People who used digital apps to message their primary-care doctors in Israel generally rated the experience positively, but they gave lower marks to perceived response time and the ability to receive quality care and follow-up without a clinic visit. The findings come from a cross-sectional survey by Ido Morag, published in PLOS Digital Health.

The study focused on asynchronous text messaging: patients send a message through a health maintenance organization app and receive a reply later from their regular primary-care physician. It did not examine phone or video consultations. The distinction matters because users must decide what to write, which request category to choose and whether a concern is suitable for messaging while they wait for a response.

The survey included 500 Hebrew-speaking adults in Israel who had used an HMO app to contact their regular doctor in the previous six months. Across nine ratings of features such as trust, clarity, convenience and response time, the average was 4.11 on a five-point scale. The intention to keep using the service scored highest, at 4.52. Perceived response time and perceived quality of care and follow-up without a clinic visit scored lowest, each at 3.90.

Those scores capture users’ perceptions, not measured reply times or clinical quality. The nine-item average was a secondary summary, not a validated single measure of satisfaction. The results therefore suggest that active users see value in the service while identifying aspects that could warrant closer study; they do not show that messaging produced better care or that patients experienced actual delays.

Request type was associated with differences in ratings. Administrative requests made up 64.2% of the sample and, after adjustment for measured participant and request characteristics, were associated with a higher overall average than clinical requests. Administrative requests also had higher ratings for convenience, usability and intention to continue using messaging. These are observational associations: they cannot establish that the type of request caused the difference. Other associations varied by survey item, and some were weakened in sensitivity analyses.

The findings also raise questions about how people choose a channel for urgent concerns. About one in five respondents said their typical requests were highly urgent, but the adjusted overall rating did not differ by urgency. The survey did not assess whether those messages were appropriate for an asynchronous channel or what happened after they were sent. The results cannot establish that messaging was unsafe or that users received late replies.

Morag presents clearer request categories, guidance about appropriate use and escalation, expected response-time information, and separate routing for administrative and clinical requests as ideas for future evaluation. They remain hypotheses: the survey did not test any workflow change or determine whether it would affect safety, clinical outcomes, staff workload or user experience.

The sample also limits what can be concluded about access. Participants were existing users recruited through an opt-in online panel, and the survey was in Hebrew. It cannot represent people who do not use these apps or those who face barriers involving language, digital literacy, devices, internet access, disability, privacy or confidence with technology. The study was cross-sectional and relied on self-reported experience, so its associations should not be interpreted as causal or generalized to all patients, HMOs or forms of remote care.

The study offers a closer look at messaging as a designed service, rather than treating all remote care as one experience. Whether different workflows can address users’ weaker ratings while supporting appropriate care will require prospective research that measures actual service performance and includes people who face barriers to digital access.