Why do we measure? How do we decide what to measure? Metrics are commonly used to judge performance of individual personnel, complete systems or individual applications. They have their uses, some valid and, well, some rather useless.
In the software development field, the general term used to define what metrics to measure is
The most important consideration when deciding on which metrics to measure is how it improves the end users experience. What else matters as much? If I'm a medical professional with 15 minutes to spend with a patient I cannot afford to sit and chit chat for 5 minutes while my application loads, let alone jot down written notes while it updates. What happens when it freezes and the user has to log in again and re-launch the application? The impact of software application performance during an emergency event is tremendous. As a healthcare application end user, timeliness and accuracy are big concerns.
The application performance metrics you measure for a healthcare application should be focused on improving the customer end-user experience and positively affecting the quality of patient care. Metrics that measure how software applications perform is critical to improving patient care, satisfying regulatory requirements and allowing medical professionals to do a critical job without undue interference.
Defining useful metrics
We're able to measure just about anything in software application performance, including applications, system hardware and database performance. The bigger question is what needs to be measured? Begin by finding what metrics in the healthcare application offer the opportunity to improve the end-user experience. Metrics should discover areas of code weakness and performance issues within and between applications or suites. Then, find out what can repair and improve your application quality. Improve the code to improve the end-user experience.
Spend the time to define a common set of metrics to start collecting data on for the application. Next, a useful and standard way to present or display the data is needed. Make it as simple and clear cut as possible so it can be easily read and understood. If the application runs on a cloud-based or other distributed system, consider measuring performance timings or response times from end user data. Measure the same response times on different hardware platforms or different browsers, depending on what the system uses. Consider leveraging the testing or customer support resources to obtain data or find a willing end user who can monitor system response times.
Depending on if the application integrates with other related suites or another vendors, consider measuring response times between applications. Where are the slow spots? Does the system have trouble switching over? Where are the areas in an application workflow that slow a physician down or make a nurse exit and log back in to get an updated patient record before she can administer patient medication? Consider the interaction of the software application with medical devices, such as heart monitors and respiratory or other critical health-monitoring devices.
Both time and data accuracy measures should be considered for any healthcare software application. Exactly how it gets done depends on the application's design and how it works in an integrated workflow among other applications and medical devices. Consider creating a team including a representative from each application development area to define and design metrics. Try them out and then change them if it is not providing information that allows improvement for the end-user experience.
The user-experience effect
The U.S. government's emphasis on healthcare reform and regulations to improve patient care has proven that measuring medical providers' use of evidence-based strategies and patients' health outcomes have changed the medical profession. It is actually changing the core culture of healthcare delivery and providing a growing acceptance that healthcare is not random, but can be objectively measured. Similar to software, if it can be measured, it can then improve the end user or patient experience.
Software applications used in healthcare are a large part of that change. Performance is critical because medical personnel perform time-sensitive, life and death operations based on, or at least influenced by, healthcare software. Granted, there are numerous methods of improving patient safety and care, but when you think about it, every single aspect of medical care increasingly relies on healthcare software applications.
For example, you're a patient heading in for day surgery. First, the admission staff uses an application to check you in and take all your necessary personal information including allergies and medications you're currently taking. Next, your physician places medication and lab orders so you're ready to go once you're admitted. The nurse assigned to care for you relies on the healthcare application to be accurate and timely when she or he administers medications. Any changes the physician makes to your patient record needs to reach the nurse or other caregiver quickly and accurately. In other words, you could miss a medication dose or get an extra dose if the application doesn't perform in a manner that alerts your caregiver to changes on your record. The workflow between medical personnel at each level is critical to giving the patient the best, safest and most accurate care possible.
Application performance metrics for a healthcare application should focus on improving customer and end-user experience and positively impact the quality of patient care. Application performance metrics provide objective evidence indicating where your application can be improved and if it has improved. When defining metrics, pick measures that expose issues within the application as well as issues between applications used in a healthcare workflow and any affected medical devices. Create reports that display the data you measure in an easy-to-understand display.
Application performance metrics and monitoring is a useful and effective way to improve the quality of patient outcomes by improving the quality of the application end-users experience.
This was first published in December 2013