Unless you’ve cancelled your newspaper subscription and listen exclusively to music on your way in to work, it’s almost impossible to have missed the recent flap over new recommendations around the age at which women should begin to receive regular mammograms. These new guidelines, issued by the U.S. Preventive Services Task Force, have raised a firestorm of criticism in recommending that women begin to receive regular mammograms at age 50, instead of age 40 as had been espoused previously by the group. A number of groups, including the American Cancer Society, have not endorsed the recommendations and continue to support the position that such screening should begin at age 40 for all women.
The implications of the findings and various positions taken on the controversy have been well covered and are not the subject of this post. I did, however, find an interesting article in Monday’s New York Times talking about the nature of the data itself, and the process that the Task Force went through in collecting it, that I think provides an interesting take and has implications for how HR uses data in Workforce Planning and Analytics. The article, Behind Cancer Guidelines, Quest for Data , describes the efforts of the Task Force to leverage the latest data and analytic techniques. In so doing, it hoped to clarify the often confusing and contradictory guidance given to women trying to make informed decisions about when to receive routine screening for breast cancer.
Instead, the group opened its work up to criticism not because of the data itself (which has not been widely challenged), but thanks to its interpretation of it and the actionable recommendations it made in conjunction with the findings. There have also been complaints about the reasoning behind the recommendation, particularly with regards to the perceived negative impact of a “false positive” test in needlessly provoking anxiety and stress in those women who receive such a result. Some feel that the amount of weight given to this kind of emotional reaction suggests the inability of women to adequately ‘handle’ what turns out to be a “false positive” result.
So what are the implications for HR practitioners trying to leverage their workforce data to support strategic decision making at their organizations? The travails of the U.S. Preventative Services Task Force provide us with a few key points to consider in using HR data:
The best data and analysis tools are not enough: The task force thought its work would be pretty straight forward. No guidelines had been issued since 2002, and in the years since new scientific and analytical techniques had been developed to improve the analysis of data. Similarly, many companies now have much broader and deeper HR data than they have ever had before, with more sophisticated tools to analyze it. However, as we see here, this is not enough to ensure success in guiding critical decisions in an organization.
The environment in which data is presented is critical: The Task Force presented its findings smack dab in the middle of a country-wide battle over healthcare reform, along with its implicit discussions around controlling costs and the rationing of care. Interest groups from across the political spectrum reacted quickly and loudly to the findings. HR data, particularly hard hitting information that challenges assumptions and rebuts myths, will often engender a similar reaction when released in an organization. Always know the different interests that will be impacted by your findings, and anticipate their reaction to them.
Don’t assume your interpretation of the data will be shared by everyone else: As noted earlier, one of the reasons the task force recommended the change was due to its perception of the damage done by “false positive” tests in which women without cancer were told that they might, in fact, have it after all. While the incidence of false positives is not in dispute (10% for women in their 40’s, with 1% leading to unnecessary biopsies), the Task Force’s interpretation that this justified changing the age of routine screening from 40 to 50 most definitely was. For HR practitioners telling a story with data, vet your interpretations with others (preferably outside of HR), before drawing your final conclusions. The way you see the world (and same piece of data) may be very different than the way it is viewed by the line or the C-Suite.
Whether the Task Force is “right” or “wrong” in this case is not the point. Instead, the story is that data and analysis can provide clarity but not certainty. When developing HR strategy or forecasting future workforce needs, the latest data and tools can be a great help, but ultimately business leaders will have to interpret what it means and decide how to react. As we teach in our Workforce Planning workshop: “Mathematical analysis and collections of massive amounts of data may be useful, but ultimately, the determination of “how we should be staffed” is a human decision.” Ultimately, someone has to make the call, and data alone won’t get you there.
Tags: HR data, Predicitve Modelling