Welcome to the MedMetrics Blog

The MedMetrics blog provides comments and insights regarding the world of Workers’ Compensation, principally, issues that are medically-related. The blog offers viewpoints regarding issues affecting the industry written by persons who have long experience in the industry. Our intent is to offer additional fabric, perspective, and hopefully, inspiration to our readers.

Search The MedMetrics Blog

Thursday, May 25, 2017

Seek Counsel from the Data for WC Medical Management

by Karen Wolfe

“Given the avalanche of information that has become available to ­businesses over the past several years, data-driven decision-making (DDDM), the practice of basing business decisions on data analysis rather than intuition, has become a critical tool to help organizations reduce risk, avoid costly mistakes and take advantage of opportunities.”[1] In Workers’ Comp, there are scores of reasons to seek council from the data.
Predictive Analytics
Nevertheless, raw data must be analyzed to be useful. Predictive analytics looks for trends and patterns in historic data. Such analysis can predict probable ultimate medical costs in new claims when similar conditions occur. Predictive analytics uncovers cost drivers that might include organizational traits, timeliness of action, or specific serious injuries. Moreover, analysis of past performance by medical doctors offers decision support for selecting best providers going forward. These are only a few examples.

Data-driven decision-making in medical management is a powerful tool, providing knowledge and guidance for those who make direction-changing decisions in real time. Yet, all data-driven decision-making efforts are dependent on the quality and limitations of the data used.[2] Those relying on information provided by analytics must be confident the data is accurate and complete.


Data reliability

If data quality is inconsistent, incomplete, or has errors, conclusions derived from it cannot be trusted by those who would rely on it for decision-making. For instance, medical provider performance analysis relies heavily on provider data accuracy. Missing data items such as NPI (National Provider Identifier from CMS, Centers for Medicare and Medicaid Services) prevents accurate individual or entity identification. Individual providers cannot be accurately distinguished. Still other data quality issues are concerning.

Misspelling addresses creates duplicate entities in the data, thereby skewing analysis. Some of the supporting data is attributed to a provider name and address spelled one way and the rest of the data is attributed to another that is spelled differently, but is actually the same person. Separate files for the same provider prevent fair analysis of performance because the data for both is incomplete.
Similarly, data omissions lead to difficulties in interpreting the data. Missing key data elements such as medical provider specialty can make the data ineffectual for evaluating performance.
Errors are often caused by manual data entry. Years ago, typists competed for jobs based on how fast they could type a paragraph without errors. Today’s data entry personnel should be evaluated on accuracy, as well. Make accuracy a performance measurement.
Data quality
Often data is transmitted to an organization from outside entities. Hard copy forms are translated to digital formats using optical character recognition, OCR. The organization wanting to analyze the data for use in decision-making did not create the data, but it should not fall victim to it. The organization should select critical data elements and proceed to correct them in the data. The resources required are easily justified by the end game: applying analytics to create accurate decision support.
Fix the flaws
Data-driven decision-making in WC medical management is powerful. Frontline professionals and other stakeholders are made accurate and efficient. That means costly mistakes and re-dos are avoided. Moreover, interventions and actions are timely, all resulting in significant cost savings for the organization. If the data producing knowledge and decision support is not accurate and complete, fix it!

Karen Wolfe is the founder and President of MedMetrics®, LLC, a Workers’ Compensation, predictive analytics-informed medical loss management and technical services company. MedMetrics offers intelligent medical management systems that link analytics to operations, thereby making insights actionable and the results measureable. karenwolfe@medmetrics.org

[1] Heires, K. Flaws in the Data. Risk Management. April 3, 2017.
[2] Ibid.