By Dr. Sarah Watters, Behavioral Scientist

‍

Let's have a quick chat about surveys.

‍

You can't just ask a question. 

‍

Well, you can. But one can also just give an answer.

‍

In healthcare, where important questions with important consequences are frequently posed, whether for a health risk assessment, a post-acute stay questionnaire, or a member experience survey, it's critical that we get accurate, well-thought-out answers. 

‍

We want answers that can help inform decisions to be made. 

‍

Unfortunately, so often in healthcare—stressed for resources and time—the way we go about asking questions to better understand our patients' needs and circumstances is riddled with biases. As a result, decisions to be made—big or small—are informed by just an answer and can lead us astray: spending in the wrong places or prioritizing the wrong initiatives, for example. 

‍

Several common biases can affect responses and can broadly be categorized by:

‍

1. When the questions are asked 

• Is the member being called out of the blue to answer a (probably long!) list of questions? If so, they may be unfocused, preoccupied and want to get through the call with haste (can you blame them?). This is likely to result in the acquiescence bias, which is the tendency to agree with all "Yes/No" or "Agree/Disagree" questions.
• Asking essential questions before or at the end of an appointment may also compromise the quality of responses. Before the visit, the member may feel apprehensive about today's appointment and provide what they deem as socially acceptable responses. At the conclusion of the visit, the member may be tempted to give answers that are different from reality based on an update they've just received about their health.

‍

2. Who is asking the questions?

• Is it someone to whom the member is trying to provide a good impression?
• Is it someone with whom the member is unfamiliar?
• Has the person asking the question insinuated that there is a 'right' response? 
• As above, social desirability bias has a strong pull across all of our lives but poses a particular challenge when it comes to addressing health concerns. Particularly those that are less talked about, like uncomfortable medication side effects or clarifying questions that might appear to be 'too basic' to want to ask. 

‍

3. Why are the questions being asked?

• In the absence of context and understanding how and why their responses matter, members may be more likely to adopt extreme responding patterns (e.g. answering the same option for each question, similar to the acquiescence bias above, or end aversion, consistently responding with 'middle of the road' options).
• The member may also be less compelled to be thoughtful in their response and lean on the most salient—and likely recent—experience they can recall, illustrating a form of availability bias.

‍

So, how can we combat these biases, get beyond just an answer, and truly tap into how our members feel? 

‍

Informative and accurate responses require attention, meaning the respondent isn't worried about how many more questions there are, or who is waiting or listening. 

‍

Wellth's daily digital touchpoints cover these bases in 3 key ways:

‍

1. We have their attention!

• No one is calling on the phone while the member is preparing dinner, taking care of others, or even watching Netflix!
• We don't need to find a 'good time' because we have the time already carved out.
• By leveraging digital connection, there's no concern over loud environments, poor connections, or difficulty hearing. 

‍

2. Steady drip

• Instead of having members answer all questions at once, which results in fatigue and jeopardizes response quality, we can include a few each day to make the experience more enjoyable.
• With the member knowing they're not set up for a laundry list of questions, we also curb response patterns like speeding (rapidly selecting a response without much thought) and end aversion (consistently providing middle-of-the-road responses), which compromise response quality and validity. 

‍

3. No judgment

• Removing the need to be on the phone or interacting in person lends itself to a less judgmental response environment and higher-quality answers.
• By embedding questions within the Wellth app, we bookend questions with relevant information explaining why and how information is being used. As in most industries, transparency is key; it builds trust and compels the member to be forthcoming.
• We can also complete a tight, positive feedback loop, thanking the member for their responses and increasing the likelihood that they'll respond (truthfully) in the future.

‍

Outreach to patients is costly from both a time and resource perspective. Some studies estimate manual outreach to cost around $45 per care gap closed, though numbers may be far higher depending on the population of interest. Taking into consideration labor costs, time, resources, and efforts across not only HRAs but CAHPS and mock CAHPS pushes outreach costs to the hundreds of thousands or even millions. Gagliari et al. (2020) note that healthcare organizations can spend "many hours fruitlessly trying to contact patients by phone and leaving messages when patients did not answer calls."

‍

With that in mind, we should be doing everything possible to ensure that the value elicited from this outreach is maximized—and, in the case of surveys, valid. When we get just an answer to a question, decisions can be poorly informed and expose us to tangible and intangible costs down the road. 

‍

The first step toward better understanding patient needs and receiving honest, valuable feedback is asking the right questions at the right time through the right channels. Wellth's daily touchpoints provide an opportunity to elicit high-quality responses from members, removing many of the most prominent response biases that plague typical survey methods used by healthcare professionals today. 

‍

Have a question? We'd love to answer (or ask) it! 

‍