Abstract
Recent initiatives by healthcare payers to mandate the collection of Social Determinants of Health (SDoH) data underscore the importance of efficient and accurate data capture methods in improving patient care. Traditional paper-based questionnaires, while widely used, present challenges in terms of cost, accuracy, and completeness of data entry into Electronic Health Records (EHRs). This study investigates the application of artificial intelligence (AI), specifically document understanding models and large language models (LLMs), to automate the transformation of paper-based SDoH questionnaires into structured, machine-readable formats suitable for EHR integration. Utilizing a test dataset derived from the Cambridge Health Alliance SDoH questionnaire, available in eight languages, this study explored the feasibility of using Microsoft Azure Document Intelligence and OpenAI’s GPT-4 and GPT-3.5 Turbo, which were selected for their advanced capabilities in document understanding and language processing, respectively, for this purpose. The findings show that GPT-4 outperforms GPT-3.5 Turbo across various metrics, including accuracy and consistency, albeit at a higher cost. This feasibility study highlights the potential of using AI as a relatively accurate and potentially cost-effective alternative to manual data entry of SDoH data collected using paper-based questionnaires. It also suggests there will be challenges such as data privacy and security considerations as well as the integration of AI-generated data into EHR systems that merit further research.
Competing Interest Statement
The authors have declared no competing interest.
Funding Statement
This study did not receive any funding
Author Declarations
I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.
Yes
I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.
Yes
I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).
Yes
I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.
Yes
Footnotes
Description This study explores the feasibility of using artificial intelligence (AI) to transform a response to a paper-based Social Determinants of Health (SDoH) questionnaire into a structured representation that could theoretically be persisted to an electronic health record (EHR). It suggests there is potential for AI, particularly document understanding and large language models, to offer a cost-effective, accurate alternative to manual data entry of responses to paper-based questionnaires.
Data Availability
All data produced are available at https://github.com/oekekezie/poc-ai-sdoh-questionnaire