Comparing point-of-care A1C and random plasma glucose for screening diabetes in migrant farm workers

Abstract

ObjectiveTo compare point-of-care (POC) glycosylated hemoglobin (A1C) and random plasma glucose (RPG) as a POC screening tool for prediabetes and diabetes in migrant farm workers of eastern North Carolina. DesignProspective, observational, single-center study. SettingFederally qualified community health center in eastern North Carolina, from August to October 2011. ParticipantsMigrant farm workers 18 years or older who resided in a migrant camp in eastern North Carolina. InterventionDiabetes screening using POC A1C and RPG via fingerstick followed by venipuncture A1C and basic metabolic panel in individuals with a positive screening. Main outcome measuresPositive predictive value (PPV) of POC A1C and RPG, incidence of positive screening, incidence of confirmed diagnosis, concordance rate of the screening tools, and correlation between POC A1C and laboratory A1C. Results206 workers participated in the screenings; screening identified 39 individuals with a POC A1C greater than 5.7% and 1 individual with both an RPG of 200 mg/dL or more and a POC A1C greater than 5.7%. Of the 39 individuals found to have a positive screening, 24 presented to Carolina Family Health Centers, Inc., for follow-up venipuncture; however, 1 participant did not have a venipuncture A1C, leaving 23 individuals with complete data. Two participants were diagnosed with diabetes and 17 with prediabetes. POC A1C had a PPV of 82.6%; however, the PPV of RPG could not be calculated due to the number of participants lost to follow-up. POC A1C correlated well with laboratory A1C regardless of time to follow-up. ConclusionPOC A1C should be considered for diabetes screening in high-risk populations. If the screening had been performed with RPG alone, 38 individuals would have gone undetected. Early identification of individuals with elevated blood glucose will likely decrease the risk of long-term complications.