Thanks for your response Kate. You've definitely brought up some important considerations. Apologies for typos, as I am on mobile :)
Measuring impact on caregivers - first and foremost, we can measure retention and engagement within the app - if people are consistently talking in groups and coming back, I think that's a sign that we are making progress in reducing the social isolation of caregivers. Another thing we can do is integrate polls and self-Check quizzes into the app. We initially built this feature to integrate diagnostic forms in a substance use context, but it works for any questionare. We can juxtapose an individual answers with their engagement and sentiment scores from our app. Which brings us to the final, and most innovative measure - the sentiment scores our algorithms give messages themselves, which allow us to get an idea of how a caregiver might be feeling based on the emotional sentiment (e.g. more positive or negative). While this is very early stage, I think it is one of the most promising parts of our technology, that we can automatically extract critical content from messages themselves. WIth access to more data, we will get a understanding of how caregivers as a whole respond to different topics - for example, our existing work has drawn interesting links between things like sleep or profanity and how a persons emotional sentiment related to their discussion of such topics.
On the financial aide we currently use a B2B SaaS approach where we license the platform to a central organization, which provides the group moderators. We customize and brand a version of the app for this central organization to better reach their population of interest. In this case there is no charge to end users. It's worth nothing that groups don't have to be moderated by clinicians - it depends entirely on the preference of the organization we work with. We have had groups moderated by peer supporters, counselors, social workers, and psychiatrists - since we provide a technology platform, we can customize to fit best with what's already on the ground. With regards to caregivers for dementia patients, I think healthcare systems and perhaps even payors would see value in this - healthy caregivers lead to better outcomes for both the caregiver and the cared for. Ancedotally, when I left John's Hopkins to work on this technology, the psychiatristry department was undergoing a big push specifically for caring for the caregiver type solutions, so there is clearly some level of interest. Of course just because we used a certain model before, does not mean we have to use it now. I'd welcome feedback both on the technology and the sustainablity strategy. Just writing this I though that a model that charges caregivers a small amount could perhaps work in training and compensating a peer support to moderate groups. I do however, feel that our technology provides enough value to be sustainable without continuous grants or donations. I hope this answered your questions! Let me know if I can provide any more context.