There is an article about the wellness program will help improve employees on many levels. This wellness program includes mental health resources, fitness incentives and nutrition coaching.  This program will increased employee productivity by 25%.  The employees were 40% less likely to take sick days.  Employee satisfaction levels raised to 88%.  Best of all, this program paid for itself within six months.  These claims sound fantastic at first, don’t they? 

The reality of the situation is that the Workshop tried this.  We are a medium sized toy manufacturer with a contract to ship as many Etch-Drawers as possible before the Holiday rush.  Many of us participate in the wellness program, and I wanted to let you know the real story.

They claim that average productivity before the wellness program was 80 Etch-Drawers and after the wellness program it was 100 units a week.  So 80 Etch-Drawers with an increase in 25% productivity is: (100 – 80) /80 *100 = 25%   What they don’t take into consideration is that the wellness program took place before our big holiday rush.  That means, after we participated in the wellness program, our deadlines were much stricter because we had to ship Etch-Drawers for the Christmas rush.  Another thing I wanted to mention is that some of the resources occurred during the workday.  I remember one day when I had to take a break from toy making to see a nutritionist and work out on the treadmill.  My numbers were so low that day.  Oh, that graphic in the article had a truncated y-axis.  It made productivity look so exaggerated.  It wasn’t technically lying, but it was definitely deceptive. 

The statistics mentioned that our employee satisfaction rose to 88% when they surveyed 75 employees with a margin of error of ±8%.  That means it could be anywhere from 80% to 96%.   The truth is that most of us had satisfaction ratings around the same level, maybe higher.  No one ever measured it before the wellness program started. Also, between you and me, this blogger thinks that they didn’t actually ask workers in every department, every work shift (day, evening and night shift) and at each role at the company.  It is hard to say because it is not addressed in the article.    

For sick days, 6% of the participants in the wellness program took 5 or more sick days.  10% of the non-participants in the wellness program took 5 or more sick days.  The relative decrease is calculated  (10 – 6)/10 * 100 = 40%.  That means participants were 40% less likely to take 5 or more sick days.  These percentages represent aggregated data.  They combine all participants and non participants into only two categories without showing differences in job title, shift, starting health conditions and departments. Aggregation can hide variations in the groups making it hard to know if the program really works.  The absolute risk reduction is 10% – 6% or 4% decrease in employees taking 5 or more sick days.   

I did want to point out, that this program was presented as a voluntary opportunity.  I was able to participate, but coworkers that were taking care of parents, children or coworkers that worked second jobs could not afford to take time away from work.   This raises a broader impact issue, as the program  may have helped employees who had the flexibility to participate while excluding those that had other commitments like family or another job.

A better way to phrase the article should have been. 

This wellness program offers lots of resources that could help employees.  It shows promise with satisfaction, sick days and productivity.  It is difficult to determine since the data is affected, missing baselines, participant level and methodology were not clearly reported.

I walked away wondering who the initial LinkedIn post really benefited. It appears that the wellness company gained the most from the positive data. Companies have a responsibility to present statistics transparently, baseline comparisons, and participation differences, so readers can make informed judgments.