Numeracy in Patients with Hemophilia

Background: Numeracy, defined as the ability to handle basic probability and numerical concepts including computation, estimation, logic, and problem solving, is an under-recognized component of health literacy. Numeracy has been shown to influence performance of health tasks in non-hemophilia populations. Little is known about numeracy in the hemophilia population. Since hemophilia treatment requires understanding of numerical concepts to manage factor replacement, it is likely that numeracy also influences performance of health tasks by patients with hemophilia. A greater understanding of numeracy status and the characteristics influencing numeracy in the hemophilia population may allow healthcare providers to better influence health task performance. The objective of this study is to explore numeracy in the hemophilia population using two different tests of numeracy and to evaluate characteristics that are associated with low numeracy.

Methods: Using a cross-sectional design, adults with moderate or severe hemophilia A or B who spoke and read English were enrolled at their annual visit at the Emory/Children's Health Care of Atlanta Hemophilia Treatment Center (HTC). Numeracy was measured using the validated Schwartz Woloshin (SW) test requiring answers in words and the unvalidated stick figure test requiring answers using images. Subjects were considered numerate with the SW numeracy test if all three questions were answered correctly or with the stick figure numeracy test if all four questions were answered correctly. Demographic and socioeconomic characteristics collected included age, race, ethnicity, household income (more or less than $50,000), level of education completed (more or less than completion of college), and duration of time followed at this HTC. Clinical information including type and severity of hemophilia, history of viral infections, history of depression, and use of chronic medication were abstracted from the medical records.

Descriptive statistics of each variable and bivariate associations between numerate status and each dependent variable were calculated. Multivariable modeling using logistic regression was performed using the validated SW numeracy test as the dependent variable.

Results: Of 91 enrolled participants with complete data, all were men. Most had hemophilia A [n=82 (90%)] and severe disease [69 (76%)]. Median age was 34 years [interquartile range (IQR) 18]. Sixty-three (69%) were Caucasian; 5 (6%) were Hispanic; 55 (61%) reported income of <$50,000; 33 (36%) had received an undergraduate degree or higher. Median duration of time followed at the HTC was 17.0 years [IQR 18]. Twenty-four (26%) were HIV positive; 54 (59%) were HCV positive; and 19 (21%) had history of depression. Forty-one (45%) used at least one chronic medication other than factor replacement. Using the SW numeracy test, there were 22 (24%) participants who were numerate. Using the stick figure numeracy test, 60 (66%) were numerate. Only 20 (22%) of all the participants answered all seven questions correctly; two participants (2%) were numerate on the SW test but not on the stick figure test; 40 participants (44%) were numerate on the stick figure numeracy test but not on the SW test; 29 (32%) were not numerate on either test.

On bivariable analysis, SW numeracy was associated with higher education (p<0.01), higher income (p=0.035), and the use of chronic medication (p=0.048). On multivariable analysis, after adjusting for age, race, and ethnicity, SW numeracy was associated with higher education (OR 6.21, 95% CI = 1.95-19.76), use of chronic medication (OR 4.31, 95% CI = 1.29-14.34), and less time followed at the HTC (OR 0.92, 95% CI = 0.86-0.97).

Conclusion: Among patients with hemophilia, a significant proportion of patients were not numerate. Patients with less than a college education were more likely to not be numerate. Accordingly, many patients with less than a college education may struggle to understand basic numeracy concepts and this may influence their understanding of dosing, factor pharmacokinetics and probability. The impact of numeracy on health outcomes and the utility of the SW and stick figure numeracy tests to help guide patient-centered discussions that involve mathematical concepts are important areas of future research.