Original Article

Health-Literacy-Sensitive Diabetes
Self-Management Interventions: A
Systematic Review and Meta-Analysis
Su Hyun Kim, PhD, RN Anna Lee, RN

ABSTRACT
Keywords Background: Low health literacy is a potential barrier to self-management among patients with
diabetes, diabetes. A variety of strategies for low health literacy have been proposed for diabetes self-
self-management, management interventions, but interventions accommodating low health literacy have been het-
health-literacy erogeneous in terms of content and have produced mixed results.
interventions, Aim: To systematically review health-literacy-sensitive diabetes management interventions, with
systematic review, a focus on identifying strategies for accommodating patients with low health literacy, and to
meta-analysis examine the efficacy of these interventions to improve health outcomes.
Methods: PubMed, CINAHL, and EMBASE were searched for intervention studies published
between January 2000 and January 2015. Two authors separately identified full-texts according
to the inclusion criteria and assessed study quality using the quantitative components of the Mixed
Methods Appraisal Tool. The final list of studies to be analyzed was made through discussion.
The meta-analysis was conducted using a random effects model.
Results: Thirteen studies were selected from the 490 studies found in our initial search. We
identified a range of strategies for accommodating those with low health literacy in diabetes self-
management interventions, which encompassed four domains: written communication, spoken
communication, empowerment, and language/cultural consideration. Using at least one of the
spoken communication strategies led to positive cognitive/psychological, self-care, and health
outcomes. We found that, overall, health-literacy-sensitive diabetes management interventions
were effective in reducing HbA1C level in the meta-analysis.
Linking Evidence to Action: Healthcare providers should consider active implementation of
strategies for accommodating people with low health literacy in diabetes self-management in-
terventions. The routine use of spoken communication strategies would be necessary to achieve
the best health outcomes in diabetes self-management interventions. More research is needed
to determine the individual effects of the key strategies that improve health and reduce health
disparity.

INTRODUCTION diabetes is higher among patients with low health literacy, as

Low health literacy, defined as a lack of the capacity to ob-
tain, process, and understand basic health information and
services needed to make appropriate health decisions (Insti-
tute of Medicine, 2004), has been considered as a potential
barrier to improving health outcomes. According to a system-
atic review on health literacy, low health literacy is associated
with poor health outcomes and poor use of healthcare services
(Berkman, Sheridan, Donahue, Halpern, & Crotty, 2011). In
particular, patients with diabetes, who must engage in constant
self-management, must have adequate health literacy to apply
the requisite knowledge, decision-making skills, and problem-
solving skills for effective diabetes management (Fransen, von
Wagner, & Essink-Bot, 2012). The risk of complications from
they are less likely to have adequate knowledge of diabetes
and related self-care activities than are those with appropriate
health literacy (Al Sayah, Majumdar, Williams, Robertson, &
Johnson, 2013).
Researchers, in response, have proposed a variety of inter-
ventions to improve health outcomes and reduce the health
disparity associated with low health literacy. The strategies fre-
quently recommended to healthcare providers are to improve
the usability of health information by using plain language, fo-
cusing on actions, limiting the number of messages, acknowl-
edging cultural differences, supplementing instructions with
pictures, and checking that patients understand (The Office of
Disease Prevention and Health Promotion, 2015).

324 Worldviews on Evidence-Based Nursing, 2016; 13:4, 324333.


C 2016 Sigma Theta Tau International

However, in the literature, interventions developed for
patients with low health literacy not only have been het-
erogeneous in content and context but also have produced
mixed results (Bailey et al., 2014; Pignone, DeWalt, Sheridan,
Berkman, & Lohr, 2005). In one narrative systematic review,
the interventions for reducing literacy-related differences in
diabetes outcomes were summarized as patient education,
self-management support, disease management, and feedback
of health-literacy screening results to providers (Bailey et al.,
2014); however, this review did not provide sufficient informa-
tion on the strategies adopted for accommodating patients with
low health literacy or the effectiveness of these interventions.
Thus, efforts to improve diabetes care for people with low
health literacy would benefit from a detailed systematic review
on the components of diabetes interventions or the strategies
specifically designed for patients with low health literacy. Deter-
mining the effectiveness of the strategies for accommodating
low health literacy through a meta-analysis would also support
evidence-based practice for patients with low health literacy. In
this systematic review and meta-analysis, we examined health-
literacy-sensitive diabetes management interventions, with an
emphasis on identifying strategies for accommodating patients
with low health literacy. We also evaluated the effectiveness of
these interventions for improving health outcomes and reduc-
ing the health disparity associated with low literacy.
METHODS
Data Sources
We conducted a literature search to identify studies pub-
lished between January 2000 and January 2015 using PubMed,
CINAHL, and EMBASE. We consulted with two librarians who
have expertise in searching academic articles to identify key-
words and Medical Subject Heading (MeSH) terms. We em-
ployed several combinations of keywords and MeSH search
terms in each electronic search engine as follows: health lit-
eracy [MeSH] OR (health [All Fields] AND literacy [All
Fields]) OR health literacy [All Fields] OR numeracy [All
Fields]; diabetes mellitus, type 2 [MeSH] OR type 2 dia-
betes [tw] OR type II diabetes [tw]; self-care [MeSH] OR
self-care [tw] OR self-care [tw] OR self-management [tw] the effectiveness of the interventions on HbA1C in people with
OR self-management [tw] OR disease management [tw]
OR disease-management [tw]. We also reviewed the refer-
ence lists of included studies and systematic reviews to find
additional studies. Any disagreements among researchers dur-
ing the literature search were resolved by discussion and a
consensus was reached.
Eligibility Criteria
Inclusion criteria were studies that (a) clearly described that
the intervention was developed or adapted for patients with
low health literacy; (b) included patients with type 2 diabetes;
(c) measured health literacy levels of each participant; (d) used
an experimental design; (e) were peer-reviewed and published
in English; and (f) measured at least one of the following out-
Worldviews on Evidence-Based Nursing, 2016; 13:4, 324333.

C 2016 Sigma Theta Tau International
Original Article
comes: glycosylated hemoglobin (HbA1C), systolic blood pres-
sure (SPB), lipid profile, diabetes knowledge, self-efficacy, or
self-care behaviors. Studies solely aiming to develop or validate
instruments were excluded.
Study Quality Assessment
We used the quantitative domain of the Mixed Methods Ap-
praisal Tool (MMAT) to examine study quality (Pluye et al.,
2011). This tool consists of four criteria for each type of study
design, such as clear descriptions of the randomization and al-
location concealment, completed outcome data, and low with-
drawal rate in randomized controlled trials (RCTs). All items
were rated dichotomously (yes = 1, no = 0). The answers
were summed and converted into a score ranging from 0% to
100%. We assessed the study quality separately and resolved
disagreements through discussion.
Data Extraction, Synthesis, and Analysis
Two authors separately reviewed abstracts and candidate full-
texts from the electronic search according to the eligibility cri-
teria. Decisions to include abstracts for full-text review as well
as the final list of studies were made through discussion. After
obtaining the final list of studies, we extracted the characteris-
tics of each study, including design, intervention, participants,
and findings (Appendix S1, available with the online version of
this article). To organize the diverse and heterogeneous strate-
gies for accommodating patients with low health literacy, we
categorized the strategies according to a previous study that de-
veloped an inclusive list of health-literacy precautions (DeWalt
et al., 2011; Table 1).
We used the Comprehensive Meta-Analysis software ver-
sion 2.2 (Biostat, NJ, USA) to conduct the meta-analysis. We
used a random effects model that assumed study-level and
sampling variance to calculate effect sizes and standardized
mean differences in the change of HbA1C between groups
(Cooper, 2010). Multiple effect sizes were calculated when a
study included multiple intervention groups. In this review, we
obtained the overall effect size for the outcome of HbA1C from
eight intervention studies that measured HbA1C. To identify
differing health-literacy statuses, we calculated separate effect
sizes for people with low and high health literacy from three
studies that stratified intervention effects by health-literacy sta-
tus. Health-literacy status was determined based on health-
literacy scores in the original study. The use of meta-analytic
techniques was not possible for other outcome measures such
as knowledge, self-efficacy, or self-care, because of the high
diversity in measurement.
RESULTS
The electronic search yielded a total of 484 studies, and
6 studies were found using other resources (Figure 1). Through
a review of the abstracts, 42 studies were identified as relevant
for full-text review, and 13 studies were selected for final review
325
 326
Table 1. Strategies of Diabetes Self-Management Interventions for Patients with Low Health-Literacy

Empowerment
Spoken communication Written communication Language/cultural consideration

Clear Teach-back Follow-up with Easy-to-read Effective health Encouragement Behavioral Action Motivational
communication Method patients material education method of questions activation plans interviewing

Randomized controlled trials

Cavanaugh (2009)
Health-Literacy-Sensitive Diabetes Interventions

 
Crowley (2013)  
Gerber (2005)   
Khan (2011)   
Moussa (2013)  
Rothman, Malone et al. (2004)    
Randomized controlled trials with more than two arms
Hill-Briggs (2011)  
Negarandeh (2013)   
Schillinger (2009)   
One group pretestposttest
Kandula (2009)  
Rothman, DeWalt et al. (2004)    


Swavely (2014)    
Wallace (2009)   

C 2016 Sigma Theta Tau International

Worldviews on Evidence-Based Nursing, 2016; 13:4, 324333.
 Original Article

Identification
Articles identified through database Additional articles identified
searching through other sources
Screening n = 484 n=6

Articles after removal of duplicates

n = 414
Articles excluded for not satisfying
eligibility criteria
n = 372
Full-text articles assessed
Eligibility

for eligibility Full-text articles excluded for

n = 42 following reasons (n = 29)
Intervention did not specifically
target health literacy (n = 8)
No measure of health literacy (n = 4)
No health outcomes (n = 6)
Study protocol (n = 6)
Included

Articles included in Duplicated study (n = 3)

quantitative synthesis No control group (n = 1)
n = 13 No peer review (n = 1)

Articles excluded for not having

HbA1C as outcome
n=4

Articles included in meta-

analysis (HbA1C)
n=8

Articles excluded for not stratifying

outcomes by health literacy level
n=6

Studies included in
subgroup meta-analysis
(HbA1C)
n=3

Figure 1. PRISMA Flow Diagram.

because they met the eligibility criteria. Among the 13 studies in
the narrative analysis, eight were included in the meta-analysis
for the effectiveness of the interventions on HbA1C, and three
studies were included in the subgroup meta-analysis by health-
literacy status.
Study Characteristics
In terms of the overall quality of the studies, two studies sat-
isfied all four criteria (100%) and seven studies satisfied three
criteria (75%; Appendix S2, available with the online version
of this article). The criterion most often violated was clear
description of allocation concealment. All except one study
were conducted in the United States. Six studies were RCTs
with control groups, three studies had more than two inter-
vention groups in the RCT design, and four studies used a
one-group pretestposttest design. The majority of the inter-
ventions were educational programs on diabetes management
delivered in-person, via telephone, or via computer multime-
dia. The remaining studies included medication management
facilitation through contact with primary care providers (Crow-
ley et al., 2013), and problem-solving training (Hill-Briggs et al.,
2011).
Regarding the study population, two studies recruited only
participants with low health literacy (Moussa, Sherrod, &

Worldviews on Evidence-Based Nursing, 2016; 13:4, 324333. 327


C 2016 Sigma Theta Tau International
Health-Literacy-Sensitive Diabetes Interventions
Choi, 2013; Negarandeh, Mahmoodi, Noktehdan, Heshmat,
& Shakibazadeh, 2013). The remaining studies enrolled par-
ticipants with low and high health literacy, where 30%60%
had low health literacy. Health literacy was most commonly
measured using the Short Test of Functional Health Literacy
in Adults (STOFHLA) and the Rapid Estimate of Adult Liter-
acy in Medicine (REALM). Only six studies selected patients
according to a criterion of HbA1C level, which ranged from
6.5% to 8.0% (Cavanaugh et al., 2009; Hill-Briggs et al., 2011;
Moussa et al., 2013; Rothman, DeWalt et al., 2004; Rothman,
Malone et al., 2004; Schillinger, Handley, Wang, & Hammer,
2009).
The outcomes measured in these studies were highly di-
verse. We organized the outcomes into three domains accord-
ing to the framework of health literacy and its associations
with diabetes mechanisms and outcomes (Bailey et al., 2014;
Appendix S3, available with the online version of this article): (a)
cognitive or psychological outcomes (knowledge, self-efficacy,
activation, and perceived susceptibility); (b) self-care outcomes
(overall self-care behavior, diet, exercise, medication, problem-
solving, glucose testing, and foot care); and (c) health outcomes
(HbA1C, blood pressure, lipid profile, body mass index, body
function, and distress).
Most studies measured only one or two outcome domains.
Only four studies assessed the all three domains in terms of
how they were directly or indirectly influenced by the health-
literacy-sensitive diabetes intervention (Hill-Briggs et al., 2011;
Khan et al., 2011; Swavely, Vorderstrasse, Maldonado, Eid, &
Etchason, 2014; Wallace, Carlson, Malone, Joyner, & Dewalt,
2010). Knowledge and HbA1C were the most frequently mea-
sured outcomes in eight and nine studies, respectively. After
the interventions, six studies reported improved knowledge
(Hill-Briggs et al., 2011; Kandula et al., 2009; Moussa et al.,
2013; Negarandeh et al., 2013; Swavely et al., 2014; Wallace
et al., 2009) and five studies reported enhanced HbA1C level
(Cavanaugh et al., 2009; Hill-Briggs et al., 2011; Rothman,
DeWalt et al., 2004; Rothman, Malone et al., 2004; Swavely
et al., 2014).
Knowledge was measured at 3 months or less postinter-
vention, except in two studies, which measured knowledge at
6 months postintervention (Gerber et al., 2005) or 12 months
postintervention (Gerber et al., 2005; Hill-Briggs et al., 2011).
HbA1C was assessed at 312 months follow-up. Six studies
examined whether the interventions had differing effects be-
tween people with low and those with high literacy across the
three outcome domains (Cavanaugh et al., 2009; Crowley et al.,
2013; Gerber et al., 2005; Khan et al., 2011; Rothman, DeWalt
et al., 2004; Rothman, Malone et al., 2004; Schillinger et al.,
2009).
Strategies of Health-Literacy-Sensitive Diabetes
Management Intervention
We identified a wide range of strategies for accommodating
those with low health literacy in diabetes self-management
328
interventions (Table 1). Most interventions utilized multiple
strategies. To organize these strategies, we categorized them
into four domains, which we adopted and modified from a pre-
vious study (DeWalt et al., 2010): (a) written communication
(easy-to-read materials and effective use of health education);
(b) spoken communication (clear communication, use of the
teach-back method, and follow-up with patients); (c) empower-
ment (encouragement of questions, behavioral activation, ac-
tion plans, and motivational interviewing); and (d) language or
cultural consideration. We defined spoken communication
as the use of oral communication during the patientclinician
interaction (DeWalt et al., 2010). When a multimedia platform
with a simple and easy-to-use interface was used as an alter-
native to traditional educational methods, we categorized it as
easy-to-read material and effective use of health education
methods under written communication (DeWalt et al.,
2010). Strategies to facilitate patients to take responsibility for
their health care and to take care of themselves during diabetes
management were categorized as empowerment (DeWalt
et al., 2010). Addressing language and cultural differences in
communication with patients from ethnic minority groups by
language assistance, improvement of cross-cultural communi-
cation skills, or considerations of cultural beliefs and customs
was categorized as language or cultural consideration.
The most commonly used domain for low health literacy
was written communication. Eight of these studies developed
easy-to-read materials to improve readability and understand-
ing of written information, such as using easy language, rele-
vant illustrations, and simple format (Cavanaugh et al., 2009;
Gerber et al., 2005; Hill-Briggs et al., 2011; Kandula et al., 2009;
Khan et al., 2011; Moussa et al., 2013; Negarandeh et al., 2013;
Wallace et al., 2009). Four studies tested new educational ma-
terials using computer multimedia in conjunction with spoken
instruction and video due to the limited ability of print materi-
als to reach patients with low health literacy (Gerber et al., 2005;
Kandula et al., 2009; Khan et al., 2011; Moussa et al., 2013).
Using written communication strategies with computer mul-
timedia improved knowledge (Kandula et al., 2009; Moussa
et al., 2013) and perceived susceptibility (Gerber et al., 2005)
but did not affect self-care or health outcomes (Gerber et al.,
2005; Kandula et al., 2009; Khan et al., 2011; Moussa et al.,
2013).
In terms of spoken communication strategies, five studies
used the clear communication strategy, such as using
common words, limiting content to 35 key points, repeating
key points, and drawing pictures when speaking to patients
(Cavanaugh et al., 2009; Negarandeh et al., 2013; Rothman,
DeWalt et al., 2004; Rothman, Malone et al., 2004; Swavely
et al., 2014). Four studies utilized the teach-back method to
ensure that participants understood diabetes education (Ne-
garandeh et al., 2013; Rothman, DeWalt et al., 2004; Rothman,
Malone et al., 2004; Swavely et al., 2014). Follow-ups with
patients were conducted in four studies through phone calls
or in-person meetings to check changes or actions to ensure
diabetes self-management (Rothman, DeWalt et al., 2004;
Worldviews on Evidence-Based Nursing, 2016; 13:4, 324333.

C 2016 Sigma Theta Tau International

Rothman, Malone et al., 2004; Schillinger et al., 2009; Wallace
et al., 2009). The seven studies that developed interventions
incorporating spoken communication to accommodate those
with low health literacy, which were typically used in conjunc-
tion with other strategies, reported significant improvements
in cognitive or psychological outcomes (Cavanaugh et al.,
2009), self-care (Cavanaugh et al., 2009; Negarandeh et al.,
2013; Swavely et al., 2014; Wallace et al., 2009), and health
outcomes (Cavanaugh et al., 2009; Rothman, DeWalt et al.,
2004; Rothman, Malone et al., 2004).
Regarding empowerment, three studies incorporated the
strategy of behavioral activation that features the application
and usability of educational information to facilitate patients
engagement and action (Hill-Briggs & Smith, 2008) in dia-
betes interventions (Hill-Briggs et al., 2011; Rothman, DeWalt
et al., 2004; Rothman, Malone et al., 2004). With behavioral
activation, information delivered to patients was centered on
behavior rather than on medical content not directly related to
patient behavior. Similarly, two studies used interventions that
required patients to devise action plans to improve their self-
efficacy in maintaining health in their daily lives (Schillinger
et al., 2009; Wallace et al., 2009). The strategy of encouraging
questions was used in one study; specifically, a conversation
map was used to facilitate discussion and patients participa-
tion in self-care of diabetes (Swavely et al., 2014). The strategy
of motivational interviewing was used in one study with the
goal of assisting patients in working through their ambivalence
concerning behavior change (Crowley et al., 2013).
In addition, five studies developed educational content and
communication tailored to an ethnic minority population that
considered their health beliefs and cultural practices and de-
veloped educational materials in languages other than English
(Crowley et al., 2013; Gerber et al., 2005; Khan et al., 2011;
Schillinger et al., 2009; Swavely et al., 2014).
Effectiveness of Health-Literacy-Sensitive Diabetes
Management Intervention
The meta-analysis of nine intervention trials with 1,874 sub-
jects identified a small but significant reduction in HbA1C
level (0.18%; 95% CI: 0.36 to 0.004) for health-literacy-
sensitive interventions in comparison to usual care (provision
of routine medical services in clinics or hospitals), regardless
of patients health-literacy status (Figure 2). In the subgroup
meta-analysis including three studies (572 subjects) that strati-
fied by health-literacy status, health-literacy-sensitive interven-
tions had a moderate significant effect on HbA1C among pa-
tients with low health literacy compared to the control group
(0.42%; 95% CI: 0.81 to 0.04), but had no significant effect
on patients with high health literacy (0.13%; 95% CI: 0.80
to 0.54).
DISCUSSION
In this review, we identified a diverse range of strategies for
accommodating patients with low health literacy in diabetes
Worldviews on Evidence-Based Nursing, 2016; 13:4, 324333.

C 2016 Sigma Theta Tau International
Original Article
self-management interventions. These strategies fell into four
domains: (a) written communication, (b) spoken communica-
tions, (c) empowerment, and (d) tailoring communication to
patients language and cultural practices and beliefs.
The most commonly used strategy was developing easy-to-
read materials as a form of written communication, in isolation
or in combination with other strategies. In this review, all in-
terventions using written communication as a main strategy
were delivered through computer multimedia or web-based
programs adapted for easy access and content. Written commu-
nication strategies were effective in improving some cognitive
and psychological outcomes but not self-care or health status.
This finding indicates that diabetes education programs utiliz-
ing multimedia devices are effective in improving knowledge
in patients who have low health literacy within a short-term
time period. However, in the long term, such programs are
insufficient for producing change in health behavior or better
health outcomes. Although patients with low health literacy
showed a willingness to learn and use a computer to improve
their diabetes knowledge and manage their diabetes (Moussa
et al., 2013), without face-to-face interaction as an additional
strategy, there may be limitations in achieving positive health
outcomes on a long-term basis.
However, using at least one of the spoken communica-
tion strategies led to positive results on cognitive or psycho-
logical outcomes as well as self-care and health outcomes
(Cavanaugh et al., 2009; Rothman, DeWalt et al., 2004;
Rothman, Malone et al., 2004; Swavely et al., 2014; Wal-
lace et al., 2009). This finding underscores the critical im-
portance of oral communication strategies in interactions
with patients with low health literacy for diabetes education.
As shown in a recent qualitative study, even though dia-
betes educational materials are generally accompanied by easy-
to-follow instructions or are used with educational technol-
ogy, many challenges for improving health behavior remain,
such as language discordance and lack of consideration of
health beliefs (Mohan, Riley, Boyington, & Kripalani, 2013).
Spoken communication centered on patients accommodat-
ing low health literacy may improve patients health through
discussion about their illness experiences as well as agree-
ments on treatment options regarding diabetes management
(Stewart et al., 2000). To achieve the best health outcomes
in diabetes self-management interventions, the routine use
of spoken communication strategies, such as the teach-back
method, clear communication, and follow-up with patients, is
necessary.
Behavioral activation and action plans, both empowerment
strategies, were shown to be effective in improving self-care
behavior and glucose control in the intervention groups in
this review (Rothman, DeWalt et al., 2004; Rothman, Mal-
one et al., 2004; Swavely et al., 2014; Wallace et al., 2009).
The studies using two intervention arms showed that the ef-
fectiveness of empowerment strategies was stronger when us-
ing a more intensive format, or with frequent contact with
healthcare providers (Hill-Briggs et al., 2011; Schillinger et al.,
329
Health-Literacy-Sensitive Diabetes Interventions

Study name Statistics for each study Std diff in means and 95% CI
Std diff Standard Lower Upper
in means error Variance limit limit Z-Value p-Value
Cavanaugh (2009) -0.436 0.144 0.021 -0.718 -0.154 -3.032 0.002
Crowley (2013) -0.069 0.106 0.011 -0.276 0.138 -0.655 0.512
Gerber (2005) 0.186 0.128 0.016 -0.065 0.438 1.450 0.147
Khan (2011) -0.381 0.202 0.041 -0.777 0.015 -1.886 0.059
Rothman, DeWalt et al. (2004) -0.487 0.138 0.019 -0.757 -0.217 -3.532 0.000
Schillinger (2009)Ia -0.061 0.138 0.019 -0.332 0.211 -0.438 0.661
Schillinger (2009)Ib 0.120 0.138 0.019 -0.149 0.390 0.875 0.381
Rothman, Malone et al. (2004) -0.497 0.101 0.010 -0.694 -0.300 -4.941 0.000
Swavely (2014) -0.061 0.131 0.017 -0.319 0.197 -0.464 0.642
-0.183 0.091 0.008 -0.362 -0.004 -2.006 0.045
-1.00 -0.50 0.00 0.50 1.00
Heterogeneity Q = 34.6, df = 8 (p < .001), I2 = 76.88
Favours intervention Favours control

2.1 Intervention effects for patients overall

IaAutomated telephone self-management support with nurse follow-up
IbGroup sessions facilitated by physician and health educators

Study name Statistics for each study Std diff in means and 95% CI
Std diff Standard Lower Upper
in means error Variance limit limit Z-Value p-Value
Gerber (2005) -0.059 0.172 0.030 -0.397 0.278 -0.344 0.731
Rothman, DeWalt et al. (2004) -0.731 0.230 0.053 -1.183 -0.280 -3.176 0.001
Rothman, Malone et al. (2004) -0.749 0.145 0.021 -1.033 -0.465 -5.168 0.000
-0.508 0.237 0.056 -0.973 -0.043 -2.142 0.032

-1.00 -0.50 0.00 0.50 1.00

2
Heterogeneity Q = 10.49, df = 2 (p = .005), I = 80.93 Favours intervention Favours control

2.2 Intervention effects for patients with low health literacy

Study name Statistics for each study Std diff in means and 95% CI
Std diff Standard Lower Upper
in means error Variance limit limit Z-Value p-Value
Gerber (2005) 0.516 0.195 0.038 0.134 0.898 2.650 0.008
Rothman, DeWalt et al. (2004) -0.201 0.174 0.030 -0.541 0.139 -1.161 0.246
Rothman, Malone et al. (2004) -0.682 0.157 0.025 -0.990 -0.374 -4.344 0.000
-0.130 0.341 0.116 -0.799 0.539 -0.381 0.703

-1.00 -0.50 0.00 0.50 1.00

Heterogeneity Q = 22.94, df = 2 (p = .000), I2 = 91.28 Favours intervention Favours control

2.3 Intervention effects for patients with high health literacy

Figure 2. Interventions effects on HbA1 C level for the overall pool of patients (2.1), patients with low health
literacy (2.2), and patients with high health literacy (2.3).

2009). By focusing on patients actual behavior changes rather
than on the ways of delivering information, empowerment
strategies appeared to help patients successfully enact be-
havior change (Seligman et al., 2007). The repeated use of
empowerment strategies might lead to more successful out-
comes. More research is needed to understand the effects of
diverse empowerment strategies on health outcomes through
behavior change and identify the minimum intensity of
interventions by mapping them to a behavioral theory, such
as social cognitive theory (Seligman et al., 2007).

330 Worldviews on Evidence-Based Nursing, 2016; 13:4, 324333.


C 2016 Sigma Theta Tau International

In the meta-analysis, the small significant reduction in
HbA1C level suggests that multiple strategies for accommodat-
ing patients with low health literacy in diabetes management
interventions successfully improves overall glycemic control
in patients with diabetes. The multiple strategies used in these
interventions, however, limited our ability to draw conclusions
on which strategy is the most effective in improving health
outcomes or reducing health disparities. We identified only
one study that compared two types of educational strategy for
patients with low health literacythe pictorial image and teach-
back methodsthat showed similar effects on knowledge and
self-care behavior, as compared to usual care (Negarandeh et al.,
2013).
In the subgroup meta-analysis stratifying patients by health-
literacy level, a significant moderate effect on HbA1C level
was found only in people with low health literacy. This in-
dicates that health-literacy-sensitive diabetes management in-
terventions are more beneficial for such patients than for those
with higher health literacy. The meta-analysis findings sup-
ported the results from another narrative systematic review
indicating the effectiveness of diabetes interventions in reduc-
ing literacy-related differences in health outcomes (Bailey et al.,
2014). However, interpretations of this finding must be made
cautiously because the subgroup analysis was conducted with
only three studies measuring differing intervention effects on
HbA1C between patients with low and those with high health
literacy (Gerber et al., 2005; Rothman, DeWalt et al., 2004;
Rothman, Malone et al., 2004).
This review has several limitations. First, there are limita-
tions derived from the characteristics of the selected studies
for the current review. These limitations included nonconceal-
ment of allocation in many RCTs, poor descriptions of the
strategies accommodating low health literacy, use of multi-
ple strategies without theoretical justification, which makes it
difficult to separate their effects, and overall lack of analysis
of intervention effects stratified by health-literacy level. Several
studies did not recruit patients with diabetes based on the crite-
rion of glucose control and health-literacy status, which might
have influenced the validity of the findings. Second, our review
excluded studies that did not measure health literacy directly,
and we may have failed to identify interventions that could
be advantageous to patients with low health literacy. Third, the
meta-analysis was not possible for outcomes other than HbA1C
because of the high variability in measurements of these out-
comes.
In order to examine the effects of strategies for accommo-
dating patients with low health literacy on health disparities,
further research should stratify the effects by health-literacy
level. Strategies for accommodating those with low health lit-
eracy are applicable as a universal precaution in healthcare
practice. However, results stratified by health-literacy level
would determine whether the intervention helped to reduce
the differences in outcome associated with health-literacy sta-
tus (Pignone et al., 2005). In addition, there must be more
rigorous intervention studies that carefully specify the indi-
Worldviews on Evidence-Based Nursing, 2016; 13:4, 324333.

C 2016 Sigma Theta Tau International
Original Article
vidual strategies for accommodating patients with low health
literacy to identify which strategies are the most effective in
obtaining positive health outcomes.
This review has important clinical implications for
healthcare providers. Given the positive effect of health-
literacy-sensitive interventions on glycemic control, health-
care providers should actively incorporate strategies for
accommodating patients with low health literacy in diabetes
self-management interventions. Healthcare providers might
prioritize the implementation of spoken communication strate-
gies if they have difficulty modifying their diabetes education
with complex and lengthy strategies to accommodate patients
with low health literacy. Afterwards, they can consider adding
strategies such as written communication, empowerment, and
language or cultural consideration.
The more-positive outcomes were associated with interven-
tions that utilized strategies of spoken communication and
empowerment as compared to multimedia interventions. The
healthcare providerpatient relationship may be an important
influence on patients health outcomes and needs to be taken
into account in diabetes management interventions (Kaplan,
Greenfield, & Ware, 1989). Thus, healthcare providers should
consider utilizing interpersonal strategies for patients with low
health literacy rather than relying on multimedia devices. Addi-
tional evidence from further research could facilitate the appli-
cation of strategies for accommodating those with low health
literacy in diabetes self-management interventions.
CONCLUSIONS
In this review, we identified various strategies for accom-
modating patients with low health literacy in diabetes self-
management interventions. These strategies fell into the do-
mains of written communication, spoken communication, em-
powerment, and language or cultural consideration. We found
that, overall, health-literacy-sensitive diabetes management in-
terventions were effective for improving glycemic control.
Further research is necessary to identify key strategies for ac-
commodating patients with low health literacy, and to test the
effectiveness of these strategies stratifying patients on health-
literacy level. Healthcare providers should consider active im-
plementation of strategies for accommodating patients with
low health literacy in diabetes self-management interventions
in their clinical practice. WVN
LINKING EVIDENCE TO ACTION
r Healthcare providers should consider active im-
plementation of strategies for accommodating
patients with low health literacy in diabetes self-
management interventions in their clinical prac-
tice.
331
Health-Literacy-Sensitive Diabetes Interventions
r Diabetes education programs utilizing multime-
dia devices, without face-to-face interaction as an
additional strategy, might be insufficient to achieve
positive health outcomes for patients with low
health literacy on a long-term basis.
r To achieve the best health outcomes in diabetes
self-management interventions in patients with
low health literacy, the routine use of spoken com-
munication strategies is necessary.
r Behavioral activation and action plans, both em-
powerment strategies, are effective in improving
diabetic self-care behavior and glucose control in
patients with low health literacy.
Author information
Su Hyun Kim, Associate Professor, College of Nursing, Re-
search Institute of Nursing Science, Kyungpook National Uni-
versity, Korea; Anna Lee, Doctoral candidate, The University of
North Carolina at Chapel Hill, NC
This research was supported by Basic Science Research Pro-
gram through the National Research Foundation of Korea
(NRF) funded by the Ministry of Science, ICT, and Future
Planning (NRF-2014R1A1A3051163).
Address correspondence to Dr. Su Hyun Kim, College of
Nursing, Kyungpook National University, Daegu 700-422,
Korea; suhyun_kim@knu.ac.kr
Accepted 31 October 2015
Copyright 
C 2016, Sigma Theta Tau International
References
Al Sayah, F., Majumdar, S. R., Williams, B., Robertson, S., & John-
son, J. A. (2013). Health literacy and health outcomes in diabetes:
A systematic review. Journal of General Internal Medicine, 28(3),
444-452. doi: 10.1007/s11606-012-2241-z
Bailey, S. C., Brega, A. G., Crutchfield, T. M., Elasy, T., Herr,
H., Kaphingst, K., . . . Schillinger, D. (2014). Update on health
literacy and diabetes. The Diabetes Educator, 40(5), 581-604. doi:
10.1177/0145721714540220
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. doi: 10.7326/0003-4819-155-2-201107190-00005
Cavanaugh, K., Wallston, K. A., Gebretsadik, T., Shintani, A.,
Huizinga, M. M., Davis, D., . . . Rothman, R. L. (2009). Ad-
dressing literacy and numeracy to improve diabetes care: Two
randomized controlled trials. Diabetes Care, 32(12), 2149-2155.
doi: 10.2337/dc09-0563
Cooper, H. (2010). Research synthesis and meta-analysis. Thousand
Oaks, CA: SAGE Publications, Inc.
Crowley, M. J., Powers, B. J., Olsen, M. K., Grubber, J. M., Ko-
ropchak, C., Rose, C. M., . . . Bosworth, H. B. (2013). The Choles-
terol, Hypertension, and Glucose Education (CHANGE) study:
332
Results from a randomized controlled trial in African Ameri-
cans with diabetes. American Heart Journal, 166(1), 179-186. doi:
10.1016/j.ahj.2013.04.004
DeWalt, D. A., Broucksou, K. A., Hawk, V., Brach, C., Hink, A.,
Rudd, R., & Callahan, L. (2011). Developing and testing the health
literacy universal precautions toolkit. Nursing Outlook, 59(2), 85-
94. doi: 10.1016/j.outlook.2010.12.002
DeWalt, D. A., Clallahan, L. F., Hawk, V. H., Broucksou, K. A.,
Hink, A., Rudd, R., & Brach, C. (2010). Health literacy univer-
sal precautions toolkit. Washington, DC: Agency for Healthcare
Research and Quality.
Fransen, M. P., von Wagner, C., & Essink-Bot, M. L. (2012).
Diabetes self-management in patients with low health liter-
acy: Ordering findings from literature in a health literacy
framework. Patient Education and Counseling, 88(1), 44-53. doi:
10.1016/j.pec.2011.11.015
Gerber, B. S., Brodsky, I. G., Lawless, K. A., Smolin, L. I., Arozul-
lah, A. M., Smith, E. V., . . . Eiser, A. R. (2005). Implementa-
tion and evaluation of a low-literacy diabetes education com-
puter multimedia application. Diabetes Care, 28(7), 1574-1580.
doi: 10.2337/diacare.28.7.1574
Hill-Briggs, F., Lazo, M., Peyrot, M., Doswell, A., Chang, Y. T.,
Hill, M. N., . . . Brancati, F. L. (2011). Effect of problem-solving-
based diabetes self-management training on diabetes control in a
low income patient sample. Journal of General Internal Medicine,
26(9), 972-978. doi: 10.1007/s11606-011-1689-6
Hill-Briggs, F., & Smith, A. S. (2008). Evaluation of diabetes and
cardiovascular disease print patient education materials for use
with lowhealth literate populations. Diabetes Care, 31(4), 667-
671. doi:10.2337/dc07-1365
Institute of Medicine. (2004). Health literacy: A prescription to end
confusion. Washington DC: The National Academies.
Kaplan, S. H., Greenfield, S., & Ware, J. E. (1989). Assess-
ing the effects of physician-patient interactions on the out-
comes of chronic disease. Medical Care, 27(3), S110-S127. doi:
10.1097/00005650-19803001-00010
Kandula, N. R., Nsiah-Kumi, P. A., Makoul, G., Sager, J., Zei, C.
P., Glass, S., . . . Baker, D. W. (2009). The relationship between
health literacy and knowledge improvement after a multimedia
type 2 diabetes education program. Patient Education and Coun-
seling, 75(3), 321-327. doi: 10.1016/j.pec.2009.04.001
Khan, M. A., Shah, S., Grudzien, A., Onyejekwe, N., Banskota, P.,
Karim, S., . . . Gerber, B. S. (2011). A diabetes education multi-
media program in the waiting room setting. Diabetes Therapy,
2(3), 178-188. doi: 10.1007/s13300-011-0007-y
Mohan, A. V., Riley, M. B., Boyington, D. R., & Kripalani,
S. (2013). Illustrated medication instructions as a strategy to
improve medication management among Latinos: A qualita-
tive analysis. Journal of Health Psychology, 18(2), 187-197. doi:
10.1177/1359105312440300
Moussa, M., Sherrod, D., & Choi, J. (2013). An e-health inter-
vention for increasing diabetes knowledge in African Ameri-
cans. International Journal of Nursing Practice, 19, 36-43. doi:
10.1111/ijn.12167
Negarandeh, R., Mahmoodi, H., Noktehdan, H., Heshmat, R.,
& Shakibazadeh, E. (2013). Teach back and pictorial image
educational strategies on knowledge about diabetes and med-
ication/dietary adherence among low health literate patients
with type 2 diabetes. Primary Care Diabetes, 7(2), 111-118. doi:
10.1016/j.pcd.2012.11.001
Worldviews on Evidence-Based Nursing, 2016; 13:4, 324333.

C 2016 Sigma Theta Tau International

Pignone, M., DeWalt, D. A., Sheridan, S., Berkman, N., & Lohr,
K. N. (2005). Interventions to improve health outcomes for pa-
tients with low literacy. A systematic review. Journal of General
Internal Medicine, 20(2), 185-192. doi: 10.1111/j.1525-1497.2005.
40208.x
Pluye, P., Robert, E., Cargo, M., Bartlett, G., OCathain, A., Grif-
fiths, F., . . . Rousseau, M. C. (2011). Proposal: A mixed methods
appraisal tool for systematic mixed studies reviews. Retrieved from
http://mixedmethodsappraisaltoolpublic.pbworks.com
Rothman, R. L., DeWalt, D. A., Malone, R., Bryant, B., Shin-
tani, A., Crigler, B., . . . Pignone, M. (2004). Influence of patient
literacy on the effectiveness of a primary care-based diabetes
disease management program. JAMA, 292(14), 1711-1716. doi:
10.1001/jama.292.14.1711
Rothman, R., Malone, R., Bryant, B., Horlen, C., DeWalt,
D., & Pignone, M. (2004). The relationship between liter-
acy and glycemic control in a diabetes disease-management
program. The Diabetes Educator, 30(2), 263-273. doi: 10.1177/
014572170403000219
Schillinger, D., Handley, M., Wang, F., & Hammer, H. (2009).
Effects of self-management support on structure, process, and
outcomes among vulnerable patients with diabetes: A three-
arm practical clinical trial. Diabetes Care, 32(4), 559-566. doi:
10.2337/dc08-0787
Seligman, H. K., Wallace, A. S., DeWalt, D. A., Schillinger, D.,
Arnold, C. L., Shilliday, B. B., . . . Davis, T. C. (2007). Fa-
cilitating behavior change with low-literacy patient education
SUPPORTING INFORMATION
Additional supporting information may be found in the online version of this article at the publishers web site:
Appendix S1. Characteristics of the included studies.
Appendix S2. Results of Mixed Methods Appraisal Tool (MMAT) quality rating.
Appendix S3. Summary of outcomes of health-literacy-sensitive interventions for diabetes management.
Worldviews on Evidence-Based Nursing, 2016; 13:4, 324333.

C 2016 Sigma Theta Tau International
Original Article
materials. American Journal of Health Behavior, 31, S69-S78. doi:
10.5555/ajhb.2007.31.supp.S69
Stewart, M., Brown, J. B., Donner, A., McWhinney, I. R., Oates,
J., Weston, W. W., & Jordan, J. (2000). The impact of patient-
centered care on outcomes. The Journal of Family Practice, 49(9),
796-804. doi: 10.1097/MCC.0b013e328364d50f
Swavely, D., Vorderstrasse, A., Maldonado, E., Eid, S., & Etchason,
J. (2014). Implementation and evaluation of a low health literacy
and culturally sensitive diabetes education program. Journal for
Healthcare Quality, 36(6), 16-23. doi: 10.1111/jhq.12021
The Office of Disease Prevention and Health Promotion. (2015).
Quick guide to health literacy. Retrieved from http://www.
health.gov/communication/literacy/
Wallace, A. S., Carlson, J. R., Malone, R. M., Joyner, J., & De-
walt, D. A. (2010). The influence of literacy on patient-reported
experiences of diabetes self-management support. Nurs-
ing Research, 59(5), 356-363. doi: 10.1097/NNR.0b013e3181ef
3025
Wallace, A. S., Seligman, H. K., Davis, T. C., Schillinger, D.,
Arnold, C. L., Bryant-Shilliday, B., . . . DeWalt, D. A. (2009).
Literacy-appropriate educational materials and brief counseling
improve diabetes self-management. Patient Education and Coun-
seling, 75(3), 328-333. doi: 10.1016/j.pec.2008.12.017
doi 10.1111/wvn.12157
WVN 2016;13:324333
333
Copyright of Worldviews on Evidence-Based Nursing is the property of Wiley-Blackwell and
its content may not be copied or emailed to multiple sites or posted to a listserv without the
copyright holder's express written permission. However, users may print, download, or email
articles for individual use.