Implementation of Fingerprint Technology for Unique Patient Matching and Identification in Kenya: This cross-sectional study was conducted at a HIV care and treatment facility in Western Kenya. An open-source fingerprint application was integrated within an implementation of the Open Medical Records System (OpenMRS) which is currently in use at the study setting (Preprint)
BACKGROUND Unique patient identification remains a challenge in many healthcare settings within low- and middle-income countries (LMICs). Without national-level unique identifiers for whole populations, countries rely on deterministic and probabilistic patient matching approaches that have proven suboptimal in LMICs. Affordable bio-metric-based approaches, implemented with consideration of contextual ethical, legal and social implications (ELSI), have a potential to address patient identification challenges and to improve care quality, patient safety and reporting accuracy. However, limited studies exist to evaluate actual performance of biometric approaches and perceptions towards these systems within LMIC contexts. OBJECTIVE To evaluate performance and acceptability of fingerprint technology (FPT) for unique patient matching and identification in the LMIC setting of Kenya METHODS This cross-sectional study was conducted at a HIV care and treatment facility in Western Kenya. An open-source fingerprint application was integrated within an implementation of the Open Medical Records System (OpenMRS) which is an open source electronic medical records system (EMR) and currently in use at the study setting. OpenMRS is nationally-endorsed and deployed for HIV care in Kenya and in over 40 countries, hence potential for ease of translating findings across multiple countries. Adult participants over 18 years of age were conveniently sampled and enrolled into the study. Participants’ left thumbprints were captured, stored and used to retrieve and match patient records. FPT performance was evaluated using standard measures namely: Sensitivity, False Acceptance Rate (FAR), False Rejection Rate (FRR), and Failure to Enroll Rate (FER). Wald test was used to compare the accuracy of the FPT to the EMRs’ probabilistic matching technique. Time to retrieval and matching of records was compared using the independent samples t-test. A survey was administered to evaluate patient acceptance and satisfaction with use of the FPT. RESULTS 300 participants were enrolled, mean age was 36.3 years (SD 12.2) and 174/300 (58%) were female. FPT per-formed as follows: sensitivity 89.3%, FAR 0%, FRR 11%, and FER 2.3%. FPT mean record retrieval speed was 3.2s (SD 1.1) vs. 9.5s (SD 1.9) with demographic-based record retrieval in the EMR (p<.001). Survey results revealed participants’ comfort (96.3%) and willingness (90.3%) to use the FPT. CONCLUSIONS Fingerprint Technology (FPT) performed very well in identifying adult patients within a LMIC setting. Patients reported a high level of satisfaction and acceptance of the technology. Serious considerations need to be given to use of FPT for patient identification in LMICs, but this has to be done with strong consideration on ELSI and security issues.