FDA self-selection studies are critical to the regulatory evaluation process, particularly for over-the-counter (OTC) drugs and direct-to-consumer (DTC) medical devices. These studies evaluate whether typical users can accurately choose if a health product is right for them (and their health needs) based on the labeling and without the help of a healthcare professional. The label may be the outer package found on a shelf in a drug store or, for e-commerce health products, information provided on the website.
The U.S. Food and Drug Administration (FDA) issued guidance in 2013 titled Self-Selection Studies for Nonprescription Drug Products, which is an excellent reference for those developing and conducting self-selection studies.
The choice to purchase and use a non-prescription drug or medical device is in the user’s hands. Self-selection studies reduce the risk of misuse (promoting safety) and aim to ensure that consumers can make informed decisions about their health purchases. They are particularly relevant for products with contraindications (e.g. not for people with diabetes), specific use criteria (e.g. a fertility device for females), or complex instructions that could lead to adverse reactions if misunderstood.
Self-selection studies are generally conducted as part of a 510(k) premarket notification for over-the-counter products, including medications that intend to switch from prescription to OTC. While the FDA’s guidance from 2013 is written for non-prescription drugs, it can be adapted and applied to Class II medical devices sold to consumers on shelves or online or to prescription drug products when requested by the FDA.
If you want to learn more about the self-selection study process with some insider tips about FDA self-selection studies guidance, read on for a deeper look at how to make the most of this process.
Study Objectives
The primary objective of self-selection studies is to assess whether people, after reading the product label, can make a correct decision about using or not using a health product based on their own (or, in some cases, hypothetical) medical condition(s).
Start by defining your communication objectives.
The communication objectives are the key concepts the consumer needs to understand to decide: 1) whether to use a health product and 2) when and how to use it properly (i.e., safely and effectively). These objectives help define the content that must be covered in the labeling and lay the foundation for the self-selection study protocol.
Communication objectives usually include 1) understanding the purpose of the drug or device, 2) understanding the contraindications (i.e., who the product is not safe for), and 3) understanding when someone would or would not use the device.
The communication objectives focus your efforts in the self-selection study. Other content unrelated to the communication objectives is less relevant to the study process.
Study Population
Whenever possible, self-selection studies are conducted with the typical users of the drug or device instead of the general population.
Define your study sample.
In self-selection studies, selecting an appropriate population is essential since the results should reflect how a diverse group of typical consumers will interpret and action product labeling.
The criteria for inclusion should capture individuals in the product’s target demographic. Determine your target population and any subgroups of interest. You may outline your inclusion criteria based on age, sex, health conditions, medication use, and relevant lifestyle factors (i.e., smokers).
Consider including both those who should and should not use the product. Exclusion criteria should be minimal because OTC health products are available to anyone on-shelf or online.
Recruiting subgroups with uncommon characteristics, such as people affected by certain health conditions, can be tricky. Conducting a virtual study can broaden the geographic reach of your recruitment and increase success in getting your target population. Another option, if recruiting for a specific characteristic is challenging, is to recruit a broader typical user (e.g. adult females) and ask your self-selection questions based on hypothetical scenarios.
For an unbiased sample, participants should be naive to the study product, not experts regarding the product type, and have no affiliation with the study sponsor. Ensuring diversity of your sample based on race/ethnicity, education level, and geographic location is critical to testing the product across a representative population. Your recruitment screener can collect these important data points to ensure a naive and diverse sample.
Including participants with lower literacy skills in these studies is essential to ensure the product labeling is accessible and understandable to users with diverse literacy levels. Assessing literacy levels can be done using validated tools like the short version of the Test of Functional Health Literacy in Adults (S-TOFHLA) or the Rapid Estimate of Adult Literacy in Medicine (REALM) test.
The sample size should be justifiable.
The sample size should provide a statistically sufficient number of cases to assess self-selection decisions across the study population. The sample size should consider elements such as the clinical importance of incorrect responses and estimates of correct responses in pilot testing or similar studies.
Statistical Considerations and Data Analysis
When designing and analyzing self-selection studies, consider the statistical factors which may impact your data’s overall success and effectiveness. Here are some statistical considerations to factor into your study design and analysis.
Determine your primary endpoints.
The primary endpoint for these studies is usually the proportion of participants who make a correct self-selection decision based on the product label. Correct choices depend on the label’s key elements (e.g. contraindications, age, health conditions) and may involve a single or multiple factors.
For example, a product may be contraindicated for anyone taking blood thinners (i.e., a single factor). Or, a product may be indicated for adult females who are menstruating (i.e., multiple factors).
There are different ways to evaluate self-selection. Some standard primary endpoints include:
- Correct decisions made across the entire population
- Correct decisions among those choosing to use the health product
- Correct decisions among those who should not use the health product
Define your success criteria.
Success in your study is defined as reaching a predefined target for correct self-selection (i.e., percent of endpoints demonstrated correctly). Someone with statistical expertise can help determine the target. It should also have a clinical rationale and account for variability in responses.
Take time to develop a detailed study protocol.
The study protocol must specify correct decisions, acceptable mitigating factors, and the target success criteria. This must be documented before the study is conducted. Mitigating factors are acceptable reasons for an incorrect response–they should be defined before the research and detailed in the data at the subject level.
For example, a participant with a contraindication to the product might say, “I would talk to my doctor before deciding whether to use the product,” instead of making a correct self-selection decision not to use the product, and this response may be deemed acceptable.
The study will need enough participants to address the primary objective reliably and meet the success criteria.
Plan your data analysis before the study begins.
A detailed statistical analysis plan should be designed with the help of experts in data science and statistics and outlined in the protocol prior to study implementation. It’s important to include methods for handling missing data and analyzing failed-to-complete cases.
Questionnaire Design
Designing your self-selection questionnaire can involve qualitative and quantitative data collection. Qualitative questions often explore participants’ thought processes and reasoning behind their self-selection decisions, while quantitative questions compute the accuracy of those decisions based on predefined criteria.
The initial self-selection question is usually, “Is it okay for you to use this health product?” This is followed by a non-leading question to explore their response, such as, “Why do you say that?” Avoid leading questions. Further open-ended questions can help provide additional information for analysis.
Questions about the participant’s medical history relevant to the decision to use the health product should be asked after completing the self-selection questionnaire. This is so you don’t introduce bias before conducting the study. For example, suppose the product is contraindicated for people with diabetes. In that case, you don’t want to ask if they have diabetes just before launching the self-selection questionnaire, as that may prompt them to focus on labeling elements related to diabetes.
Scenario-based questions should be straightforward and optimized based on user feedback.
Sometimes, hypothetical scenarios may be needed to address specific and uncommon contraindications. Scenarios should be simple and realistic, followed by “Is it okay for the person in the scenario to use the health product? Why or why not?”
Pilot test your scenario-based questions with a group of participants, including those with lower literary levels, and revise the questions as needed based on participant feedback. This user-centred approach to questionnaire design will help avoid incorrect responses due to bad questions versus genuine label comprehension issues.
Consider having the FDA review your key label elements, study design/protocol, and questionnaire before launching your study. While this might set you back on your launch date, it can be game-changing to your study’s success.
Study Conduct
The self-selection study should reflect how consumers will interact with a product in real-world settings. Multiple strategies can be used. For example, studies can be conducted in person or remotely through moderated live-televideo interviews or unmoderated and self-administered using an online label and questionnaire.
Choose the best study design to meet your needs and promote success.
Your study design will depend on various factors. Suppose you need a clinician to provide a physical validation assessment to verify specific symptoms or clinical features (e.g., the presence of skin findings, like nevi). In that case, you will need to conduct an in-person study. If you need to recruit from a wide sample of participants, a remote study may be more appropriate to get the sample you need.
During the study, participants must make self-selection decisions independently without assistance or coaching to provide accurate results. As such, when moderated, interviewers should avoid leading questions or language that may influence participants’ decisions. Standardized scripts and neutral questioning can help minimize this type of bias.
Data Collection and Recording
When recording and analyzing qualitative responses from self-selection studies, responses should be recorded word-for-word to avoid misinterpretation. To accurately analyze open-ended questions, you can categorize and code responses systematically to identify common themes. The process for classifying and coding qualitative responses should be defined before data collection and documented in the protocol. The quantitative data elements to be recorded and analyzed should also be specified beforehand.
Final Report
The final report for self-selection studies is a document that should go over all aspects of the study, from design to results. Key components include:
- Detailed description of study design, conduct, and result interpretation
- Study objectives, endpoints, and success criteria
- Sample design, recruitment efforts, and response rates
- Details on non-responders and incomplete cases
- Study methodology
- Study results, including self-selection rates and rates for key subgroups (e.g., by literacy, sex, age, race, and risk factors)
- Interpretation of results based on pre-defined success criteria and statistical plan
When presenting findings, use quantitative and verbatim open-ended responses to analyze self-selection decisions comprehensively. This may include the percentage of correct self-selection decisions and root cause evaluations for the incorrect decisions. From there, you can best interpret the results and assess the implications if a product is approved as-is or the labeling may require changes.
Our high five:
At SoundRocket, we have experience designing and implementing self-selection studies. Here are our top five tips for conducting a successful self-selection study.
- If you’re preparing a 510(k) submission with an FDA-mandated self-selection study, these studies take time to plan and execute. Get started early.
- Consult experts on how to plan and design a self-selection study for your health product.
- Your self-selection study relies on the quality of your label. Use experts in consumer-facing health content to develop your label and its design. Pilot test your labeling with typical users to assess and optimize your health product before the study.
- Document your plans from recruitment to reporting in a protocol before launching your study. Consider running this by the FDA before you launch your study. We recommend running a small pilot study as a trial (to identify any pain points) before you kick off the full implementation.
- Be ready to learn about labeling changes that may be revealed by a self-selection study, and know that incorporating these changes will improve the overall product.
Self-selection studies are critical to ensuring that consumer-facing health product labeling and other key informational materials support people in making safe and effective healthy product choices independently.