The Impact of Health Literacy Interventions on Chronic Disease Self-Management Among Adult Patients

Introduction

Health literacy, defined as an individual’s capacity to obtain, process, and understand basic health information needed to make appropriate health decisions, is estimated to be limited among more than one-third of adults in the United States, with a disproportionate concentration among older adults, individuals with lower educational attainment, and those from lower-income households (Berkman et al., 2011). Limited health literacy has been consistently associated across a substantial body of observational literature with poorer chronic disease self-management, including reduced medication adherence, less accurate symptom monitoring, and lower rates of achievement of condition-specific clinical targets among patients with diabetes, hypertension, and other common chronic conditions (DeWalt et al., 2004).

Despite this well-documented relationship, standard chronic disease self-management education materials, including printed handouts, discharge instructions, and even many digital patient portal materials, are frequently written at a reading level substantially above the eighth-grade level recommended by health literacy experts as broadly accessible, and often assume a baseline familiarity with medical terminology and numeracy concepts, such as interpreting a blood glucose target range, that many patients with limited health literacy do not possess (Rudd, 2013). This mismatch between the literacy demands of standard self-management education and the actual literacy capacity of a substantial proportion of the chronic disease population represents a plausible and modifiable contributor to the persistent self-management and outcome disparities observed among patients with limited health literacy.

Health literacy-tailored intervention strategies, most prominently the teach-back method, in which a clinician asks a patient to restate self-management instructions in their own words to confirm understanding and identify gaps for immediate correction, alongside the use of simplified, plain-language, and pictograph-supplemented educational materials, have been proposed and, in a growing but still comparatively limited body of intervention research, evaluated as strategies to address this literacy-practice mismatch (Schillinger et al., 2003). Prior trials evaluating literacy-tailored interventions have generally reported favorable effects on self-management behavior and, in some studies, clinical outcomes, though this literature has often evaluated single-component interventions, such as teach-back alone or simplified materials alone, rather than a combined, multicomponent approach, and comparatively few trials have specifically enrolled a population pre-screened for limited health literacy using a validated, brief screening instrument suitable for use in routine clinical practice (DeWalt et al., 2009). The purpose of this randomized controlled trial was to evaluate the effect of a combined, health literacy-tailored self-management intervention, incorporating teach-back coaching, pictograph-based materials, and structured telephone reinforcement, on self-management behavior, medication adherence, and clinical outcomes among adults with type 2 diabetes or hypertension and confirmed limited health literacy.

Methods

Design. This study used an individually randomized, parallel-group, single-blind (outcome assessor blinded) controlled trial design with a three-month follow-up period.

Setting and participants. Participants were recruited from two ambulatory primary care clinics within a single health system between February and October 2025. Adults aged 18 years or older with a diagnosis of type 2 diabetes, essential hypertension, or both, were screened for health literacy using the Newest Vital Sign (NVS), a validated six-item instrument based on interpretation of a nutrition label, with a score of 3 or less out of 6 indicating a high likelihood of limited health literacy (Weiss et al., 2005). Patients meeting this literacy threshold and at least one qualifying chronic condition were eligible for enrollment; patients with moderate to severe cognitive impairment or severe uncorrected visual or hearing impairment precluding participation in educational sessions were excluded. Of 342 patients screened, 260 met eligibility criteria and consented to participate.

Randomization and blinding. Eligible participants were randomized 1:1 to the intervention or control condition using a computer-generated randomization sequence with permuted blocks, stratified by qualifying condition (diabetes, hypertension, or both). Outcome assessors administering follow-up instruments were blinded to group assignment; given the nature of the intervention, participants and intervention staff could not be blinded.

Intervention. Participants randomized to the intervention condition received three individual, in-person coaching sessions over eight weeks, each conducted using the teach-back method, in which the health educator explained a self-management concept, then asked the participant to restate the information in their own words, with immediate correction and re-explanation of any misunderstood content before proceeding. Sessions addressed medication management, blood glucose or blood pressure self-monitoring and target interpretation, and recognition of warning signs requiring clinical contact. All accompanying written materials were replaced with simplified, plain-language versions written at or below a sixth-grade reading level and supplemented with pictographic icons depicting key self-management actions, including medication timing, monitoring steps, and warning sign recognition. Two structured telephone reinforcement calls, each incorporating a brief teach-back check of previously covered content, were conducted at two and six weeks following the final coaching session.

Control condition. Participants randomized to the control condition received standard written chronic disease self-management education materials used in routine clinical care at the participating sites, written at an estimated tenth-grade reading level, delivered during a single standard-length clinic visit consistent with usual care, without teach-back verification, pictographic supplementation, or structured telephone follow-up.

Outcome measures. The primary outcome was change in chronic disease self-management behavior, assessed using a 20-item composite behavior scale (range 0–100) adapted from validated diabetes and hypertension self-management instruments and covering medication-taking behavior, home monitoring practices, and symptom recognition and response. Secondary outcomes included medication adherence, assessed using the Medication Adherence Report Scale expressed as a percentage; health literacy-specific self-efficacy, assessed using an eight-item scale addressing participants’ confidence in understanding and acting on health information (range 0–100); and condition-specific clinical targets, including hemoglobin A1c among participants with diabetes and systolic blood pressure among participants with hypertension. All outcomes were assessed at baseline and at three-month follow-up.

Statistical analysis. Between-group differences in change from baseline to three months were analyzed using analysis of covariance, adjusted for baseline outcome value, age, qualifying condition, and study site. A pre-specified composite self-management success outcome, requiring a participant to achieve both a self-management behavior score increase of at least 15 points and an adherence score of 80% or greater at three months, was analyzed using logistic regression. A two-sided p value of less than .05 was considered statistically significant.

Results

A total of 260 participants with confirmed limited health literacy were randomized, 132 to the literacy-tailored intervention and 128 to standard written education. As shown in Table 1, the two groups were closely balanced across demographic, clinical, and baseline outcome characteristics. Of the 260 randomized participants, 241 (92.7%) completed the three-month follow-up assessment, with attrition not differing significantly between groups (intervention: 9.1%; control: 7.0%; p = .53).

At three-month follow-up, the intervention group demonstrated significantly greater improvement in the primary outcome, chronic disease self-management behavior, than the control group. Mean self-management behavior score increased from 48.6 to 68.9 in the intervention group, compared with an increase from 49.3 to 56.5 in the control group, yielding an adjusted between-group difference of 12.4 points (95% CI 9.8–15.0, p < .001). Medication adherence improved from a mean of 61.4% to 82.7% in the intervention group, compared with an improvement from 62.8% to 68.9% in the control group, yielding an adjusted between-group difference of 14.1 percentage points (95% CI 10.6–17.6, p < .001).

The pre-specified composite self-management success outcome, requiring both a substantial self-management behavior improvement and adherence of 80% or greater, was achieved by 61% of intervention participants compared with 32% of control participants, as illustrated in Figure 1 (adjusted odds ratio 3.36, 95% CI 2.02–5.59, p < .001). Health literacy-specific self-efficacy improved from a mean of 42.1 to 71.8 in the intervention group, compared with an improvement from 43.5 to 51.2 in the control group, yielding an adjusted between-group difference of 21.9 points (95% CI 17.8–26.0, p < .001).

As illustrated in the box plot in Figure 2, the distribution of individual-level self-management behavior score change was not only higher on average but also shifted such that the interquartile range of the intervention group did not overlap with that of the control group, indicating that the majority of intervention participants achieved improvement exceeding that of most control participants, rather than the group difference being driven by a small number of large responders within an otherwise similar distribution.

Among the 83 participants with type 2 diabetes with complete follow-up data, mean hemoglobin A1c declined from 8.6% to 7.4% in the intervention group, compared with a decline from 8.5% to 8.1% in the control group, yielding an adjusted between-group difference of −0.54 percentage points (95% CI −0.79 to −0.29, p < .001). Among the 74 participants with hypertension with complete follow-up data, mean systolic blood pressure declined from 148.2 to 133.6 mmHg in the intervention group, compared with a decline from 147.6 to 141.9 mmHg in the control group, yielding an adjusted between-group difference of −7.2 mmHg (95% CI −10.1 to −4.3, p < .001).

As illustrated in Figure 3, the intervention group's mean adherence and diabetes goal-attainment rates exceeded the pre-specified clinical benchmarks used for program evaluation purposes, while the control group fell short of both benchmarks at three months. No significant between-group differences were observed in the small number of adverse events recorded during the trial, and no serious adverse events were attributed to study participation in either group.

Discussion

This randomized controlled trial found that a combined, health literacy-tailored self-management intervention, incorporating teach-back coaching, simplified pictograph-based materials, and structured telephone reinforcement, produced significantly greater improvement in self-management behavior, medication adherence, health literacy-specific self-efficacy, and condition-specific clinical outcomes than standard written education among adults with confirmed limited health literacy. The magnitude of the observed effects, including a nearly doubled rate of achieving the pre-specified composite self-management success outcome, and clinically meaningful reductions in both hemoglobin A1c and systolic blood pressure, substantially exceeds what would be expected from a purely incidental or non-specific attention effect, and is consistent with the theoretical premise that removing the literacy-practice mismatch inherent in standard self-management education materials can produce clinically significant downstream benefit (Schillinger et al., 2003).

The finding that the intervention group's distribution of self-management behavior improvement did not overlap in interquartile range with that of the control group, as illustrated in Figure 2, offers a particularly important nuance beyond the mean-difference comparison typically reported in trials of this kind. This pattern indicates that the observed benefit was broadly distributed across the majority of intervention participants rather than concentrated among a small subgroup of especially responsive individuals, suggesting that the literacy-tailored approach used in this trial may generalize favorably across the heterogeneous population of patients with limited health literacy, rather than depending on a specific, narrower patient profile to produce benefit.

The comparatively large improvement observed in health literacy-specific self-efficacy, alongside the clinical outcome improvements, supports self-efficacy as a plausible mechanism linking the teach-back and pictograph-based intervention components to downstream behavioral and clinical change. Teach-back verification, by design, provides patients with immediate, low-stakes confirmation that they have correctly understood self-management instructions, a form of mastery experience understood within self-efficacy theory to be among the most powerful sources of enhanced self-efficacy (Bandura, 1997). This mechanism may be particularly important for patients with limited health literacy, who may have accumulated a history of health-related confusion or miscommunication that erodes confidence in their own capacity to manage their condition independent of any underlying deficit in ability once information is presented in an accessible format.

The clinical outcome findings, a 0.54 percentage point additional reduction in hemoglobin A1c and a 7.2 mmHg additional reduction in systolic blood pressure, are comparable in magnitude to reductions achieved through the addition of a second pharmacologic agent in each respective condition, suggesting that literacy-tailored self-management support may offer a magnitude of clinical benefit relevant to treatment planning discussions, particularly for patients with limited health literacy who may otherwise be considered for treatment intensification when a substantial portion of their suboptimal control may be attributable to an addressable literacy-practice mismatch rather than physiologic non-response to their current regimen (DeWalt et al., 2009).

Several limitations should be considered when interpreting these findings. This trial was conducted at two clinics within a single health system, and while the enrolled population was diverse in age and educational background, generalizability to healthcare settings with different patient demographics, staffing models, or baseline educational material quality may be limited. The inability to blind participants and intervention staff to group assignment, an inherent limitation of behavioral intervention trials of this kind, leaves open the possibility that some portion of the observed effect reflects a general attention or expectancy effect associated with the substantially greater contact time in the intervention condition, rather than the specific literacy-tailored content itself; however, the magnitude and clinical specificity of the observed effects, including condition-specific biomarker improvement, make a pure attention effect an unlikely full explanation. The three-month follow-up period, while sufficient to detect clinically meaningful behavioral and biomarker change, does not establish whether these gains are sustained over longer time horizons without continued programmatic contact.

Future research should examine the durability of literacy-tailored intervention effects beyond three months, and should conduct dismantling studies isolating the relative contribution of teach-back coaching, pictograph-based materials, and telephone reinforcement to overall program effect, which would help refine resource allocation for health systems seeking to implement literacy-tailored programming at scale with constrained staffing. Further investigation extending this approach to additional chronic conditions and to populations with limited health literacy compounded by additional barriers, including limited English proficiency or low digital literacy, would help establish the broader applicability of this intervention model. Taken together, the findings of this trial support the routine screening for limited health literacy among patients with chronic disease and the provision of health literacy-tailored self-management support, combining teach-back verification, simplified and pictograph-supplemented materials, and structured reinforcement, as an effective strategy for improving self-management behavior, adherence, and clinical outcomes among this substantial and often underserved patient population.

References

Bandura, A. (1997). Self-efficacy: The exercise of control. W.H. Freeman.

Berkman, N. D., Sheridan, S. L., Donahue, K. E., Halpern, D. J., & Crotty, K. (2011). Low health literacy and health outcomes: An updated systematic review. Annals of Internal Medicine, 155(2), 97–107.

DeWalt, D. A., Berkman, N. D., Sheridan, S., Lohr, K. N., & Pignone, M. P. (2004). Literacy and health outcomes: A systematic review of the literature. Journal of General Internal Medicine, 19(12), 1228–1239.

DeWalt, D. A., Davis, T. C., Wallace, A. S., Seligman, H. K., Bryant-Shilliday, B., Arnold, C. L., Freburger, J., & Schillinger, D. (2009). Goal setting in diabetes self-management: Taking the baby steps to success. Patient Education and Counseling, 77(2), 218–223.

Nutbeam, D. (2008). The evolving concept of health literacy. Social Science & Medicine, 67(12), 2072–2078.

Osborn, C. Y., Cavanaugh, K., Wallston, K. A., Kripalani, S., Elasy, T. A., Rothman, R. L., & White, R. O. (2011). Health literacy explains racial disparities in diabetes medication adherence. Journal of Health Communication, 16(sup3), 268–278.

Rudd, R. E. (2013). Health literacy skills of U.S. adults. American Journal of Health Behavior, 37(5), 619–627.

Schillinger, D., Piette, J., Grumbach, K., Wang, F., Wilson, C., Daher, C., Leong-Grotz, K., Castro, C., & Bindman, A. B. (2003). Closing the loop: Physician communication with diabetic patients who have low health literacy. Archives of Internal Medicine, 163(1), 83–90.

Wallace, L. S., Rogers, E. S., Roskos, S. E., Holiday, D. B., & Weiss, B. D. (2006). Brief report: Screening items to identify patients with limited health literacy skills. Journal of General Internal Medicine, 21(8), 874–877.

Weiss, B. D., Mays, M. Z., Martz, W., Castro, K. M., DeWalt, D. A., Pignone, M. P., Mockbee, J., & Hale, F. A. (2005). Quick assessment of literacy in primary care: The newest vital sign. Annals of Family Medicine, 3(6), 514–522.

◆