses are important, but can sometimes unnecessarily encroach on each others’ spaces. Research methods involve large data samples with statistical testing of hypotheses, a strong focus on removing measurement error and improving data accuracy. The goal is to provide strong empirical evidence to inform clinical decision making.
Performance measurement measures work well at a system level when a few simple metrics are needed to measure system performance (think 4hr ED targets, surgery waiting lists) and the goal is to provide a global snapshot. However, these lose relevance when you look at the service or hospital level and you need a broader comprehensive metrics strategy to take into account the existent complex systems and interrelated parts of healthcare delivery. Broader metrics strategies use the four lenses: experience, process, impact and balancing measures.
Performance improvement takes a consistent and regular approach to measurement, continuously sampling processes while looking for trends, improvement and changes in variability. We often use run charts to measure changes over time. Those using a performance improvement mindset are much more comfortable with imperfect data and more focused on continuous testing over time.
What level is your data literacy at?
So, while you don’t need to be a data guru, you do need to know enough to ask good questions and make informed decisions.
The Data Literacy Project is a great place to get a litmus test on your data maturity. The self-assessment takes less than five minutes and the resulting learning roadmap will give you helpful podcasts, tools, ideas and other resources to start your upskilling journey, based on your starting point.
Where to next?
I strongly believe that increases in data literacy across our health workforce will be of significant benefit to health sector workers, to patients, and to the system.
If you’re interested in growing your data competence, check out Emerson.Live where we teach data skills specifically for health care workers: https://www.emerson.live/courses