Taking the Guesswork out of Medical Crowdfunding

• In Event: The Future Will Be Terrible

Abstract

Online crowdfunding platforms have become an increasingly significant source of revenue for individuals seeking financial assistance for medical expenses (Renwick and Mossialos 2017). Crowdfunding website formats permit patients and their families to build web pages soliciting donations from online publics, substantiating their requests with narrative and images. These appeals circulate via social media and other online networks, sometimes generating significant return.

However, the crowdfunding mechanism may introduce unanticipated inefficiencies in health care funding (Agrawal et al. 2014). Contributors to medical crowdfunding campaigns make allocations in the absence of full information, relying on self-reported accounts that emphasize appeals to sentiment (Paulus and Roberts 2018). Crowdfunding choices thus allocate resources in the absence of information about a patient’s medical prognosis and objective financial need, and do not permit donors to assess a given campaign’s worthiness or likelihood of success relative to others. This is concerning in light of a recent finding that crowdfunding campaigns are frequently applied to unproven therapies that insurance companies will not cover (Vox et al. 2018), such that backers may unwittingly fund unscrupulous health care providers and medical long shots. Crowdfunding thus has significant potential to helping patients make up financial shortfalls and afford critical medical services, but a lack of mechanisms to establish a patient’s medical futurity makes crowdfunding a gray area for prospective donors.

I propose to reduce or even eliminate this uncertainty by introducing the power of algorithmic evaluation into crowdfunding campaigns. Algorithms – mathematical or computerized protocols for solving problems – are capable of premising predictions and decisions on the basis of big data (Cormen et al. 2009, Goffey 2008). The proposed proprietary algorithm will predict the results of medical procedures to be covered by a crowdfunding campaign and evaluate the relative utility of a donation.

While the medical “credit-worthiness” of a given individual cannot be fully predicted, this algorithm will go beyond medical prognosis, including social and behavioral factors in its assessment of a patient’s health futurity. Its evidence-based metric is based on a review and assessment of data including patient medical charts and health history, genetic profile, access to insurance, financial position, and history of education and employment. Further analysis factors in social capital, behavioral health, and lifetime exposure to sources of environmental risk. As studies in public health and medical anthropology have shown, these factors importantly predict a patient’s probability of surviving illness, recovering health, and returning to their previous capacity for social contribution.

By significantly reducing the inefficiencies of the crowdfunding model, this metric will facilitate the achievement of its full potential. Qualitative ratings like ‘Great odds’ and ‘Outside chance’ will inform donors of the likely outcome of the treatment to be funded, and suggest how far their dollar will go in influencing the patient’s health outcome. Finally, the metric will further boost giving by introducing prospective donors to campaigns similar to those they have viewed and funded in the past, “pushing” recommendations on the basis of disease to be treated, patient biography, level of need, and/or likelihood of success.

Author

• Martha Lincoln, San Francisco State University