JMIR Publications

ABSTRACT

Background:

The adoption of high-quality health apps has been slow, despite a myriad of benefits associated with their use. This is partly due to users’ concerns regarding the effectiveness, safety, and data privacy of such apps. Quality assessments with robust and transparent criteria can address these concerns and, thereby, encourage the utilisation of high-quality apps. However, a major challenge for such assessments is reaching a scale at which a substantial proportion of the over 350,000 available health apps can be evaluated.

Objective:

To support the scaling of health app quality assessments, the current study aimed to examine the preferences and willingness to pay for assessments with different value propositions among potential customers.

Methods:

Two discrete choice experiments were conducted with, respectively, 41 health app developers and 46 health system representatives (from healthcare institutions, authorities and insurers) from across Europe. Mixed logit models were applied to examine the impact of assessment attributes on participants’ choices, as well as to calculate marginal willingness to pay and predicted assessment uptake.

Results:

Among health app developers, the attributes with the largest impact on assessment choices were the associated clinical care uptake (integration into clinical guidelines and reimbursement or procurement), as well as cost (purchase price). Increased willingness to use assessed apps and app store integration of assessment results had a moderate impact on choices, while required developer time investment and time until assessment results become available made the smallest contribution. Among health system representatives, increased willingness of clinicians and patients to use evaluated apps had the greatest impact on assessment choices, followed by cost. Time until assessment result availability and percent of peers recommending the assessment made a moderate contribution, and re-assessment frequency had the smallest impact on choices. On average, health app developers were willing to pay an additional 9,020€ (95% CI 4,968€ – 13,072€) if an assessment yields guideline integration and procurement or reimbursement, while health system representatives were, on average, willing to pay 7,037€ (95% CI 4,267€ – 9,806€) more if an assessment results in a large, rather than small, increase in willingness to use apps. The predicted uptake of assessments that offer the preferred values for all attributes was 88.6% among app developers and 91.1% among health system representatives.

Conclusions:

These findings indicate that, in order to maximise uptake and willingness to pay among health app developers, it is advisable for assessments to facilitate or enable clinical guideline integration and reimbursement or procurement for high-scoring apps. Assessment scaling thus requires close collaboration with health authorities, healthcare institutions, and insurers. Further, if health system organisations are targeted as customers, it is essential to provide evidence for the assessment’s impact on patients’ and clinicians’ willingness to use health apps.