Using a Scenario-Based Hybrid Approach to Understand Participant Health Behavior

Abstract

Background:

Qualitative and mixed methods approaches are commonly used to understand participants’ interactions with real-world settings and can help health services researchers to obtain realistic details about patients’ health behaviors. However, interviews do not easily capture data about how patients perform health-related behaviors that are not part of their daily routine. A scenario-based approach is one method that can be used prospectively to explore how patients make decisions about their health-related behaviors. This approach is comprised of a set of small tailored probable circumstances with equally plausible situations, and are presented as narrative descriptions. To understand how older adults, a group at high-risk for OTC misuse, select over-the-counter (OTC) medication qualitative methods can be used.

Objectives:

This study describes a scenario-based hybrid approach that included a simulation exercise and a situational interview to understand how older adults first select and then take OTC medication.

Methods:

The scenario-based hybrid approach consisted of 1) a simulation exercise to emulate participants’ real-world experiences as they selected a medication in a store, followed by 2) a situational interview to capture how participants intended to take the medication they selected. Video recordings captured interview data as well as participants’ body language, navigation patterns, and other nuanced data that would not have been captured in audio recordings.

Results/Conclusion:

The scenario-based hybrid approach not only yielded detailed information about behavior, but also allowed investigators to discern participants’ decision-making, influences, and the rationales they use when selecting and taking OTC medications. Studies aiming to capture participants’ behavior in naturalistic situations can use these techniques to draw inferences from direct and indirect visual references that may not be captured otherwise. In this study, the goal was to understand how older adult participants select and take OTC medications. This approach allowed the research team to expediently recreate situations in which participants would purchase an OTC medication, a task that may not occur frequently and thus may not be amenable to participant observation or accurately recalled using retrospective interviewing.

Background

Innovation to improve health service delivery requires understanding how people think about and enact decisions related to health and healthcare. Increasingly, health services researchers are turning to qualitative and mixed methods in order to understand these health behaviors. ^1,2

By studying health consumers’ interactions with real-world health service delivery settings, researchers can get closer to obtaining accurate details about their health behaviors. However, conventional approaches to gathering information about these behaviors may fall short of elucidating these details. For instance, traditional interviews rely on patients’ ability to recall details about their situations or circumstances.^3,4 Although such retrospective methods produce detailed accounts of participants’ first-hand experiences, data gathered from these type of interviews can be subject to recall bias and imperfect memories ^5,6 Also, these retrospective descriptions might accurately represent participants’ thought process and actions in past situations, may not reflect what the patients would do in the future if a similar situation arises due to different thought processes in the moment. Observation techniques are also a common method for gathering information about health behaviors. Although naturalistic observations in health settings can be challenging, previous studies have leveraged these techniques to successfully capture patient experiences and behavior in community pharmacies ^7,8 However, naturalistic observation techniques require ample time to follow and observe participants. Interviews may not capture nuances in consumer behavior and observations may not be an efficient way to capture data (e.g., participant would have to contact research team when the task is being performed or researcher would have to spend time waiting for the appropriate situation to come up to then observe) so new approaches are required.

Other qualitative methods may be more suitable at elucidating health behaviors. Scenario-based approaches have been used extensively used to prospectively explore how patients make decisions.⁹ “Scenarios” are a set of small tailored probable circumstances with equally plausible future situations.^10,11 They are often presented as narrative descriptions and are designed for specific purposes. Over the years, scenario-based approach has become an important tool in foresight activities and research, making use of both, qualitative and quantitative research approaches.⁹ For most scenario-based research, the participant is asked to describe their response, action or reaction to a possible future situation, including paths of development they take to reach their decisions.¹² This ensures a systematic and structured future-oriented lens of thinking taking alternative situations into consideration.^9,12,13,14 One type of scenario-based technique, the simulation exercise, can be used to reproduce real experiences and help understand how participants solve problems and make decisions.¹⁵ During simulation exercises, the participant is placed in a manufactured, but similar situation requiring them to function in the capacity expected of them in a real event.¹⁵

Situational interviewing is an another scenario-based, data-collection technique, in which participants are presented hypothetical situations with the aim to predict an individual’s performance and behavioral intentions.¹⁶ Whereas simulation exercises ask participants to perform a task in a particular setting, situational interviewing requires a participant to respond to questions within a hypothetical scenario.

This paper will describe a hybrid scenario-based approach that uses both simulation and situational interviewing in order to capture participants’ behaviors surrounding the selecting and taking of over-the-counter (OTC) medications. In this approach, a simulation exercise was used to reveal participants’ behavioral and cognitive processes while making an OTC medication selection, whereas situational interviewing technique helped the researchers understand participants’ cognitive processes regarding how they would take an OTC medication.

Research Problem

OTC medication misuse in older adults (age 65+) is a common medication safety issue.¹⁷ For example, from the one third of older adults that take one or more OTC medication(s), 66% concurrently use OTC medications with prescription medications or dietary supplements, creating a potential for adverse drug events.^18,19 Community pharmacies, where most older adults purchase OTC medications, have the potential for decreasing medication misuse by older adults because these settings most often allow access to a pharmacist who can provide medication guidance.

The data collection methods described in this paper were developed over two studies. The first study, a pilot project, was aimed at better understanding the rate of misuse and the barriers and facilitators to medication safety that older adults encounter when making decisions surrounding OTC medication selection and use. Misuse was categorized and evaluated within four prospectively identified categories²⁰: drug-drug interactionⁱ, drug-disease interactionⁱⁱ, drug-age interactionⁱⁱⁱ, and drug-label^iv.^20,21 The authors found that there was at least one instance of potential misuse in 95% of the participants.²¹ The focus of the second study was to design, implement, and test the effectiveness of an intervention aimed at reducing OTC medication misuse in older adults.²⁰ The intervention was a new OTC section in the store dedicated to older adults and located near the pharmacy. There were two phases to this research: the first phase (beta) was to refine the intervention and implementation strategy²² and the second phase (test), was to evaluate the effectiveness of the intervention pre- and post-implementation.²⁰ The hypothesis was that a new OTC medication section stocked with a curated selection of medications that are generally safer for older adults located next to the pharmacy pick-up window would facilitate communication with a pharmacist and reduce older adult unintentional misuse of OTC medications.²⁰ The hybrid scenario-based method was used to understand how older adults select and use OTC medications pre- and post-implementation

Data collection for the pilot and second study took place at a regional chain mass-merchandise store located in the state of Wisconsin, United States of America. The pilot study was conducted at 1 store location. The main study was conducted at 4 different store locations in Wisconsin. Each store included a pharmacy, an OTC section, as well as several other mass-merchandise departments. Participants were recruited through the pharmacy from which they received their prescriptions and thus had some familiarity with the store’s lay-out. The study was approved by the Institutional Review Board at the University of Wisconsin - Madison. All participants provided informed consent prior to participation.

Description of the Hybrid Scenario-Based Approach

In the second study, the first goal was to understand how participants selected an OTC medication as they moved through a store. The second goal was to understand how participants would use the medication selected. In order to capture these two distinct goals, the research team developed a scenario-based hybrid method consisting of a simulation exercise to address goal 1 followed by a situational interview to address goal 2.

Simulation exercise

Participants were recruited by either a mailed recruitment letter (recipient list generated and letter mailed by pharmacy), word-of-mouth recruitment by pharmacist, or a posted recruitment flier in the pharmacy. Interested participants would call the research team, who would inform the participant about the study and determine if the participant was eligible for the study. Eligible participants were 65 years or older, English-speaking, and had purchased or considered purchasing an OTC medication in the past 6 months to treat either pain, an insomnia/sleep problem, a cough/cold, or allergies. This was important as participants would have had a recent experience similar to the task scenario.

Each participant met the researcher at the store entrance and signed the informed consent. The participant was presented with three conditions (cough/cold/allergy, pain, or sleep) and asked to pick the one that resonated most with them. The researcher then narrated a hypothetical scenario corresponding to the condition (see Table 1) containing a task to perform. Each scenario was carefully written and vetted by members of the research team as well as two practicing pharmacists to ensure that the scenarios were realistic and the conditions were appropriate to treat with an OTC medication. Standardized probes were built into the simulation script to aid in helping participants clarify their thoughts on their selection. Simulations generally took about 15–30 minutes.

Table 1.

Interviewer Script

Scenario
Today, I am going to describe to you one issue that is common for older adults and then I want you to show me how you would pick out a medication to treat that issue. Between sleep, pain, cough/cold or allergy, what resonates more with you?
Situation 1: Sleep Recently, you have been having (more) difficulty falling asleep or staying asleep. You are here at [store] to look for a medication that can help you sleep.
Situation 2: Pain You are having soreness and muscle aches. It is not bad enough to call your doctor. You have not taken any medication to help with these aches yet. You are here at [store] to look for a medication that can help you feel better.
Situation 3: Cough/Cold or Allergy For the past three days, you’ve had a runny nose, sore throat, felt “stuffy,” and your head is congested. You don’t have a fever and it is not bad enough to call your doctor. You have not taken any medication for your symptoms yet, but you are here at [store] to look for a medication that can help you feel better.
Now, keeping that scenario in mind, I’d like for you to show me and talk to me about how you would go about choosing a medication to help with that issue as if you had just arrived here in [store] and your goal is to find something to help. No detail is too small and there are no right or wrong answers.
Show me how you would go about choosing a medication to help you treat the problem.

Following the selection of a scenario, older adults were asked to “show and tell” how they would go about selecting an OTC medication. An audio-recorder and video-recorder were used to aid in data collection. Researchers also took notes as necessary. The researcher walked with the participant as they navigated through the store and explored, examined, and verbalized OTC choices, and then selected an OTC product. The researcher asked the participant to “think aloud” and used probes such as, “What are you thinking now?” and “Are you considering any other options here?” These probes helped elucidate the older adult’s decision-making as they selected an OTC medication. The researchers also asked specific questions about the information that participants used to make their OTC medication selection decision, probing matters like cost, previous experience with a particular medication, and label warnings to determine the specific aspects being considered by the participant.

Situational interview

The goal of the situational interview was to capture how participants intended to take the medication they selected during the simulation exercise (e.g., how many pills would they take, how often). This portion of the interview, which generally took 15–20 minutes, took place in a semi-private area located near the store pharmacy and occurred immediately following the simulation exercise.

The same researcher from the simulation exercise introduced this section by reading out the same scenario. Participants were told that questions in this section pertained to how they would take the selected medication and that it was okay to look at the medication box/bottle to answer the questions. Participants were also told that there were no right or wrong answers. The researcher asked questions about dosage, frequency of use, and if there were any safety concerns about the selected medication. Probes were used in the situational interview to gauge older adults’ rationale for their hypothetical medication use.

Video and Audio Recording of the Hybrid Scenario-Based Approach

The only difference between the pilot and second study was the technology used in the OTC selection process. During the pilot study, each interview was conducted by two people, one who served as interviewer and one as observer. Interviews were audio-recorded using a digital recorder. The observer used a Livescribe Pen and Notebook to record notes during the interview. The Livescribe system links audio to written words allowing the person to later to go back listen to the words that were being said at the same time notes are being recorded. For example, if a participant picked up two different boxes of medications and stated, “Well, these are the two I would choose between,” the observer would write down “picks up ibuprofen and acetaminophen boxes.” The observer information was then later incorporated into transcription for analysis.

During the second study, interviews were audio- and video-recorded. Participants were asked to clip the microphone to their shirt and either hold or put the recorder in their pocket. Small GoPro video cameras were used to video-record and were attached to the interviewer’s clipboard. It was least intrusive to place the GoPro on the interviewer’s clipboard so that they could direct where the camera was facing and re-direct it if a non-participant store customer or employee came into the frame. The video recording was a major source of information during the simulation exercise as it was used to record participants’ body language, navigation patterns, and other nuanced data that would not be captured in the audio recordings. Because actions and body language were not evaluated during the situational interviews, video was recorded but only audio was considered for the data analysis.

Methodological Considerations

Multiple considerations should be taken into account when using this hybrid approach in order for successful recruitment and data collection to take place.

Sensitivity to study population

In order to gain trust with participants, several strategies were used. Included in the recruitment information was the option for the participant to call the pharmacy so they could confirm that the study was legitimate. A second strategy was to send out a reminder sheet after the interview was scheduled that included a picture of the interviewer. As the first face-to-face meeting between the interviewer and participant was the day of the interview, this helped alleviate participant concerns about not being able to find the interviewer. Interviewers also called to remind participants of the interview date and time and answered any additional questions before the interview took place.

Because the population studied were older adults, it was important for the researchers to be sensitive to potential mobility and other physical limitations, such as participants in wheelchairs (taking extra time to go through the aisles), participants who didn’t bring their glasses but needed them to read the package labeling (researchers read the label for participants who would read the label but couldn’t), and participants who needed to hold onto something for balance (in these cases, shopping carts/shopping carts with scooters were used).

Study environment

The research team worked directly with the mass-merchandise store’s corporate executives, then with the specific store managers, to gain support for the study. These partnerships were established from the beginning of the project. The research team met with store managers individually so that they were aware of what was happening in their stores. Pharmacy staff were aware that they should act as they normally would in their role. Interviewers also introduced themselves to other store staff who were not directly involved with the study and answered any questions they might have to alleviate any concerns.

The store was usually busy which resulted in disturbances in the aisles during both simulation exercises and situational interviews. The private area near the pharmacy was usually noisy and sometimes made it difficult for the interviewer to hear the participant. The interviewers worked with the store to identify appropriate areas to meet participants at the store entrance to introduce the study and attach the audio recorder, and to conduct the situational interview.

Privacy protection

The use of video especially introduced privacy concerns for both non-participants in the store as well as the study participants. These were addressed through meetings with store management and through reference to the investigators’ Institutional Review Board and its video recording policies.

Scenario development

Scenarios were presented at the start of each interview and were used to frame the questions surrounding OTC medication selecting and taking. The scenarios were adapted after each phase of the study, after reviewing their relevance and scope. For example, our initial script for the second study did not specify cough/cold/allergy as a recent condition participants had. In the beta phase, we encountered conditions that were not specific (headache or migraines) or applicable (oral contrast like barium sulfate prescribed by a doctor for a CT scan) to our intervention. Although scenarios are specific to this study, they are discussed to illustrate how the presentation of a scenario can impact the focus of the interview. The researchers believe that participants should be presented with scenarios that fall within the scope of the research question as well as realistic for participants in order to obtain high-quality and relevant data.

Benefits

Interviewers were able to collect rich and nuanced data relevant to study goals. The hybrid scenario-based approach had the following advantages:

Study environment

Unlike most simulation exercises that often take place in an artificially created environment such as a laboratory, this study was conducted in a store that the participant was familiar with and had visited before. Although the scenarios presented were hypothetical in nature, by conducting the interview in the participants “home” shopping environment, observations of how they navigated and moved throughout the store more closely mimicked how they would actually conduct the task of walking into a store to select an OTC medication in real-life. Immediately following the simulation exercise was the situational interview. This interview allowed the interviewer to delve into how participants would take the OTC medication they selected. By reminding the participants about the scenario and summarizing the selection process, it was easier for participants to recall why they selected the product and how they would take the product.

Use of video and audio recordings to collect data

The unobtrusive use of recording technology allowed the researchers to collect detailed data. The Livescribe method used during the pilot was a lower-cost solution than the video recording and would work well for researchers who are working in areas where video-recording is not possible such as healthcare facilities where it may be difficult to not record other patients or healthcare professionals on camera. The GoPro method was successful in capturing a level of detail that could not have been achieved by note-taking. For instance, use of the GoPro allowed the researchers to evaluate wayfinding by recording how participants navigated the store to the OTC medication aisle as well as to the shelf where the selected OTC medication was located. The video recorded how participants interacted with the environment, including challenges to find what they were looking for and difficulties with bending down for products on low shelves or reaching products on high shelves.

The GoPro also allowed identification of what products older adults referred to (with and without touching) during their selection process. Also recorded was the deliberation process as older adults told the interviewer about the products they have taken in the past, would consider taking, and ultimately select to treat their symptoms. An additional benefit of using the GoPro is that is allowed for interviews to be conducted by just one person.

By audio-recording the interview, interviewers were able to focus on the selection thought process instead of recording responses by hand. This allowed the research team to glean the rationale behind participants’ selection to discern their motivations and rationales. Audio recordings also helped corroborate video data.

Limitations

Because the interviews were conducted the study in a “real-world” environment, researchers also became aware of several limitations to the approach.

Concern over interrupting workflow

One goal of the study intervention was to improve communication between older adults and pharmacy staff. During the interview, participants were encouraged to follow their usual course of navigation, selection, and decision-making in the store. This included asking pharmacy or staff questions if that was what they would normally do. However, there were times where some participants reported that they would usually speak to a pharmacist as their next step, but were hesitant to do so during the interview because it was not a “real” question and they did not want to bother staff if they looked busy. Although some participants did still approach the staff with questions, in other cases they were uncomfortable doing so, thus the research was not always able to observe the participant’s natural path and process.

Accuracy of responses

Researchers used probing techniques to understand participants’ thoughts on other medications on the shelf, e.g., “I see you selected this medication. Is there a reason you selected this medication instead of others on the shelf?” This sometimes led to the interruption of the participant’s usual decision-making process.

The selection process was conducted in a setting where medication selection would take place. However, due to study design logistics, medication-taking questions were asked in the pharmacy instead of in the home, the setting where medication is taken. Although following the simulation exercise with the situational interview made it easier for participants to recall the scenario, participant responses may not accurately indicate how participants would actually take OTC medications at home given these decisions are often not made until they get home and begin to take the medication. At home, participants may be influenced by their medication stock and other family members.

Analysis

The overall study with the Senior Section implementation has so far generated multiple publications describing the development and the effects of the intervention. By combining the scenario-based hybrid approach with additional methodologies, several rich analyses were conducted which include:

1. An evaluation of the effectiveness of the intervention on OTC medication misuse in older adults. A logistic regression model was used to understand association of patient characteristics for pre/post intervention with misuse outcomes.¹⁸

2. An in-depth qualitative analysis was conducted to understand how older adults’ knowledge, beliefs, and attitudes inform their decision-making regarding OTC use.²³ The rich responses brought to light older adult motivations for following OTC label instructions, strategies in case an OTC medication is not effective, as well as medication use and safety concerns.

3. A preliminary persona analysis classifying different types of older adult behaviors while purchasing OTC medications in the pharmacy.²⁴ An expanded analysis will involve identification, operationalization, and in-depth description of different OTC selection and use behavior and how their behavior impacts OTC medication misuse.

4. A cognitive task analysis to conceptually map older adults’ cognitive decision-making process as they navigate in the community pharmacy to select OTC medications (Submitted).

Conclusion

The hybrid approach that was described in this paper can be used to qualitatively answer a number of research questions. Studies aiming to capture participants’ behavior in naturalistic situations can utilize this technique to draw inferences on the basis of body language, tone of responses, indirect visual references, for example. In this study, the goal was to understand how older adult participants select and take OTC medications. The hybrid scenario-based approach developed allowed the research team to expediently recreate situations in which participants would purchase an OTC, a task that may not occur too often and thus may not be amenable to participant observation or accurately recalled using retrospective interviewing. Other research that seeks to investigate decision-making as they relate to health behaviors may benefit from such simulation exercises.