Self-Service Hospital Reporting Possibilities: Enabling Clinicians to Make Faster and More Informed Decisions

Hospital Reporting There is little doubt that we are rapidly becoming a self-service society. Need to buy food? Many large grocery chains now offer self-checkout lines, and other retailers are following suit. Going to a concert, sporting event, or the movies? You can purchase (and print) tickets online. Taking a vacation? You can plan your entire trip from your living room – no travel agent required.

Along with the availability of data, these self-service capabilities are made possible by technology that simplifies what were once more complex operations. Self-service capability is now making its way into healthcare analytics and hospital reporting – driven by the availability of large amounts of data gathered through electronic health records (EHRs) and a new generation of tools that put many analytics capabilities into the hands of clinicians and other front-line users.

Healthcare Analytics: Why It Used to Take So Long

This is a significant change in thinking. Traditionally, healthcare data analytics and hospital reporting has been exclusively under the control of IT. The complexity of the tools, concerns about who should be given access to protected health information (PHI), territorial battles over data ownership, a less technology-savvy provider/administrative base, and other factors have required clinicians and other front-line personnel to submit any and all requests for analytics to IT.

They would then have to wait for the request to work its way through the queue and for results to be returned – a process that could take several weeks, or even months, before a request turns into a solution. By then, the conditions may have completely changed, rendering the solution moot. Even if conditions don’t change, the delay created a several-week gap before the organization could come to an understanding of the current state and begin moving toward the ideal state. This process put an unhealthy barrier between the clinical side of the organization and its data, which led to frustration, distrust, and disengagement by clinicians and other front-line personnel.

With self-service analytics tools, clinicians and other users can access and analyze the wealth of data their organizations have accumulated, building and executing many queries themselves using drag and drop techniques. They can use these tools to create ad hoc reports or generate visual representations that tell a story about their data, such as stoplight graphs or heat maps.

Moving to a self-service model not only helps users get to decision-aiding information faster. It also helps lower the cost of analytics by removing the need to involve multiple parties in many of these queries.

This is not to say that a self-service model completely eliminates the need for data analysts; they will still be required for more complex and in-depth analysis of larger questions, allowing them to operate at the top of their education and training. But for simpler and more direct queries, giving clinicians and other users the tools to perform their own analytics encourages them to learn more about their areas on their own while freeing IT data analysts to focus on answering the more difficult, enterprise-affecting questions.

Self-service in Healthcare: Clinicians and Front-line Users Help Themselves

The primary value of the self-service model is it allows clinicians and front-line users to form a hypothesis about a specific area of concern, explore the data surrounding it, compare it to external benchmarking data, and gain a greater understanding of the data, all without having to involve IT.

Once they understand more about what the data might tell them and what they’re after, the users can go to IT and ask more detailed questions that the self-service tool isn’t built to answer. Of course, every query doesn’t have to lead to some greater revelation. Sometimes the user just needs to pull a simple report to get a quick answer. For example, if the cardiac unit has made a change to standardize how heart failure patients are treated, a self-service report can show whether readmissions and/or costs have gone up, down, or remained the same. Discovering the answer might be all that’s required for a report to verify that the initiative is working.

The key is to ensure flexibility in the design of the self-service tools to allow the user to produce self-contained answers as well as to lead to larger insights. It is important to manage expectations with what self-service offers and what it doesn’t.

Simple vs. Easy: Helping Users Understand the Difference

Because of the speed and simpler mechanics involved in working with self-service tools, there is a tendency to equate self-service with easy. While self-service tools don’t require the level of sophisticated technical knowledge required for IT data analyst-level reporting, there is still a learning curve, and it can be steep depending on the users’ comfort level with technology and the types of insights they want to glean.

That is why it is important to the success of the program to set user expectations properly from the beginning. They need to understand the types of insights they’ll be able to gain and what it will take to gain them – as well as what self-service will not do for them.

They may also find the reports developed via self-service tools are not as flexible as the reports generated by IT because some portion of the data they’re working with will need to be bound to business rules and vocabulary sooner. This factor may limit the reports or visualizations they can produce, especially when the source data tends to be more dynamic. The more standardization that’s required for the users, the sooner the data will need to be bound. There may be other limitations based on the nature of the database or the hardware as well. All of that should be made clear from the beginning.

Whether the organization has a self-service or a full-service reporting infrastructure, it is important to consider how users engage with IT. In both cases, if clinicians and front-line users resort to a traditional, one-off reporting mentality, they risk failing the project. There is huge value in a cross functional, permanent team that focuses a particular area and is able to initiate effective improvement projects. This allows for continuous focus, better engagement, and quicker, more accurate turn-around time.

Launching a Healthcare Self-service Initiative

Once the organization makes a decision to move forward with self-service analytics the real challenges begin. The first is who should own the project.

While there is a temptation to think of self-service analytics as an IT project because it involves data, the reality is the user community must initiate it and play a leading role. They must see the need to access and manipulate data themselves; if IT builds it on its own, the users won’t come. The organization’s leadership must be seen as the driving force, with clear goals and expectations in mind, and participate with analysis, design, development, security, rollout, and training. They must help generate enthusiasm for it, or the project will never achieve the desired results.

With a firmly established cross-functional team in place (possibly a clinician, a nurse, a data architect, a knowledge manager, etc.), it is time for IT the project team, preferably following an iterative approach, to build the tools and enable access to the data under the sponsorship of an owner. Self-service analytics tools can be built around inpatient or outpatient services, a specific department, or even a particular disease state or condition.

For example, if the organization is enabling self-service for the cardiovascular unit, the project will fall under the oversight of a cardiovascular leader who understands the clinical issues as well as the types of reports, visualizations, or other outputs most relevant and beneficial for that particular unit.

An enterprise data warehouse (EDW) helps prevent competing siloed data, and it is definitely helpful for integration – particularly if data for the reports will come from different areas, such as clinical, financial, patient satisfaction scores, supply chain, etc. An EDW enables users to obtain these different types of data securely and without the risk of creating redundant extracts from the source data. It also provides greater support for HIPPA compliance.

The big advantage to providing easy-yet-controlled access to data is this type of access ensures the right data is being used. Not having an effectively managed EDW and metadata repository with quality underlying data can lead to false conclusions. Making organizational data centrally and readily available helps ensure users are using high-quality data as the foundation for their reports.

Self-service in Healthcare: Champions and Training

When the technical aspects are enabled, the next step is for the leader who owns the self-service tools to begin championing them. The owner, with IT playing a supporting role, should demonstrate the new capabilities to others and show how they can help improve the delivery of care.

The final step is the development of a formal, ongoing training program for users that incorporates actual organizational data into the training structure (rather than generic data). Often, there is an assumption in IT that you can do a single two-hour training session and then send users on their way. That may work with some, but most users learn best by doing. Organizations successfully using self-service analytics tools have incorporated continuous course classes around self-service where this valuable asset is shared again and again with clinicians and front-line users.

There needs to be an ongoing partnership with the users to provide support and additional training where needed – especially since it’s likely that the development process for the tools will be iterative. Establishing a user group facilitated by IT, where users share new discoveries with one another, can relieve IT of some of the training burden. It can also help IT discover who the super-users are as they will be advocates for the self-service analytics program going forward.

Traditional Healthcare Reporting: Dealing with Discrepancies

As users begin taking advantage of the self-service tools, they may find discrepancies between the results from a legacy report produced by IT and their new reports. Generally, there is a tendency to assume that the new report is incorrect because it was self-generated and isn’t a “real” report created by IT. While that can happen due to users not using filters or rankings correctly, it isn’t always the case.

The primary drivers of quality are the integrity and currency of the data being used and how the report was designed. An older legacy report may not have gone through the same quality process the organization is currently using, which means it could have built-in errors that have been carried forward since it was developed. Conversely, the data and design incorporated into the self-service tools have presumably been freshly examined and validated as part of the enablement process.

Regardless of where the discrepancies come from, it is important that users understand they should point out the problems so they can be investigated and remediated. The goal is to bring accurate information and insights to them. Rooting out discrepancies between old reports and new ones is part of the process and should be expected; creating a feedback loop between end-users and operations is imperative. Upfront discussions and setting expectations around data quality are an important part of education and training.

Accelerating the Pace of a Healthcare Organization’s Decisions

The entire healthcare industry is undergoing rapid and sweeping changes. Clinicians and front-line personnel need a way to obtain information faster than the weeks it can take even the most simple of reports to work their way through an over-burdened IT queue.

Self-service analytics tools can help clinicians and front-line healthcare personnel create and run their own hospital reports, giving them the information they need to make timelier decisions. This, in turn, will help improve the quality and safety of care as well as reduce the cost. All while freeing IT to focus on larger, more complex issues. It’s a win all around.

Loading next article...