This fact is I think clearly shown in a paper that I wrote with Professor Peter Ubel and post-doctoral fellow Andrea Angott of Duke University that is forthcoming in the journal Breast Cancer Research and Treatment. (For a summary, see the abstract, press release, and a podcast I recorded about the study.)
In short, we presented women with a hypothetical decision about different combinations of treatments that can be taken after breast cancer surgery to reduce the chance that cancer comes back. Some women learned about the 4 main options (hormonal therapy, chemotherapy, both, or neither) and received survival risk information about each of the options all at once, as is customary in most doctor-patient communications on this issue. Within this group, those women with low numeracy skills (i.e., they have a hard time working with numbers), found it impossible to distinguish between situations where chemotherapy offered large versus small risk reductions.
Other women in the study, however, received the exact same information in smaller doses. These women made two simpler choices: first, a choice about hormonal therapy, and then a choice about whether to add chemotherapy to hormonal therapy. We found that all of the women in this group, even those with low numeracy skills, were able to know when the treatment offered large versus small benefits.
Four options doesn’t seem like much. It’s not like we were asking these women to choose among 15 different combinations. Yet, our research showed that it clearly was too much for many women.
This has lots of implications for many types of health communications. For example:
- Doctors talking to a patient about different approaches for managing chronic conditions often must discuss combinations of approaches. For example…
- Depression can be managed through medications, counseling, or both.
- Lower back pain can be addressed through surgery, pain medications, physical therapy, or any combination of these.
- Industries, communities, and/or individuals affected by toxic exposures often have multiple possible mitigation options to choose between.
What our study suggests is that the most effective way to present complex risk information about multiple possible concurrent options is to use a simple algorithm:
- To the best degree possible, order the options by benefit-cost ratio.
- Present the decision maker with the single behavior/option that has the best benefit-cost tradeoff. Let them learn about and make their choice about this first option by itself.
- Present the second option after recalculating comparison risk statistics to use the decision maker’s prior choice as the new baseline. Thus, if the decision maker chooses to do the first option, assume all of the risk reduction or other benefit from that option as now given and show only what the new option adds or subtracts.
- Continue until the decision maker finds the tradeoff unacceptable.
This simple approach does two things. First, it reduces the amount of risk information someone has to consider at one time.
Second, and more importantly, it allows the incremental benefit of each subsequent option to be calculated and considered by the decision maker independent of all of the others. In many situations (such as the one we studied in the paper), the incremental benefit of the second option to be considered is quite small, a fact that becomes clear only once the effect of the first option is separated out of the total risk reduction.
When people face important risks, whether about medical treatments, environmental exposures, or anything else, there is a natural tendency to want more. More data, more options. The problem is that more data and choices can easily become too much to handle.
What Ellen Peters has previously shown is that people do best with extremely focused presentations of the most decision-critical risk information. What we showed here is that a key way to accomplish this goal is to provide a decision making process that allows them to consider each choice separately rather than having to swim through a tidal wave of choices.
Because learning less risk information at once really can be more helpful.
Brian J. Zikmund-Fisher is an Assistant Professor of Health Behavior & Health Education at the University of Michigan School of Public Health and a member of the University of Michigan Risk Science Center and the Center for Bioethics and Social Sciences in Medicine. He specializes in risk communication to inform health and medical decision making.