Clinic Waiting Room-based Study of Swahili Language Artificial Intelligence-driven Symptom Assessments in Tanzanian Primary Health Care Facilities

Clinic Waiting Room-based Pilot of Swahili Language Artificial Intelligence-driven Symptom Assessments in Primary Health Care Facilities, Dar es Salaam, Tanzania

This study will evaluate the accuracy of the condition suggestions and urgency advice of the Swahili language Ada symptom assessment application (SAA), when symptoms are input by a lay-person user and a medical professional; these SAA results will then be compared to the condition suggestions and urgency advice of different tiers of doctors and a "gold standard" created by a panel.

There will be two different wings of this study: (1) where a lay-person inputs their symptoms into the Swahili language Ada symptom assessment application (SAA); (2) where the usual care doctor asks a lay-person (the patient) the same questions from the SAA and determines their own condition suggestions based off this questioning and the SAA's condition suggestions. In both wings, the patient will be seeing their usual care doctor for a diagnosis and then a study-provided physician for a diagnosis. Separately, the doctors will list a ranking of other considered condition suggestions (diagnoses) and urgency advice, which will not be shared with the patient. These condition suggestions and urgency advice will be compared to a "gold standard", which will be determined by a high level panel for each patient's case, based on the doctor notes and physical examination results of each patient's visit.

Inclusion Criteria:

• All patients who enter the clinic and that are willing/able to provide consent will be included, except those who fall under certain conditions listed below

Exclusion Criteria:

• Patients with severe injury/illness requiring immediate treatment
• Patients with traumatic injury (many of these patients require minimal anamnesis, and it is not rational to include them in a pilot study)
• Patients incapable of completing a health assessment (e.g. due to illiteracy, mental impairment or inebriation or other incapacity).