Community-Based Hemoglobin A_1C Testing in Barbershops to Identify Black Men With Undiagnosed Diabetes

Abstract

This cross-sectional study evaluates the use of hemoglobin A_1c testing at barbershops owned by black individuals for timely diagnosis of diabetes among black men and suggests appropriate methods for care.

In the United States, black men with diabetes have disproportionately high rates of diabetic complications and are less likely to survive into their 70s compared with men in other racial and ethnic groups.^1,2 The diagnosis of diabetes is often delayed, especially among black men without a regular source of primary care. In barbershops, which are places of trust among black men, community-based interventions have been successful in identifying and treating men with hypertension.³ Using point-of-care hemoglobin A_1c (HbA_1c) testing, we evaluated a community-based approach for diabetes screening in barbershops owned by black individuals.

Methods

From September 19, 2017, to January 23, 2019, customers were approached and tested at 8 barbershops owned by black individuals in Brooklyn, New York, in neighborhoods previously identified as having a high prevalence of poor glycemic control.⁴ English-speaking black men without a history of diabetes and aged 18 years or older were included in the study. Individuals with blood disorders such as sickle cell disease or those who had recently experienced blood loss were excluded from the study to avoid obtaining spurious HbA_1c results. Participants were tested with the A1CNow⁺ test (PTS Diagnostics), which provides results within 5 minutes and has a reported accuracy of 93% when correlated with HbA_1c testing of venous blood.⁵ Participants with an HbA_1c level of 6.5% or higher on a single test result were considered to have diabetes; however, a confirmatory test was not performed. Participants with an abnormal HbA_1c result (≥5.7%) were counseled about the importance of modifying their diet and physical activity, the need for medical management, and were provided contact information for local primary care clinics. The institutional review board at the NYU Grossman School of Medicine approved the study. Written informed consent was obtained from the participants.

Results

Of the 895 black men who were asked to participate in the study, 312 (34.9%) agreed to be screened and 290 (32.4%) were successfully tested. Eight men were excluded because they had a blood disorder, and 14 men had an error code during their testing (such as for insufficient or too much blood) and declined a second test to resolve the code. Of the 583 men who refused to participate in the study, 331 (56.8%) provided a reason for refusal. Of these 331 men, 187 (56.5%) reported already knowing their health status or having been checked by their doctor, and 117 (35.3%) reported either being healthy, not having the time or interest, or not wanting to know their result. In addition, 26 men (7.9%) reported being scared of needles and only 1 specifically reported not wanting to be tested in a barbershop.

The Table shows the demographic characteristics of the 290 participants who were successfully tested. Of 290 participants, 26 (9.0%) had an HbA_1c level of 6.5% or higher and 3 (1.0%) had an HbA_1c level of 7.5% or higher. The highest HbA_1c level was 7.8%. In addition, 82 participants (28.3%) had an HbA_1c level between 5.7% and 6.4%, which is the criterion for diagnosing prediabetes. Of the 26 participants with undiagnosed diabetes, 16 (61.5%) were obese. The median age of these men diagnosed with diabetes was 41 (range, 22-65) years and 11 (42.3%) had an education of high school or less.

Table. Characteristics of 290 Study Participants.

Characteristic	All Ages, No. (%)	Age Group, No. (%)
		18-29 y	30-39 y	40-49 y	50-59 y	60-79 y
Participants	290 (100)	71 (24.5)	87 (30.0)	51 (17.6)	51 (17.6)	20 (6.9)
Caribbean or West Indies origin	75 (25.9)	20 (28.2)	26 (29.9)	11 (21.6)	12 (23.5)	3 (15.0)
Foreign born	78 (26.9)	10 (14.1)	31 (35.6)	17 (33.3)	13 (25.5)	4 (20.0)
High school education or less	128 (44.3)	25 (35.2)	44 (50.6)	17 (34.0)	25 (50.0)	11 (55.0)
BMI, Mean (SD)	29.3 (6.0)	28.5 (6.0)	29.4 (6.0)	29.9 (5.8)	30.3 (6.1)	27.0 (5.7)
Obese (BMI >30)	102 (42.0)	21 (37.5)	33 (43.4)	17 (39.5)	22 (52.4)	5 (27.8)
Undiagnosed diabetes	26 (9.0)	3 (4.2)	10 (11.5)	6 (11.8)	6 (11.8)	1 (5.0)

Abbreviation: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared).

Discussion

We found that approximately one-third of men approached in barbershops owned by black individuals in Brooklyn were willing to be screened for diabetes. We also found that barbers were important health advocates; although we do not have exact numbers, some customers (who initially declined testing) agreed after encouragement from their barber. Our study sample may not be representative of other barbershops; however, the prevalence of undiagnosed diabetes (9.0%) that we found was much higher than the estimated prevalence of undiagnosed diabetes at 3.6% among New York City residents.⁶

The participation rates may not be generalizable to other community-based settings or other areas of New York City or the United States. Although point-of-care HbA_1c testing is relatively accurate, confirmatory testing is also important.

Black men who live in urban areas of the United States may face socioeconomic barriers to good health, including poor food environments and difficulty in obtaining primary care. Our findings suggest that community-based diabetes screening in barbershops owned by black individuals may play a role in the timely diagnosis of diabetes and may help to identify black men who need appropriate care for their newly diagnosed diabetes.