Domain specific testing has gained a lot of momentum in recent times. Testers are no longer focused just on the horizontal testing knowledge but also the vertical domain knowledge. Such a tester who is able to bring in the right marriage between the two is certainly able to differentiate himself from the rest. Today, while most domains are benefitting with the right use of data and data analytics solutions, the more sensitive ones such as healthcare, life sciences, banking are certainly the ones that stand out. The right set of data tools and analytics (both retrospective and predictive) are enabling players in these domains make very informed product and user decisions – all these together are bringing in much better user and market acceptance along with huge returns to the bottom line of the organization. The Chief Strategy Officer of the insurance provider Blue Cross Blue Shield presented on this topic on “The Power of Data to Transform HealthCare” in a conference that aims at transforming healthcare through IT, earlier this year.
While we often refer to the phrase data analytics in the context of data, the world of data has much more to offer and analytics is only one piece of it, albeit significant. The other core pieces include data reporting, data sciences and data quality. Herein data quality is worth talking more about – where it is important to ensure the quality of the data you deal with, to start your reporting, analytics and sciences efforts on – if the quality of the base data is not worthy, the entire effort becomes so much less relevant and significant. Often, due to lack of time and/or resources, the data quality angle may be overlooked but some amount of effort in this space goes a long way in ensuring the analytics outcomes are actionable especially for domains such as healthcare given the number of entities at play – physicians, health care provider organizations, patients, insurance providers etc.
The data quality effort in health care software testing includes big data testing so as to confirm data accuracy, relevance, completeness and security. Security is a big angle here to ensure the data is adequately masked and at the same time is realistic enough to represent true end user conditions. Such tests should be done on the raw data before any kind of analysis can be started. Data relevance is another important area to verify as often the data under test has significant gaps from what they would look like in production. Such a gap can result in distorted test results and scenario coverage as some workflows may not have been touched at all. The other problem that data quality testing should catch is to ensure all kinds of data have been accommodated for. This includes positive, negative, null and blank data.
While such data quality checks are important for any domain, it can be understandably lowered in priority with just a few core checks for most domains. For areas such as health care though, it cannot be compromised and should be an important item in the healthcare software testing efforts.