Many self-monitoring of blood glucose (SMBG)-systems demonstrate positive glucose bias at lower and negative bias at higher hematocrit (HCT) levels. Variations can thus lead to measurement errors and possibly to over- or underdosage of insulin. ISO 15197:2013 demands that manufacturers publish HCT-ranges within which BG-meters perform within certain tolerance limits. But how wide should these ranges be? In this cross-sectional study, we collect real-world data from hospital outpatients to assess the frequency of out-of-range HCT-values compared to healthy individuals.
Data from more than 1 million patients were collected from hospitals in Leeuwarden (NL) and Prague (CZ) and stratified by outpatient status, sex and age. Data were compared to common HCT labelled range limits and data from 1780 healthy Czech subjects to assess differences in distribution.
Compared to healthy subjects, outpatients from Prague were distributed with significantly higher dispersion of values. Independent of the location, low HCT-values are common in age groups known to also have high prevalence of diabetes.
Real-world data indicate that BG-meters labelled to perform only within the frequently used 30-55% range would leave thousands of patients at risk of falsely high or low results. Adequate SMBG-systems should be chosen carefully, particularly for patients with increased and decreased HCT-values.
| Age group | Location | Total number Female / Male | Number <30% HCT F / M | Percentage <30% HCT F / M | Prevalence of Diabetes (Eurostat) F / M |
| 55-64 | NL | 27329 / 33546 | 1043 / 1324 | 3.8% / 3.9% | 6.2% / 10.3% |
| 55-64 | CZ | 22212 / 19035 | 748 / 970 | 3.4% / 5.1% | 10.9% / 11.6% |
| 65-74 | NL | 32004 / 44181 | 1958 / 2849 | 6.1% / 6.4% | 12.0% / 15.0% |
| 65-74 | CZ | 33598 / 31321 | 1987 / 1924 | 5.9% / 6.1% | 18.1% / 19.5% |
| 75+ | NL | 42435 / 45080 | 3431 / 4535 | 8.1% / 10.1% | 14.4% / 19.7% |
| 75+ | CZ | 26645 / 22706 | 1923 / 2041 | 7.2% / 9.0% | 27.3% / 25.3% |
| Age group | Location | Total number Female / Male | Number <30% HCT F / M | Percentage <30% HCT F / M | Prevalence of Diabetes (Eurostat) F / M |
| 55-64 | NL | 27329 / 33546 | 1043 / 1324 | 3.8% / 3.9% | 6.2% / 10.3% |
| 55-64 | CZ | 22212 / 19035 | 748 / 970 | 3.4% / 5.1% | 10.9% / 11.6% |
| 65-74 | NL | 32004 / 44181 | 1958 / 2849 | 6.1% / 6.4% | 12.0% / 15.0% |
| 65-74 | CZ | 33598 / 31321 | 1987 / 1924 | 5.9% / 6.1% | 18.1% / 19.5% |
| 75+ | NL | 42435 / 45080 | 3431 / 4535 | 8.1% / 10.1% | 14.4% / 19.7% |
| 75+ | CZ | 26645 / 22706 | 1923 / 2041 | 7.2% / 9.0% | 27.3% / 25.3% |
T. Zima: Research Support; Self; Roche Diabetes Care Health and Digital Solutions. H. Storm: None. M. Spacek: None. S. Wardat: Employee; Self; Roche Diagnostics Corporation. O. Hauss: Consultant; Self; Roche Diabetes Care GmbH. R. Hinzmann: Employee; Self; Roche Diagnostics Corporation.
