Assessing the Food and Drug Administration’s Risk-Based Framework for Software Precertification With Top Health Apps in the United States: Quality Improvement Study
Bibliographic record
Abstract
BACKGROUND: As the development of mobile health apps continues to accelerate, the need to implement a framework that can standardize the categorization of these apps to allow for efficient yet robust regulation is growing. However, regulators and researchers are faced with numerous challenges, as apps have a wide variety of features, constant updates, and fluid use cases for consumers. As past regulatory efforts have failed to match the rapid innovation of these apps, the US Food and Drug Administration (FDA) has proposed that the Software Precertification (Pre-Cert) Program and a new risk-based framework could be the solution. OBJECTIVE: This study aims to determine whether the risk-based framework proposed by the FDA's Pre-Cert Program could standardize categorization of top health apps in the United States. METHODS: In this quality improvement study during summer 2019, the top 10 apps for 6 disease conditions (addiction, anxiety, depression, diabetes, high blood pressure, and schizophrenia) in Apple iTunes and Android Google Play Store in the United States were classified using the FDA's risk-based framework. Data on the presence of well-defined app features, user engagement methods, popularity metrics, medical claims, and scientific backing were collected. RESULTS: The FDA's risk-based framework classifies an app's risk by the disease condition it targets and what information that app provides. Of the 120 apps tested, 95 apps were categorized as targeting a nonserious health condition, whereas only 7 were categorized as targeting a serious condition and 18 were categorized as targeting a critical condition. As the majority of apps targeted a nonserious condition, their risk categorization was largely determined by the information they provided. The apps that were assessed as not requiring FDA review were more likely to be associated with the integration of external devices than those assessed as requiring FDA review (15/58, 26% vs 5/62, 8%; P=.03) and health information collection (24/58, 41% vs 9/62, 15%; P=.008). Apps exempt from the review were less likely to offer health information (25/58, 43% vs 45/62, 72%; P<.001), to connect users with professional care (7/58, 12% vs 14/62, 23%; P=.04), and to include an intervention (8/58, 14% vs 35/62, 55%; P<.001). CONCLUSIONS: The FDA's risk-based framework has the potential to improve the efficiency of the regulatory review process for health apps. However, we were unable to identify a standard measure that differentiated apps requiring regulatory review from those that would not. Apps exempt from the review also carried concerns regarding privacy and data security. Before the framework is used to assess the need for a formal review of digital health tools, further research and regulatory guidance are needed to ensure that the Pre-Cert Program operates in the greatest interest of public health.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.293 | 0.419 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.004 |
| Bibliometrics | 0.010 | 0.009 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.005 | 0.005 |
| Open science | 0.003 | 0.005 |
| Research integrity | 0.002 | 0.004 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".