Functional health literacy in patients with cancer may play a crucial role in successful treatment and outcomes. Importantly, health literacy is distinct from formal education. For example, one study found that 63% of patients with 9 to 11 years of education and 34% of patients who graduated from high school had a marginal/inadequate level of functional health literacy.13 Furthermore, low rates of adequate levels of functional health literacy are common, particularly among the elderly and those with less formal education.4,10,14-17 Among elderly patients (>60 years of age) at urban public hospitals, one study found that 81% of English speakers and 83% of Spanish speakers had marginal/ inadequate levels of functional health literacy.17 Functional health literacy is lower among older age groups even after adjusting for differences in mental status, frequency of reading the news, health status, and visual acuity.18 The physical health of participants with lower reading levels has been found to be poor compared with that of participants with higher reading levels even after adjusting for confounding sociodemographic variables.2,19 Individuals with an inadequate level of health literacy are also more likely to report depressive symptoms, explained in part by their worse health status.2,20
Health care professionals must be sensitive to the level of functional health literacy of their patients when they provide information regarding treatment options and prognoses. Analyses of the readability of patient education materials, discharge instructions, and consent forms have found that these materials are typically written at too complex a level for many or most patients.21,22 Some evidence suggests that tailoring communications for adults with low literacy can be effective.23 However, patients with a variety of health literacy levels may have difficulty understanding health information; therefore, improving communication may help patients across all levels of health literacy.
To mitigate barriers to health literacy, health care professionals should take steps when meeting with patients to ensure that communication is clear and that patients understand what is being taught to them. One recommended strategy involves the health care professional asking the patient questions toward the end of a clinical encounter to assess whether the patient recalls and understands the information or instructions provided.24 For example, the health care professional might ask the patient about the name, dose, and frequency of a medication that was just prescribed. This approach, which is often called the “teach-back” method, provides health care professionals with an opportunity to confirm patient understanding and gives patients the opportunity to solidify their understanding.
Previous studies of health literacy found that lower levels of literacy were correlated with being male,16 elderly,4,15-17 and having less formal education15,16 and income.16 The demographic and socioeconomic characteristics of our sample followed these patterns (see Table 1).
In evaluating our hypotheses, we found that an adequate level of health literacy increased the likelihood of receiving chemotherapy among patients with stage 3/4 disease, a finding that suggests greater levels of health literacy might help patients receive better care. However, we detected no other clear differences by level of health literacy in patient beliefs, preferences, or decision-making about chemotherapy. We did not find an association between level of health literacy and either cancer stage at diagnosis or vital status at last observation.
Our study had several limitations. First, our small sample limited our power to detect differences by health literacy status. Second, the S-TOFHLA might not precisely capture the desired construct of health literacy. Instead, it could be better regarded as a test of reading comprehension in a health care context rather than as a test of the broader concept of health literacy.25 Specifically, the S-TOFHLA might not evaluate aspects of health literacy other than reading, such as oral health literacy, navigation, and culture. The limitations of the instrument as a measure of health literacy could attenuate its association with some health outcomes. A third potential limitation that might be more general to health literacy research is the challenge of including sufficiently large numbers of participants with marginal/inadequate levels of literacy to detect the effects of health literacy levels. As noted, our goal was a sample of approximately 50% marginal/inadequate health literacy, but our actual sample had 30% marginal/inadequate health literacy. We targeted patients for inclusion in the health literacy substudy using formal education as a proxy, and our results reinforced the conclusion of prior research that formal education and health literacy, while related, are distinct.13 Several previous studies of health literacy using the S-TOFHLA (not all on cancer) obtained samples with even lower percentages of marginal/ inadequate levels of health literacy.15,16,26
The lower-than-expected numbers of participants with marginal/inadequate levels of proficiency in studies of health literacy suggest that selection bias might influence which patients enter these studies. To be eligible for inclusion in our sample, participants had to enroll in CanCORS, complete a baseline survey, and be administered the S-TOFHLA. Patients who died before completing any of these steps could not have participated in the sample. For included patients, the mean number of days from cancer diagnosis to CanCORS enrollment and baseline survey was 150 days and from diagnosis to S-TOFHLA administration was 640 days.
It is possible that, if the level of health literacy was associated with rates of survival, then patients with CRC who died before enrolling in CanCORS, or before they could complete the baseline survey or the S-TOFHLA, might have had disproportionately low levels of health literacy. Poor health and difficulty completing surveys among those with low levels of health literacy might systematically limit these patients’ participation in studies of health literacy. Future studies of health literacy should be designed to account for this possibility.