Accuracy of Three Commercial Home-Use Hemoglobin A1c Tests

Abstract

Background:

The COVID-19 pandemic and the rapid expansion of telemedicine have increased the need for accurate and reliable capillary hemoglobin A1c (HbA1c) testing. Nevertheless, validation studies of commercially available products suitable for home use have been in short supply.

Methods:

Three commercial home-use capillary blood sample HbA1c tests (Home Access, CoreMedica, and A1cNow+) were evaluated in 219 participants with type 1 or type 2 diabetes (4–80 years years of age, HbA1c 5.1%–13.4% [32–123 mmol/mol]) at four clinical sites. Comparisons were made between HbA1c measurements from the commercial tests and paired venous samples for which HbA1c was measured at two central reference laboratories. The primary outcome was percentage of commercial HbA1c values within 5% of the corresponding reference values.

Results:

HbA1c values were within 5% (relative difference) of paired reference values for 82% of Home Access samples, 29% of CoreMedica samples, and 46% of A1cNow+ samples. Absolute differences were within 0.3% of the reference value for 75% of Home Access samples, 28% of CoreMedica samples, and 44% of A1cNow+ samples and exceeded 0.5% for 8%, 55%, and 37%, respectively.

Conclusions:

None of the commercial home-use HbA1c tests produced the National Glycohemoglobin Standardization Program goal of ≥90% measurements within 5% of a DCCT venous reference. However, the Home Access product performed substantially better than the CoreMedica or A1cNow+ products. Telemedicine is likely to persist as a mainstay of diabetes care well after the COVID-19 era. As such, accurate home-based HbA1c assessment represents an urgent need for the diabetes community.

Introduction

Hemoglobin A1c (HbA1c) measurement is the current mainstay of diabetes treatment monitoring. Specifically, HbA1c assays certified by the National Glycohemoglobin Standardization Program (NGSP) have been used for glycemic management, and HbA1c levels have been shown to have a strong association with microvascular and macrovascular diabetes complications.¹ Routine monitoring of HbA1c, at least two to four times per year, is recommended by the American Diabetes Association (ADA). While other metrics of glycemia, such as data obtained from continuous glucose monitoring (CGM), are highly valuable, access to these technologies is not yet universal for all patients with type 1 or type 2 diabetes. Due to the COVID-19 pandemic, in-person visits for individuals with diabetes have decreased and telemedicine visits have rapidly expanded.² Consequently, health care teams have not been able to achieve the monitoring of glycemic status recommended by the ADA.³

We previously reported the results of a study to evaluate two academic laboratories' fingerstick capillary blood collection kits suitable for home use, for which the blood specimens were sent to the academic laboratories for measurement of HbA1c. The study demonstrated that the kits were easy to use and the accuracy of HbA1c measurements from these specimens equaled that of specimens obtained from venipuncture, with more than 95% of the capillary HbA1c values being within 5% of the venous HbA1c values.⁴ Although neither of these sample collection kits is available for home use yet, several home-use commercial HbA1c tests are currently available in the United States.

However, rigorous studies of their performance are limited. Therefore, as part of the study, we evaluated the accuracy of three commercially available home HbA1c tests and compared their accuracy to that from venous collection and laboratory measurement. This information could not only aid medical decision making by health care teams and individuals with diabetes but also spur the development of improved home HbA1c tests, which would be of great value during the COVID-19 pandemic and beyond.

Methods

The study was conducted at four diabetes centers located in Florida, Georgia, Missouri, and Idaho. The protocol was approved by a central institutional review board. Participants ≥18 years old and parents/guardians provided informed consent; assent was obtained from participants 7 to 17 years old. Participants were ≥2 years old with a diagnosis of type 1 or type 2 diabetes with no other known medical condition, such as those that alter the red blood cell lifespan, or medications that could affect HbA1c.

As noted earlier, the evaluation of commercially available home HbA1c tests was part of a larger study that aimed to evaluate the accuracy of test kits from the two research laboratories and included 240 participants.⁴ These individuals were offered the option of also participating in the evaluation of three commercially available home-use HbA1c products: Home Access^® A1c Test (Home Access Health Corporation), CoreMedica Blood-DS™ (CoreMedica Laboratories), and A1c Now^®+ (PTS Diagnostics).

The Home Access product requires the user to fill the well of a collection cassette with “several” large capillary blood drops (at least 20 mcL). For context, blood glucose meters generally require 0.3–0.6 mcL. The blood is air-dried for 15 min before the cassette is sealed into a desiccant-containing pouch and sent for analysis through a United States Postal Service (USPS) prepaid return mailer. Analysis is performed using a Beckman Coulter AU640e Chemistry Analyzer, which is an immunoassay, in a clinical laboratory located at Home Access Health Corporation. The Home Access product is an FDA-cleared HbA1c test method.

The CoreMedica product requires the user to fill the well of a collection cassette with four large capillary blood drops. After 3 min, the cassette is sealed inside a USPS prepaid return mailer and sent for analysis. Analysis is performed using a “latex enhanced immunoturbidimetric assay” at a CoreMedica Laboratories facility. The CoreMedica product is not FDA cleared for HbA1c measurement.

The A1cNow+ product requires the user to fill the blood collector using one drop of capillary blood (∼5 mcL). The blood collector is inserted into a shaker device and shaken six to eight times to mix the blood with the solution inside the shaker body. A separate cartridge is removed from a foil pouch and inserted into an analyzer unit, and then the solution inside the shaker is injected into the cartridge. Analysis is completed and an HbA1c measurement value displayed on the analyzer within 5 min. The analysis mechanism involves a dry reagent strip inside the cartridge and a 4-channel reflectance photometer inside the analyzer.

The A1cNow+ product is the “professional” version (i.e., intended for clinical use) of a system that is also offered as an over-the-counter product (A1cNow) for home use. The A1cNow+ product is an FDA-cleared and CE-marked HbA1c test method with NGSP certification (per www.ngsp.org, accessed May 23, 2022).

Upon consent, blood collection for the commercial home HbA1c tests was overseen by clinic staff according to provided instructions, with participant assistance provided as needed. Sample collection for all tests was completed on the same day during a single clinic visit for each participant. Multiple fingersticks were used to obtain sufficient blood for the different tests, and in some cases, multiple fingersticks were required for a single test.

A survey consisting of three questions on a five-or six-point Likert scale inquiring about ease of use, overall satisfaction, and level of pain associated with collection of the sample was completed by each participant (or parent) for each product used (which must be interpreted in the context that study staff and not the participant or parent obtained the blood sample and did the processing).

Venous samples were obtained by clinic staff, placed in a −80°C freezer within 4 h of collection, and shipped on dry ice weekly to the two academic laboratories, the University of Minnesota Advanced Research and Diagnostic Laboratory (ARDL) in Minneapolis, MN, USA, and the Children's Mercy Hospital Laboratory (CMH) in Kansas City, MO, USA, through FedEx Priority Overnight^® service. Upon receipt of venous blood sample, each of the two laboratories determined if the specimen was analyzable. The Tosoh G8 HPLC system was used in variant mode (Tosoh Bioscience, Inc., South San Francisco, CA, USA) to make HbA1c measurements for the venous reference samples in both academic laboratories. Variant mode allows the analyzer to detect hemoglobin variants HbC, HbD, HbE, and HbS present in the sample.

Statistical methods

Based on the NGSP standard for laboratory HbA1c testing certification, accuracy criteria were predefined as ≥90% of the analyzable commercial home HbA1c results within ±5% of the reference HbA1c values. The reference HbA1c value was taken as the average of venous samples from the two academic laboratories (ARDL and CMH) when both were available.

Separate analyses were conducted for each of the three commercial HbA1c products comparing the HbA1c value from the sample obtained to the reference value. For each comparison, summary statistics were computed for the proportion of home test values within 5% of the reference value and for the distributions of relative differences and absolute differences. The 95% limit of agreement was calculated, winsorized at the 10th and 90th percentiles to account for a small number of outliers.

Separate exploratory analyses were performed for adult and pediatric participants and high and low HbA1c values for each of the commercial HbA1c products. The effect of age was assessed using logistic regression for the proportion of samples within ±5% of the reference (binary outcome), and linear regression for the relative differences and absolute differences (continuous outcomes). Both models controlled for the reference HbA1c level. A rank normal score transformation was used for highly skewed outcomes. The false discovery rate was controlled by using the two-stage adaptive Benjamini-Hochberg method.⁵ Analyses were conducted using SAS version 9.4 (SAS Institute, Cary, NC, USA).

Results

The evaluation of the commercially available home HbA1c tests included 219 participants (Table 1) ranging in age from 4 to 80 years (111 ≥ 18 years old and 108 < 18 years old); 185 had type 1 diabetes and 34 had type 2 diabetes. Mean HbA1c (from venipuncture measured at central laboratory) was 7.9% ± 1.6% (63 ± 17.5 mmol/mol) overall (range 5.1%–13.4% [32–123 mmol/mol]), 8.2% ± 1.6% (66 ± 17.5 mmol/mol) for pediatric participants, and 7.6% ± 1.6% (60 ± 17.5 mmol/mol) for adult participants. Reference values were missing for four participants who were excluded from the accuracy analyses but included in the survey results. Therefore, a total of N = 215 participants were included in the accuracy analysis, and N = 219 participants were included in the survey analysis.

Table 1.

Participant Characteristics

	Overall (N = 219)
Age (years)
Mean (SD)	27.3 (19.9)
Range	4–80
4 to <8	13 (6%)
8 to <18	95 (43%)
18 to <25	33 (15%)
25 to <50	37 (17%)
50 to 80	41 (19%)
Female^a	106 (48%)
Race/ethnicity
Non-Hispanic White	180 (82%)
African American	9 (4%)
Hispanic or Latino	15 (7%)
Asian	2 (<1%)
Other or mixed race	13 (6%)
Type of diabetes
T1D	185 (84%)
T2D	34 (16%)
Diabetes duration (years)^b
Median (IQR)	9 (4–16)
Range	<1 to 63
Highest level of education^c
Less than high school	2 (<1%)
High school	87 (41%)
Associate's degree	21 (10%)
Bachelor's degree	66 (31%)
More than bachelor's degree	35 (17%)
Reference HbA1c^d mean (SD) [range]	7.9% (1.6) [5.1%–13.4%]

One participant reported nonbinary gender.

Diabetes duration missing for three participants.

Education information missing for eight participants; for pediatric participants, education of parent was reported.

Reference HbA1c missing for four people from both laboratories; reference not analyzable for nine additional samples at CMH, so ARDL value was used as reference.

ARDL, Advanced Research and Diagnostic Laboratory; CMH, Children's Mercy Hospital Laboratory; HbA1c, hemoglobin A1c; IQR, interquartile range; SD, standard deviation; T1D, type 1 diabetes; T2D, type 2 diabetes.

Home Access

There were 214 Home Access samples with a paired venous reference HbA1c available. Of these, two samples were determined to be nonanalyzable due to improper collection (N = 1) or too old upon receipt (>21 days) as per laboratory policy (N = 1). Among the 212 analyzable Home Access samples, 82.5% (95% confidence interval [CI] 76.8–87.4) of A1c measurements were within 5% (relative difference) of the reference.

The median difference was −0.2% (interquartile range [IQR]: −0.3% to 0.0%, mean [standard deviation; SD] −0.19% [0.25%], Fig. 1a) and the median absolute difference was 0.2% (IQR: 0.1%–0.4%), with 34% within 0.1% and 75% within 0.3% (Table 2). R ² was 0.97. Across the range of HbA1c levels (Table 3 and Fig. 1b) and for pediatric and adult participants (Table 3), point estimates ranged from 72.9% to 91.2% for the percentage of commercial Home Access values being within 5% of the reference values. Accuracy tended to be better for adult than pediatric participants (P = 0.04, Table 3).

FIG. 1.

Comparison of Home Access, CoreMedica, and A1c Now+ measurements with reference venous HbA1c measurements. Bar graphs show the distribution of differences between the commercial test HbA1c values and venous reference values for Home Access (A, N = 212), CoreMedica (C, N = 205), and A1c Now+ (E, N = 194). Plots (with reference HbA1c on x axis) show the differences between the commercial test HbA1c values and the venous reference values for Home Access (B, N = 212), CoreMedica (D, N = 205), and A1c Now+ (F, N = 194). To represent clinical significance, B, D, and F dotted lines represent −0.3 and 0.3 limits; solid line represents no difference between commercial test and venous reference. HbA1c, hemoglobin A1c.

Table 2.

Commercial Home-Use Hemoglobin A1c Test Accuracy Analysis

Comparison measure	Home Access versus Reference (N = 212)	CoreMedica versus Reference (N = 205)	A1cNow+ versus Reference (N = 194)
Difference^a within ±5.0% (95% CI)	82.5% (76.8%, 87.4%)	28.8% (22.7%, 35.5%)	46.4% (39.2%, 53.7%)
95% Limit of agreement	(−0.57, 0.19)	(−0.13, 1.37)	(−1.16, 0.34)
Difference^a
Mean (SD)	−0.19 (0.25)	0.64 (0.48)	−0.39 (0.66)
Median	−0.2	0.6	−0.4
Range	−1.0 to 0.6	−0.6 to 2.7	−3.8 to 3.2
Distribution
<−0.5%	16	1	65
−0.5% to <−0.4%	34	0	36
−0.4% to <−0.3%	73	2	41
−0.1% to <0.1%	72	28	33
0.2% to <0.3%	14	27	12
0.4% to <0.5%	2	35	0
>0.5%	1	112	7
Cumulative distribution of absolute value of difference (%)
Identical values	10	5	6
≤0.1%	34	14	17
≤0.3%	75	28	44
≤0.5%	92	45	63
≤1.0%	100	84	87
R ²	0.97	0.90	0.81

Difference = Commercial HbA1c test minus reference HbA1c. A positive difference denotes that the commercial home test had a higher HbA1c value.

Table 3.

Accuracy Analysis in Subgroups Based on Reference Hemoglobin A1c and Age Category

Home Access versus Reference
	Reference HbA1c			Age
	<7.0% (N = 68)	7.0 to <9.0% (N = 96)	≥9.0% (N = 48)	Pediatric (<18 years) (N = 102)	Adult (≥18 years) (N = 110)	P-value for age^a
Difference^b within ±5.0% (95% CI)	91.2% (81.8%, 96.7%)	81.3% (72.0%, 88.5%)	72.9% (58.2%, 84.7%)	79.4% (70.3%, 86.8%)	85.5% (77.5%, 91.5%)	0.11
95% Limit of agreement	(−0.38, 0.20)	(−0.56, 0.15)	(−0.72, 0.14)	(−0.61, 0.13)	(−0.51, 0.23)
Difference^b						0.008
Mean (SD)	−0.06 (0.21)	−0.21 (0.21)	−0.31 (0.29)	−0.25 (0.24)	−0.13 (0.24)
Median (IQR)	−0.1 (−0.2, 0.0)	−0.2 (−0.4, −0.1)	−0.4 (−0.5, −0.2)	−0.2 (−0.4, −0.1)	−0.1 (−0.3, 0.0)
Range	−0.4 to 0.6	−0.7 to 0.2	−1.0 to 0.3	−1.0 to 0.6	−0.7 to 0.5
Absolute difference^c						0.04
Mean (SD)	0.17 (0.12)	0.24 (0.17)	0.36 (0.22)	0.28 (0.20)	0.22 (0.16)
Median (IQR)	0.2 (0.1, 0.2)	0.2 (0.1, 0.4)	0.4 (0.2, 0.5)	0.2 (0.1, 0.4)	0.2 (0.1, 0.3)
Range	0.0 to 0.6	0.0 to 0.7	0.0 to 1.0	0.0 to 1.0	0.0 to 0.7

CoreMedica versus Reference
	Reference HbA1c			Age
	<7.0% (N = 64)	7.0 to <9.0% (N = 95)	≥9.0% (N = 46)	Pediatric (<18 years) (N = 98)	Adult (≥18 years) (N = 107)	P-value for age^a
Difference^b within ±5.0% (95% CI)	28.1% (17.6%, 40.8%)	28.4% (19.6%, 38.6%)	30.4% (17.7%, 45.8%)	26.5% (18.1%, 36.4%)	30.8% (22.3%, 40.5%)	0.46
95% Limit of agreement	(−0.19, 1.34)	(−0.15, 1.36)	(−0.01, 1.40)	(−0.10, 1.43)	(−0.16, 1.30)
Difference^b						0.11
Mean (SD)	0.56 (0.42)	0.65 (0.52)	0.75 (0.48)	0.70 (0.50)	0.60 (0.47)
Median (IQR)	0.7 (0.3, 0.9)	0.6 (0.3, 0.9)	0.8 (0.4, 1.0)	0.7 (0.4, 0.9)	0.6 (0.3, 0.9)
Range	−0.3 to 1.3	−0.6 to 2.7	0.0 to 2.6	−0.6 to 2.7	−0.3 to 2.6
Absolute difference^c						0.14
Mean (SD)	0.59 (0.38)	0.67 (0.49)	0.75 (0.48)	0.72 (0.47)	0.61 (0.45)
Median (IQR)	0.7 (0.3, 0.9)	0.6 (0.3, 0.9)	0.8 (0.4, 1.0)	0.7 (0.4, 0.9)	0.6 (0.3, 0.9)
Range	0.0 to 1.3	0.0 to 2.7	0.0 to 2.6	0.0 to 2.7	0.0 to 2.6

A1c Now+ versus Reference
	Reference HbA1c			Age
	<7.0% (N = 59)	7.0% to 9.0% (N = 92)	≥9.0% (N = 43)	Pediatric (<18 years) (N = 91)	Adult (≥18 years) (N = 103)	P-value for age^a
Difference^b within ±5.0% (95% CI)	54.2% (40.8%, 67.3%)	45.7% (35.2%, 56.4%)	37.2% (23.0%, 53.3%)	58.2% (47.4%, 68.5%)	35.9% (26.7%, 46.0%)	0.008
95% Limit of agreement	(−0.76, 0.31)	(−1.17, 0.26)	(−1.45, 0.28)	(−1.09, 0.32)	(−1.22, 0.35)
Difference^b						0.13
Mean (SD)	−0.09 (0.67)	−0.49 (0.57)	−0.60 (0.69)	−0.43 (0.56)	−0.37 (0.74)
Median (IQR)	−0.2 (−0.4, 0.1)	−0.4 (−0.7, −0.2)	−0.7 (−1.1, −0.2)	−0.3 (−0.6, −0.1)	−0.4 (−0.7, −0.1)
Range	−0.9 to 3.2	−3.8 to 0.7	−2.0 to 1.6	−3.8 to 0.3	−2.3 to 3.2
Absolute difference^c						0.008
Mean (SD)	0.42 (0.52)	0.54 (0.53)	0.76 (0.51)	0.47 (0.52)	0.62 (0.54)
Median (IQR)	0.3 (0.1, 0.5)	0.4 (0.2, 0.7)	0.7 (0.3, 1.1)	0.3 (0.2, 0.6)	0.5 (0.2, 0.8)
Range	0.0 − 3.2	0.0–3.8	0.0–2.0	0.0–3.8	0.0–3.2

P-value for comparisons in age based on a logistic or linear model adjusting for reference HbA1c. Nominal (uncorrected) P values were adjusted for multiple comparisons using the Benjamini-Hochberg adaptive false discovery rate procedure. Continuous covariate age included in model in continuous form.

Difference = Commercial HbA1c test minus reference HbA1c.

Absolute value of the difference.

The survey completed by the 160 participants and 32 parents (with 26 completers unspecified) indicated that 48% thought the product was somewhat or very easy to use, while 19% thought it was very difficult (Supplementary Table S1).

CoreMedica

There were 211 CoreMedica samples with a paired venous reference HbA1c available. Of these, six samples were determined to be nonanalyzable due to noninterpretable result (N = 3), being lost/compromised in transit (N = 2), or insufficient amount of blood (N = 1). Among the 205 analyzable CoreMedica samples, 28.8% (95% CI 22.7–35.5) of HbA1c measurements were within 5% (relative difference) of the reference. The median difference was 0.6% (IQR: 0.3%–0.9%, mean [SD] 0.64% [0.48%], Fig. 1c) and the median absolute difference was 0.6% (IQR: 0.3%–0.9%), with 14% within 0.1% and 28% within 0.3% (Table 2). R ² was 0.90.

Across the range of HbA1c levels (Table 3 and Fig. 1d) and for pediatric and adult participants (Table 3), 26.5%–30.8% of CoreMedica values were within 5% of the reference. The HbA1c values appeared to be similarly inaccurate for both pediatric and adult participants (P = 0.14, Table 3).

The survey completed by the 159 participants and 30 parents (with 26 completers unspecified) indicated that 71% thought the product was somewhat or very easy to use and 3% thought it was very difficult (Supplementary Table S1).

A1cNow+

There were 202 A1cNow+ samples with a paired reference available. Of these, six samples were not analyzed due to device error codes or value not being displayed, and for two others, the site failed to record the value. Among the 194 analyzable A1cNow+ samples, 46.4% (95% CI 39.2–53.7) of HbA1c measurements were within 5% (relative difference) of the reference. The median difference was −0.4% (IQR: −0.7% to −0.1%, mean [SD] −0.39% [0.66%], Fig. 1e) and the median absolute difference was 0.4% (IQR: 0.2%–0.7%), with 17% within 0.1% and 44% within 0.3% (Table 2). R ² was 0.81. Across the range of HbA1c levels (Table 3 and Fig. 1f) and for pediatric and adult participants (Table 3), 35.9%–58.2% of A1cNow+ values were within 5% of the reference. Accuracy tended to be better for pediatric than adult participants (P = 0.008, Table 3).

The survey completed by the 154 participants and 27 parents (with 25 completers unspecified) indicated that 96% thought the product was somewhat or very easy to use and none thought it was very difficult (Supplementary Table S1).

Discussion

HbA1c, a reflection of average glycemia over the lifespan of a red blood cell (∼3 months), is used to assess glycemic status for determining appropriate pharmacologic treatment and to predict the risk of complications in individuals with diabetes. Thus, commercial HbA1c tests for home use are a valuable tool for routine monitoring. However, if they are to be used for altering medical management, then their accuracy is essential. Large studies with rigorous primary endpoints are lacking. We have demonstrated that none of the three commercial products tested meets the NGSP standard, which is to have at least 90% of values within 5% of the reference sample.

However, we observed marked differences among commercial products with Home Access performing well above CoreMedica and A1cNow+. The NGSP standard was chosen as the primary endpoint in this study because (1) results are standardized in primary and secondary laboratories against the DCCT results and (2) accurate results are important for diabetes management for patients and their medical team. Meeting the NGSP standards with home kits for collection of a fingerstick blood sample is achievable as demonstrated by the performance of the two academic laboratory kits also tested in the study. Their accuracy not only exceeded the NGSP standard (>95% of capillary samples were within 5% of the reference value) but also was comparable to the accuracy of measurements from venous blood draws.⁴

Of the three commercial kits tested, Home Access test performed the best with 82% of results within 0.5 percentage points of the reference, with only a slightly negatively skewed result. That said, none of the commercial kits provided the same degree of accuracy as similarly tested academic kits.⁴ Accuracy appeared slightly higher in adults than youth. In contrast, A1cNow+ and CoreMedica results were substantially worse with only 46% and 29%, respectively, meeting the NGSP standard. Clinically meaningful differences between the capillary test and the venous reference were common, with 37% of A1cNow+ samples and 55% of CoreMedica samples discordant from the reference values by more than 0.5 percentage points and over 10% of samples being more than 1.0 percentage point different.

There also was clear skewing of the results, with the majority of CoreMedica tests reporting falsely high HbA1c values, and the majority of A1cNow+ tests reporting falsely low HbA1c values. Accuracy for both tests was low across the age range and HbA1c range of participants. While an absolute difference of 0.5 percentage points may seem a small difference, it could change management, especially for those with values closer to the ADA target. HbA1c targets are personalized, but, in general, a goal of <7% (<53 mmol/mol) balances the objectives of optimally reducing risk for long-term complications and avoiding severe hypoglycemia, and HbA1c testing should be performed every 3–6 months.

Both negative and positive bias in home HbA1c values could influence patient and provider decisions. A falsely elevated HbA1c value could result in insulin dose increases, which may increase risks of hypoglycemia. Falsely low HbA1cs could prevent patients from recognizing factors that are negatively impacting glycemic control and implementation of necessary changes. There also are negative emotional connotations associated with HbA1c and some patients approach diabetes visits with trepidation; thus, an inaccurate result may only increase anxiety or diabetes distress. However, point-of-care testing can reduce anxiety associated with venous blood draws, which is common among children and adolescents.

A benefit of home-use HbA1c tests is empowering patients to improve their diabetes care by using their own data. Home-use HbA1c tests are also of contemporary relevance during the COVID-19 pandemic as they could be used for outcome measures in clinical trials assessing diabetes-specific therapies. This must be weighed against the suboptimal accuracy of the test. The A1cNow+ has more immediate feedback, while the other two require shipping the sample to a laboratory.

However, the ease of home collection remains for all commercial products tested. Home Access and CoreMedica samples were shipped using USPS. There were delays in sample arrival, likely due to travel delays caused by the pandemic, while this study was taking place, although this did not appear to affect the results. Satisfaction and ease-of-use questionnaire results were overall positive, but highest for the A1cNow+ product, which we speculate may be related to its lower blood volume requirements and immediate feedback.

Limitations of this study include the research staff involvement in the sample collection, which especially limits the applicability of survey data. In addition, there is increased likelihood of preanalytical/sample collection error with real “home” use. As such, real-world accuracy might be even lower than what was found in this study. The study cohort was predominately non-Hispanic White and there were too few African American participants for a separate subgroup analysis.

Point-of-care HbA1c testing by capillary blood sampling is used in many clinicians' offices to make real-time diabetes management changes during an office visit. Devices such as the DCA Vantage^® Analyzer and the Afinion™ HbA1c Dx have been demonstrated to have a high degree of accuracy, sufficient for making treatment decisions for people with diabetes.⁶ However, in-person office visits have been greatly reduced because of the COVID-19 pandemic.

As such, fewer patients are receiving point-of-care HbA1c testing to guide their diabetes management. Unrelated to the pandemic, the use of CGM-based metrics of glycemic control has been increasing. Specifically, percent time-in-range and the glucose management indicator have become increasingly meaningful measures of glycemic control. Nevertheless, CGM access issues, challenges with remote uploading of CGM and glucose data, and discontinuation of technology by patients in the setting of inadequate diabetes education demonstrate the ongoing need for multiple modalities to assess control. The development of accurate home HbA1c testing remains an important area of interest for the diabetes community.

Footnotes

Authors' Contributions

L.M.J. researched data and wrote/edited the article. L.E.B. performed statistical analyses and wrote/edited the article. J.W.L., C.K., V.B.-L., M.S., M.J.H., B.B., J.T.C., A.A.K., U.G., D.L., J.G.A., M.C., D.G., T.J., M.A.C., and R.W.B. researched data, contributed to discussion, and reviewed/edited the articles. R.W.B. is the guarantor of this work and, as such, had full access to all data in the study, and takes responsibility for the integrity of data and the accuracy of data analysis.

Author Disclosure Statement

M.A.C. reports receiving research support from Dexcom and Abbott Diabetes Care, as well as being the Chief Medical Officer for Glooko. L.M.J., L.E.B., J.W.L., C.K., V.B.-L., M.S., M.J.H., B.B., J.T.C., A.A.K., U.G., D.L., J.G.A., M.C., D.G., T.J., and R.W.B. have no disclosures relevant to this study to report.

Funding Information

Funding for the study was provided by the Leona M. and Harry B. Helmsley Charitable Trust.

Supplementary Material

Supplementary Table S1

References

The Diabetes Control and Complications Trial Research Group; Nathan

, Genuth

, Lachin

, Cleary

, Crofford

, Davis

, Rand

, Seibert

: The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med, 1993; 329:977–986.

Caballero

, Ceriello

, Misra

, et al.: COVID-19 in people living with diabetes: an international consensus. J Diabetes Complications, 2020; 34:107671.

Fragala

, Kaufman

, Meigs

, et al.: Consequences of the COVID-19 pandemic: reduced hemoglobin A1c diabetes monitoring. Popul Health Manag, 2021; 24:8–9.

Beck

, Bocchino

, Lum

, et al.: An evaluation of two capillary sample collection kits for Laboratory Measurement of HbA1c. Diabetes Technol Ther, 2021; 23:537–545.

Benjamini

, Krieger

, Yekutieli

: Adaptive linear step-up procedures that control the false discovery rate. Biometrika, 2006; 93:491–507.

Torregrosa

, Molina

, Argente

, Ena

: Accuracy of three hemoglobin A1c point-of-care systems for glucose monitoring in patients with diabetes mellitus. Endocrinol Nutr, 2015; 62:478–484.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB