The Sea of Data Monsters…Health care claim data and other scary numbers
Patti Dunham, MBA, MA, SPHR
Strategic Human Resources, Inc.

We see the headlines and read the stories day after day.  Health care costs are rising, claims are increasing, and the overall health of our nation decreasing.  So what are we doing about it?  Many companies have taken the initiative to actively manage their health care programs, even those who may not have done so in the past.  Organizations are adding a bit of consumerism to their plans, implementing wellness initiatives, and creating cost-shifting alternatives to the plan design.  But is it enough?  Are we using data to analyze our groups and make appropriate plan and program decisions?  Are we shying away from the sea of data monsters because we don’t know what we are diving into and what the data means?  As we do with other human resources activities we need to analyze not just the cost of our benefit programs but the source of the costs and use the data to tame the monster and improve the health and welfare of not only our programs but of our group. 

For anyone who has tried to dive into this sea, I agree, health care claim data is a complicated animal.  For those of you fortunate enough to have access to your claim data, you know that the monthly barrage of data is seemingly endless.  Sorting through all of the data and making sense of it is a daunting task.  However, claim data is an excellent tool to use to refine your programs and work to improve the overall health of your organization.  Effective analysis of appropriate data will help you and your organization implement programs for overall organizational improvement.  In order to have a complete analysis, there are a few things to keep in mind.

Understand what the numbers are reporting.  Most claim data is provided as a result of ICD-9, CPT-4, NCD, and HCPCS codes.  Okay, I just said that to scare you, let’s see if we can make sense of what these types of codes mean.  A provider will code a visit as “hypertension” or “sinusitis”, it is reported to the insurance company and out comes our data.  Providers typically use these “codes” to determine what the illness, injury, or visit was for.  You would then need to work with your third party administrator (TPA) or data provider to clarify where your data is coming from and how it is reported.  The TPA can also help to refine the types of reports you are getting and review the information.  Work with them and allow them to become your interpreter.  Remember, more data may not necessarily be better data but do pick and chose those that will fairly represent the claims and give you a comprehensive picture.  Again, your TPA may help to narrow this information down to what you need for a complete picture rather than allowing the data to manifest into one of those sea monsters.

Analysis of data.  Data is just that, data.  Analyze each piece of the data and then look at the entire picture.  There are a lot of factors that impact what you are seeing so try and break it down into pieces and then review it in it’s entirety. 

• Identify the trends but be careful not to make assumptions.  An increase in OBGYN visits does not necessarily mean someone is pregnant.  Identify and watch trends.
• Secondly, look around and observe data in its totality – include a review of workers’ compensation claims, absence rates, and short and long term disability claims to get a true picture of what is happening in your organization. 

Again, data received from ICD-9 codes are great but use that data in conjunction with your prescription data to make sure you have the whole picture.  If you see a high percentage of hypertension as a diagnosis, are you also seeing a lot of prescriptions treating the same diagnosis?  Are you seeing an increase in the number of back strains?  Did that “jump” occur two months after the start up of the new assembly line?  Could it be an ergonomic issue?  Look at the numbers and then look beyond.

Reviewing statistical time frames.  Review your data by month, quarter, and year to look for trends.  Have your claims increased or decreased over the same time period in the pervious month, quarter, year, or years?  Again analysis is crucial.    Consider all of the logical explanations for a jump in the number of office visits.  Look at the time frame in which you are analyzing and make sure there aren’t obvious (or not so obvious) explanations for jumps in trends due to the season.  Did you just get out of flu season?  Are you in an organization with a lot of families who just completed back to school physicals?  Using time frames will allow you to make better interpretations of the data you are receiving.

Benchmarking.  It is very important to benchmark your data to determine where you are as compared with other companies.  Although you can compare your data with your data in previous timeframes, it is also important to benchmark your data to determine where you are as compared with other companies.  What are your trends and are they similar to the trends nationally or regionally?  How about with similar-type organizations?

National and Regional data will allow you to compare your data with that of organizations (typically of similar size) nationally or regionally.  Mercer Human Resource Consulting offers a National Survey of Employer Sponsored Health Plans that will allow you the opportunity to benchmark data.  Other organizations such as Kaiser Family Foundation (www.kff.org) and the Milliman Medical Index offer sources of data to benchmark from.      

Industry data will allow you to make good comparisons with like organizations but is sometimes difficult to capture.  Go to your national organization and see if data is captured not only for compensation and benefit costs but claim data. 

Data à Action.  Most importantly we need to turn the data that we receive and have properly analyzed and turn it into action.  Program implementation should be documented and tracked so you are able to quantify successes in terms of an ROI for the program.  Examples of such programs could include some of the following.

• Provide education about those claims you are seeing most.  For example, is your number one diagnosis hypertension?  Use payroll stuffers and articles and letters sent home to provide information on preventative care and personal health management of the disease.

• Get the family involved.  Many companies have implemented wellness programs (some including financial incentives) to their employees but we should begin to include employee’s families as well.  For example, if you are supporting a wellness initiative such as weight management or smoking cessation, it may help to build a supportive environment at home by including families in the initiative.

• Provide information about prescription drug choices.  We have already used the tactic of making “non-preferred” drugs more expensive to consumers but go beyond that.  Analyze your prescription data and if you see data telling you most employees are using preferred drugs rather than generics; educate.  Make sure employees and their families are provided good information about drug tiers, costs, and equitability in prescription choices to allow them to make an informed decision.  Face it, human resources professionals get confused about “tier one”, “tier two”, and “tier three” prescriptions…imagine how our employees feel.  Communicate and educate.

• Actively work with your provider to ensure proper implementation and use of disease or care management programs.  Work with your provider to ensure early detection of high risk / manageable illnesses and “pay the extra dollar” for disease management or case workers.  According to a recent CIGNA study (2004, Health Care Report) active programs such as these can result in a potential savings of 20% for the entire cost of the illness.  These programs allow us to show measurable ROI on the programs and continue implementation. 

• Use rewards for health management.  Reward employees and their families for completing health risk assessments.  Reward for participation in wellness activities.  Reward for self-care for minor illnesses and for lifestyle and behavior changes.  Remember, rewards don’t have to be costly.  Use creative rewards for those involved in managing their health.

Finally with data, remember that no system is perfect.  Use the information you are receiving, and look at the “big picture”.  Data analysis will allow us to make more informed decisions in our future plan designs and wellness programs in an attempt to reduce our overall health care costs.  Remember, monsters always appear larger when we are in the dark.  We can all swim easily in this sea of data as long as we work with our TPAs, providers, and/or brokers to understand what we are reading and properly analyze and implement programs to target our own monsters.

Thanks to Patti Dunham, M.A., MBA, SPHR for contributing this article.  Patti is Senior HR Consultant with Strategic Human Resources (http://www.StrategicHRinc.com).  She can be contacted at Patti@StrategicHRinc.com.