Potential Health Claims Data

In order to assess health care expenditures, individual-level health claims data should be obtained and aggregated from the health plan provider(s). Specifically, the following information for each enrollee would be valuable:

• Age
• Sex
• Employment status – employee, spouse, dependent, or retiree
• Total health care costs
• Diagnosis codes (use first or second code)
• Procedure codes (for preventive services use)
• Place of Service

Health care expenditures analysis should include health claims data for employees and employee dependents. Analyzing the data for employees and their dependents separately will assist in selecting the target audience of the health promotion program. Health claims data for dependents can illuminate the health issues that contribute to an employee’s absenteeism and time off. Issues concerning an employee’s family can be costly.

Additional stratification of results by sex and age groups will also be helpful in identifying specific demographic groups with higher health care costs. Similarly, it is important to stratify results by place of service (e.g., Emergency Room [ER], inpatient services, outpatient services, etc.). This can both demonstrate the types of issues employees face as well as the cost of these services. For instance, high costs from ER services each year may indicate an opportunity for preventive care policies and programs.  

In order to identify major sources for health care expenditures, it is sufficient to look at the health care cost data from one year. However, depending on the sample size, more stable estimates would be obtained by combining health care claims data for 24 or more months.

When reporting the results, it is important to clarify the time period for which the costs are reported (e.g., one or two-year period). It is also important to recognize employee retention and turnover rates when performing annual evaluations. For a company with a low turnover rate, data will likely remain the same and the population the team is evaluating will likely be the same population to receive the program at the workplace. However, for a company with a high turnover rate, it may be useful to focus on an evaluation for those employees considered “stable” – or likely to stay with the company for a longer period of time.

Data collection and evaluation should focus on the employees that will eventually benefit from policy and programmatic changes. For a company experiencing high turnover rates, an evaluation may help address the turnover issue, as data collection may illustrate the reasons for high turnover and low employee retention. Furthermore, an evaluation and potential changes may be seen as an “investment in the company’s employees”, perhaps increasing employees willingness to stay with the company.

In organizations where multiple health plans are offered to employees, the data from multiple data sources should be combined and analyzed together. Analyzing the data separately will most likely skew the results if the health plans are very different in what they cover. For example, sicker people may choose a plan that covers more services (even if its premiums are higher) while healthier people will choose a plan with less services and lower premiums.  

It is important to recognize that health care claims information will come from the company’s health care plan providers which takes time to receive and may not come in a form that is ready to use or contain the necessary level of detail. This is another place where a company liaison can be helpful in obtaining the necessary information to conduct the analysis.