Investigating the effects of an employee wellness coaching intervention on patient engagement and healthcare costs

Abstract

Responding to the burden of rising healthcare costs, employers have turned to wellness coaching, a holistic behavioral intervention designed to help clients attain wellness-promoting goals across a range of areas. This study was conducted to test the effects of a wellness coaching intervention on indicators of patient engagement and healthcare costs. Through the use of propensity scoring techniques, employees in a variety of large organizations throughout a local community in the Midwest who participated in a wellness coaching intervention were randomly matched with noncoached participants on key characteristics to create a meaningful comparison group to test treatment effects. Medical claims data from 300 coached participants were compared with 964 noncoached employees. Differences in current and projected healthcare costs and indicators of patient engagement with the healthcare system were examined. Results revealed that coached participants saw an increase in current and projected healthcare costs and indicators of patient engagement. Indicators of patient engagement mediated the relationship between coaching participation and healthcare cost outcomes. Over time, current and projected healthcare costs decreased whereas indicators of patient engagement remained the same. It appears that wellness coaching is a beneficial intervention that may save employers money on healthcare expenditures down the road.

Keywords:

• wellness coaching
• workplace behavioral health intervention
• patient engagement
• medical claims data
• healthcare costs

Notes

1 One hundred and twenty-four coached participants were removed from the data set because they began the coaching intervention before their first medical claims data were recorded. In other words, they lacked true baseline data. In addition, 36 coached and noncoached participants were removed because they only contained 1 month of medical claims data, hence, they did not have pre- and post-treatment data. This resulted in a total of 13,739 untreated participants and 323 treated participants.

Additional information

Funding

This study was funded in part by an award from the Regenstrief Center for Healthcare Engineering, Discovery Park, Purdue University. The funding source facilitated access to the de-identified data set used in this study, but played no role in the analysis or interpretation of the data.