Validation of a Point-of-care Device Measuring Ferritin With Capillary Blood (FerPOC)

Measurement of iron is important for identifying both low and high iron levels in the body. Measuring a protein in the blood called ferritin is currently the easiest way to determine a patient's iron status. However, the test requires a blood sample that is be sent to the laboratory for measurement, which can cause delays in obtaining the result. The investigators are testing whether a point of care device that measures ferritin levels using a capillary blood sample is accurate compared to current methods of testing. This could allow measurement of ferritin immediately in the clinic with only a drop of blood, and enable treatment decisions based on these results.

All patients will undergo testing with the study intervention (point of care) and routine testing with serum ferritin (venous blood sample).

The proportion of subjects where the difference between the two different methods is less than 15; this will be measured by calculating the difference between the two results. The proportion for the absolute difference between the two measurement methods on the same subject that is less than 15 ng/mL will be calculated. Non-inferiority test for one proportion will be performed to examine, if, the proportion is greater than 95%, where the test p value and the one-sided 97.5% confidence interval will be reported. An interim analysis will be performed after the first 50 subjects are enrolled to test the assumptions made.

Distribution of the differences between the two measurement methods using the normality test, mean difference and its 95% CI, and outlier detection; other methods may be used as appropriate

Comparison of the differences between ferritin levels as measured by the standard test and POC test in a Repeated Measures Design

Applies to patients who have multiple measurement points during study follow-up. The differences in ferritin levels between the two methods will be studied using a mixed model to account for the between-subject variation and the within-subject variation. The mean difference and its 95% CI, the Bonferroni-adjusted p-values and the plot of mean vs time will be reported.

Applies only to patients with at least two repeated measures. Agreement between the two measurement methods will be quantified and summarized using probability of agreement.

The Receiver Operating Characteristic curves (ROC) analysis will be performed to report the diagnostic accuracy for the new measurement method. The optimal cut-off value and the associated accuracy measures (sensitivity, specificity, predictive values and etc.) will be reported.

Inclusion Criteria: able to read and understand English (the consent form is in English) attending ambulatory hematology clinic and/or outpatient medical care Exclusion Criteria: -