Skip to main content
NCBI home page
The Gerontologist logoLink to The Gerontologist
. 2017 Mar 10;58(4):686–695. doi: 10.1093/geront/gnw254

A Multi-faceted Approach to Promote Comprehension of Online Health Information Among Older Adults

Jessie Chin 1,✉, Darcie D Moeller 1, Jessica Johnson 1, Elise A G Duwe 1, James F Graumlich 2, Michael D Murray 3, Daniel G Morrow 1
PMCID: PMC6044368  PMID: 28329840

Abstract

Purpose of the Study

Older adults’ self-care often depends on understanding and utilizing health information. Inadequate health literacy among older adults poses a barrier to self-care because it hampers comprehension of this information, particularly when the information is not well-designed. Our goal was to improve comprehension of online health information among older adults with hypertension who varied in health literacy abilities.

Design and Methods

We identified passages about hypertension self-care from credible websites (typical passages). We used a multi-faceted approach to redesign these passages, revising their content, language, organization and format (revised passages). Older participants read both versions of the passages at their own pace. After each passage, they summarized the passage and then answered questions about the passage.

Results

Participants better remembered the revised than the typical passages, summarizing the passages more accurately and uptaking information more efficiently (less reading time needed per unit of information remembered). The benefits for reading efficiency were greater for older adults with more health knowledge, suggesting knowledge facilitated comprehension of information in the revised passages.

Implications

A systematic, multi-faceted approach to designing health documents can promote online learning among older adults with diverse health literacy abilities.

Keywords: Health literacy, Intervention, Domain knowledge, Self-care, Hypertension

Older adults are more likely than younger adults to have chronic illness. At the same time, while older adults are highly variable on many dimensions, they also tend to have fewer cognitive resources needed for accomplishing complex self-care activities related to chronic illness, such as taking medications, monitoring symptoms, and modifying lifestyle behaviors. They tend to have lower levels of health literacy, defined as the ability to find and understand information needed to make health decisions (Baker, Gazmararian, Sudano, & Patterson, 2000; Nielsen-Bohlman, Panzer, & Kindig, 2004). Older adults with less health literacy tend to be less successful in carrying out health behaviors such as taking medication or utilizing health services, and in turn have more adverse health outcomes (Berkman, Sheridan, Donahue, Halpern, & Crotty, 2011; Wolf, Gazmararian, & Baker, 2005). The links among health literacy, health behaviors, and outcomes are due in part to the fact that older adults with lower health literacy have more difficulty understanding health information (Chin et al., 2011; Chin, Madison, et al., 2015).

Self-care challenges related to older adults’ understanding of self-care information may be exacerbated by evolving ecologies of literacy. The information needed for self-care is increasingly (only) available on the web, requiring older adults to access the web to find, understand, and act on the information they need to manage their health (Sharit, Hernández, Czaja, & Pirolli, 2008). Research in this area has focused on age-related differences in access to the web and other digital resources (Zickuhr & Madden, 2012) as well as on age differences in the ability to search the web for health information (Morrow & Chin, 2015). There is also some evidence that older adults are less able than younger adults to understand and evaluate web-based information (Chin & Fu, 2010; Sharit et al., 2008; Taha, Czaja, Sharit, & Morrow, 2013).

The goal of our study was to develop a systematic approach to improve web-based health information so that it is better understood and used by older adults with varying health literacy abilities. This goal requires a theoretical framework that identifies how health literacy related abilities influence comprehension. Although conceptualized from a variety of theoretical frameworks (Berkman, Davis, & McCormack, 2010), health literacy is often viewed functionally as patients’ resources (sensory and cognitive abilities) needed to manage their health demands (Nielsen-Bohlman et al., 2004). Process-based models of health literacy focus on how these patient resources influence the ability to understand and make decisions about the information needed to perform self-care and other health care tasks (Baker, 2006). The Process–Knowledge model identifies broad cognitive resources related to health literacy, including processing capacity (PC; i.e., fluid mental ability such as psychomotor speed, processing speed and working memory capacity), general knowledge (GK; i.e., verbal ability), and health knowledge (HK; Chin et al., 2011). These components have different trajectories across the lifespan: PC tends to decline whereas general and HK often sustains or increases with age (Baltes, 1997; Beier & Ackerman, 2005).

These age-related changes in abilities influence comprehension of self-care information. According to theories of comprehension (Kintsch, 1998), there are three levels of text representation: surface (recognizing words), textbase (semantic integration, binding concepts into propositions) and situation model (a mental model of the situation described by the text). Declines in PC may impair surface- and textbase-level processing by reducing the ability to integrate concepts (Stine-Morrow, Miller, Gagne, & Hertzog, 2008). Knowledge, on the other hand, may make some processes more efficient and promote conceptual integration and use of the situation model (Chin, Payne, et al., 2015; Stine-Morrow & Miller, 2009). Therefore, knowledge may compensate for PC limits on comprehension of health information (Chin, Madison, et al., 2015; Chin, Payne, et al., 2015). According to this view of health literacy, comprehension of health information may be improved by designing texts to reduce the demands of comprehension on limited PC and to build on patients’ general and HK relevant to comprehension.

Health literacy interventions related to redesigning health documents often focus on improving the readability of text, operationalized as lower grade-level scores that reflect use of simpler language (shorter words and sentences, more common words; Nielsen-Bohlman et al., 2004). These studies have produced mixed results, with some studies showing no benefits for lower literacy adults (Berkman et al., 2011). This may reflect a focus on shallow linguistic features of the text, even though comprehension is influenced by a range of text/document, as well as reader characteristics (Kintsch, 1998; Meyer, 2003; Wright, 1999).

We used a multi-faceted approach to analyze and revise passages about hypertension self-care in order to improve older adults’ comprehension of information commonly found on the web. In general, discourse can be analyzed at the level of content (the information conveyed by the passage), organization (effective arrangement of the content), language (vocabulary and grammar) used to convey the content, and presentation medium/format (visual text, graphics, or speech; Brewer & Lichtenstein, 1981). Based on this general approach, the content, organization, language and medium of each passage was revised, following guidelines from the literature on patient education (Doak, Doak, & Root, 1995), theories of discourse comprehension (Hill-Briggs, Schumann, & Dike, 2012; Lorch, Lemarié, & Grant, 2011), as well as the Process–Knowledge model of health literacy (Chin et al., 2011; Chin, Madison, et al., 2015). Similar multi-level approaches have been used to revise medication instructions commonly used in pharmacies, which improve comprehension of and memory for medication information (Boudewyns et al., 2015; Morrow et al., 2005; Pander Maat & Lentz, 2010).

We focused on hypertension because it is a common chronic illness among older adults (Go et al., 2013) and requires a range of self-care activities such as taking medication and measuring blood pressure. We addressed two research questions. First, do older adults better remember self-care information after reading the redesigned rather than typical passages? Second, do some older adults especially benefit from the redesigned passages? This is suggested by the finding that older adults with lower health literacy are less likely to understand typical self-care information (Chin et al., 2015). Guided by the Process–Knowledge framework, we examined whether PC and knowledge moderated the effect of passage redesign on memory for self-care information.

Methods

Participants

One hundred and twenty eight older adults (mean age = 70.8 years, SD = 7.7) were recruited for the study through advertisements posted in public libraries, health and recreation centers, senior centers, and hospitals in Peoria and Champaign, IL, United States. All participants provided informed consent and were compensated for their participation. Seventy-nine were women (61.7%), and 95 were diagnosed with hypertension (74.2%). Participants with hypertension were identified by self-reported physician diagnosis and use of medications prescribed for high blood pressure. For patients with hypertension, the average duration of illness was 16 years (SD = 13.4 years). Most had completed high school (85.2%); the rest had completed some high school (8.6%) or did not enter high school (6.3%). And about one third of participants reported often or sometimes reading online health information (34.4%).

Measures

Health literacy was measured by a commonly used standardized test, STOFHLA (Short Test of Functional Health Literacy in Adults; Baker, Williams, Parker, Gazmararian, & Nurss, 1999). Although most participants had adequate health literacy, 12% had marginal health literacy (N = 12), and 5.5% had inadequate health literacy (N = 7).

A PC composite measure was constructed by averaging the standardized scores of the processing speed and working memory measures. Processing speed was measured by the Letter and Pattern Comparison (Salthouse, 1991), and working memory by the Reading Span (Stine & Hindman, 1994). GK was measured by the Advanced Vocabulary Task (Ekstrom, French, Harman, & Dermen, 1976). HK was measured by a test of hypertension knowledge that was adapted from Gazmararian, et al. 2003. It consisted of 33 true/false and four multiple-choice questions (Cronbach α = .90; Chin, Fu, & Kannampallil, 2009).

Passages

Selecting the Typical Passages

Nine passages (667–1,176 words) were used in the study. These passages were selected from credible websites often used by patients, chosen based on website reputability (government and Health On the Net-certified sites) and Google page rank. The sites included National Institute on Aging, American Heart Association, the Mayo Clinic, American Family Physician, University of Maryland Medical Center, WebMD and New York Times. Thus, the passages were “typical” in the sense that they were representative of information about hypertension found on the web. From each website, 4–5 source passages were identified for each of five topics: introduction to hypertension, causes and risk factors, its complications, lifestyle changes to improve blood pressure, and the pharmaceutical treatment. For each topic, one prototype “typical” passage was created to be representative of the set of source passages. The typical passages did not differ from their source passages in terms of length (number of words, sentences, and paragraphs), text format (i.e., font size and style), presence of graphics, and readability score (Flesch-Kincaid readability grade level and reading ease). Typical and source passages were also comparable in terms of rated difficulty (ratings from 10 older adults with hypertension).

Revising Typical Passages

We analyzed and revised content, organization, language and medium of typical passages. Content revision was guided by recommendations from three medical experts (two physician and one pharmacy research-practitioners). The medical experts were experienced practicing clinicians who were also health literacy researchers. Therefore, they had in-depth knowledge of hypertension and extensive experience communicating with older patients The medical experts sometimes recommended eliminating extraneous information or elaborating some points in the revised passages with the key concepts and major contents similar across two passage types. Organization revision was to present the information in a more appropriate order (guided by experts’ decisions) and create text structures, such as the placement of paragraph breaks, titles and headers, and bullet lists, for signaling the information. Language revision was based on consensus (trained students and medical experts) about easy to understand word choice and sentence structure.

For passage medium, a concept outline was added in order to signal the most important concepts in the passage and the relations among these concepts. Such “advanced organizers” have been found to improve students’ memory when placed before the passage (Nesbit & Adesope, 2006). An earlier study also found that concept outlines improved memory for hypertension self-care passages for older adults who had more HK (Gao et al., 2011).

Two types of the passages did not differ in any text characteristics except that the revised passages were longer than the typical ones in terms of the number of words (Table 1). In a pilot study, 10 older adults gave feedback on the typical and revised versions of the passages (not labeled as such), and also indicated which they preferred. Participants indicated that they would prefer to receive self-care information in the revised format.

Table 1.

Text Characteristics of the Typical and Revised Passages

Typical Revised
Mean (SD) Mean (SD) t
Number of words 742.25 (73.78) 1008.25 (150.83) −3.34*
Number of syllables 1236.25 (252.04) 1254.5 (244.52) −2.39
Number of sentences 49 (10.95) 62.25 (10.65) −1.43
Flesh-Kincaid grade level 9.98 (2.18) 8.63 (1.61) 2.53
Open in a new tab

Note: *p < .05.

Passage Memory Measures

Question Accuracy

After reading each passage, participants answered 13 multiple-choice questions. For example, “ACE inhibitors and ARBs both block hormones that (a) widen your blood vessels, (b) constrict your blood vessels, (c) decrease the amount of fluid in your blood, (d) allow calcium to enter the cells of your arteries,” where (b) was the correct answer. Only information that was in both versions of each passage was tested.

Passage Summary

Summary protocols were transcribed verbatim and then segmented into syntactically complete units that conveyed an idea (Johnson, 1970). Each segment was coded using an established scheme (Adams, Labouvie-Vief, Hobart, & Dorosz, 1990) regarding the relatedness and accuracy of the passage information. A segment would first be categorized as a related or unrelated segment depending on whether the idea conveyed was related to the passage. For a related segment, it would be further categorized as a correct or incorrect segment depending on whether the related idea mentioned was correct or incorrect. Correct ideas included both the information directly from the passages and inferences from the information conveyed by the passages. Unrelated segments included information from other passages, participants’ personal stories, or their comments about the passages or the study (Percentage agreement among three raters: 91.8%).

Reading Efficiency

Reading efficiency, or the reading time per unit of remembered information, was also analyzed because it reflects participants’ efficiency of reading strategies (information gained for the amount of effort devoted to comprehension). Following the model of Self-Regulated Language Processing (Stine-Morrow et al., 2008), this reading efficiency measure was used to analyze the efficiency of resource allocation policies to the text as a whole (Miller, 2009). More efficient readers should be able to strategically allocate their effort to the passage to optimize their comprehension performance.

Reading time was measured for each passage (with six missing observations due to technical problems). To control for differences in passage length, we used reading time per word to create the reading efficiency measure. For passage summary, reading efficiency was operationalized as the reading time per word divided by the number of correct segments recalled for each passage; for question accuracy, reading efficiency was operationalized as the reading time per word divided by the accuracy score for each passage (Miller, 2009). In other words, reading efficiency represented the amount of time readers needed for the uptake of one unit of correct information. Measures were log transformed for later analyses.

Experimental Design

All participants read four passages, each about a different hypertension topic (causes and risk factors, complications, lifestyle changes or pharmaceutical treatment). Two of these passages were typical and two were revised. The presentation order of the topics was counterbalanced across participants. Presentation of passages was also blocked by passage type (typical and revised), with block order counterbalanced across participants.

Procedure

Participants completed the demographic, cognitive, hypertension knowledge and health literacy measures, and then read the passages on a computer. The text was displayed in black Arial 12-point font on a white background (Figure 1). They read five passages in total: the practice introduction passage, two typical passages and two revised passages. Participants were instructed to read the passages for comprehension and to be able to answer questions about the content. They read one passage at a time and could scroll up and down the screen at their own pace, with a limit of 9 min per passage. All participants finished reading within the 9-min limit. After reading each passage, participants first summarized the main points of the passage, and then answered multiple-choice questions.

Figure 1.

Figure 1.

Open in a new tab

Screenshots of a typical (left) and revised passage (right) on the display.

Results

Correlates of Reading Efficiency

Guided by the Process–Knowledge model (Chin et al., 2011), we examined the relationship among HK, GK, PC, and reading efficiency. As seen in Table 2, participants with better PC, GK and HK spent less time to uptake one unit of information in terms of the correct segments recalled (r = −.43, r = −.42, r = −.24, all ps < .01) and the question accuracy (r = −.33, r = −.46, r = −.25, all ps < .01). The two reading efficiency measures were also correlated (r = .64, p < .001).

Table 2.

Correlations Among Individual Difference Variables and Reading Efficiency Performance

Age PC GK HK REq REs STOFHLA
Age −0.02 0.24* 0.07 −0.08 −0.09 −0.05
PC 0.26* 0.17 −0.33* −0.43* 0.38*
GK 0.31* −0.46* −0.42* 0.40*
HK −0.25* −0.24* 0.20**
REq 0.64* −0.45*
REs −0.33*
Open in a new tab

Notes: GK = general knowledge, HK = health knowledge; PC = processing capacity; REq = log transformed reading efficiency score based on question accuracy (reading time per word divided by the accuracy score of each passage); REs= log transformed reading efficiency score based on summary performance (reading time per word divided by the number of correct segments recalled for each passage); STOFHLA = Short Test of Functional Health Literacy in Adults.

*p < .01. **p < .05.

Effects of Passage Revision

We used paired t tests to investigate the affect of passage revision on comprehension performance.

Question Accuracy

Participants better remembered the revised (M = 0.74, SD = 0.14) compared to the typical passages (M = 0.70, SD = 0.11; t(127) = −3.08, p < .01). They also read the revised passages more efficiently (M = 0.50, SD = 0.23) than the typical ones (M = 0.60, SD = 0.25; t(121) = 5.98, p < .001).

Summary Accuracy

First, there was no difference in the total number of segments and the number of related segments recalled for typical and revised passages (Total segments: t(121) = −0.87, p = .39; typical: M = 11.18, SE = 0.62; revised: M = 11.56, SE = 0.68; Related segments: t(121) = −1.62, p = .11; typical: M = 8.04, SE = 0.53; revised: M = 8.73, SE = 0.56). Therefore, we report analyses for raw number of segments recalled rather than percentage of segments recalled (analyses using percentage of segments recalled produced similar results).

We used the number of correct segments recalled per passage as the measure of summary accuracy. Participants’ summaries contained more correct ideas from the revised passages (M = 7.34; SE = 0.46) than from the typical passages (M = 6.64; SE = 0.47) (t(121) = −1.95, p = .05). Further, participants needed less time to achieve the same level of summary accuracy for revised passages (M = 0.08; SE = 0.01) than for typical ones (M = 0.11; SE = 0.01) (t(110) = 2.46, p < .05). These results on summary accuracy were consistent with the results on question accuracy.

Who Benefits More From Passage Revision?

Linear mixed effects models were used to analyze the effects of passage type (typical vs revised) as well as individual difference variables (age, PC, GK, HK) on the reading efficiency measure when controlling the random effects of subjects and passages. We used PROC MIXED function in SAS to analyze the data.

Question Accuracy

We first examined the fixed effects of passage type, PC, and their interaction on reading efficiency (Model 1 in Table 3). There was no interaction, showing that participants better remembered the revised passage regardless of PC level (B = −0.01, SE = 0.01, t = −1.33, p = .18). Similarly, passage type did not interact with GK on reading efficiency (Model 2 in Table 3), suggesting that participants better understood the revised passages than the typical ones regardless of GK level (B = 0.001, SE = 0.01, t = 0.21, p = .84).

Table 3.

Effects of Individual Difference Variables and Types of Passages on Reading Efficiency (Question Accuracy) in Both Typical and Revised Passages

Model 1 Model 2 Model 3 Model 4
B t B t B t B t
Intercept −0.35 (0.03) −13.51* −0.34 (0.03) −13.09* −0.34 (0.03) −12.75* −0.40 (0.12) −3.23*
Item predictors
 Pass −0.04 (0.01) −7.81* −0.04 (0.01) −8.08* −0.04 (0.005) −8.13* −0.04 (0.01) −7.79*
Subject predictors
 Age 0.001 (0.002) 0.47
 PC −0.08 (0.02) −4.05* −0.05 (0.02) −2.75*
 GK −0.08 (0.01) −5.22* −0.06 (0.02) −3.54*
 HK −0.05 (0.02) −3.33* −0.03 (0.02) −1.58
Cross-level interaction
 PC × Pass −0.01 (0.01) −1.33
 GK × Pass 0.001 (0.01) 0.21
 HK × Pass −0.01 (0.01) −2.23* −0.01 (0.01) −1.94*
Open in a new tab

Notes: GK = general knowledge, HK = health knowledge; Pass = types of passages; PC = processing capacity.

*p < .05.

However, a similar analysis revealed a significant interaction of HK and passage type (Model 3 in Table 3; B = −0.01, SE = 0.01, t = −2.23, p < .05). Participants with more rather than less HK benefited more from the revised passages. Moreover, this interaction remained significant after the effects of age, PC and GK were controlled (B = −0.01, SE = 0.01, t = −2.23, p < .05; see Model 4 in Table 3). We plotted the time needed to uptake one unit of information from the typical and revised passages for participants who had 1 SD below and above the mean HK performance in Figure 2.

Figure 2.

Figure 2.

Open in a new tab

Interaction of health knowledge and types of passages on reading efficiency (raw reading time per unit of information uptake).

Summary Accuracy

Similarly, as shown in Table 4, PC, GK and HK facilitated the uptake of correct ideas (PC: B = −0.18, SE = 0.04, t = −4.61; GK: B = −0.14, SE = 0.03, t = −4.88; HK: B = −0.09, SE = 0.03, t = −2.73, all ps < .05). In addition, less time was needed to uptake correct ideas from revised than from typical passages (PC: B = −0.06, SE = 0.01, t = −5.55; GK: B = −0.06, SE = 0.01, t = −5.21; HK: B = −0.06, SE = 0.01, t = −5.33, all ps < .05). However, there was no interaction between the passage type and the individual difference variables, suggesting that participants benefitted from passage revision regardless of their levels of cognitive resources (PC: B = 0.002, SE = 0.02, t = 0.1, p = .92; GK: B = −0.01, SE = 0.01, t = −1.24, p = .22; HK: B = −0.003, SE = 0.01, t = −0.25, p = .80).

Table 4.

Effects of Individual Difference Variables and Types of Passages on Reading Efficiency (the Number of Correct Segments Recalled) in Both Typical and Revised Passages

Model 1 Model 2 Model 3
B t B t B t
Intercept −1.27 (0.06) −21.46* −1.24 (0.06) −20.38* −1.25 (0.06) −20.22*
Item predictors
 Pass −0.06 (0.01) −5.55* −0.06 (0.01) −5.21* −0.06 (0.01) −5.33*
Subject predictors
 PC −0.18 (0.04) −4.61*
 GK −0.14 (0.03) −4.88*
 HK −0.09 (0.03) −2.73*
Cross-level interaction
 PC × Pass 0.002 (0.02) 0.1 −0.01 (0.01) −1.24
 GK × Pass −0.003 (0.01) −0.25
 HK × Pass
Open in a new tab

Notes: Pass = types of passages; PC = processing capacity; GK = general knowledge, HK = health knowledge.

*p < .05.

Discussion

The study shows that older adults’ memory for hypertension self-care information available on the internet can be improved by a systematic, multi-faceted approach to revising the information, which results in documents that reduce comprehension demands on health-literacy-related cognitive resources. Better memory for self-care information may in turn translate into more effective health-related decisions, behaviors, and outcomes because memory for self-care information has been found to predict health behaviors (Dewalt, Berkman, Sheridan, Lohr, & Pignone, 2004). Of course this link needs to be demonstrated for the present passages in future research.

We found that the revised passages enabled older adults with more HK to obtain information more efficiently. According to the Process–Knowledge model (Chin et al., 2011), PC and knowledge interact to influence age differences in comprehension because these abilities have different age-related trajectories (declines for PC vs gains for knowledge). Guided by this approach, we revised the passages to reduce comprehension demands on PC (e.g., by simplifying language and more clearly signaling organization), as well as to promote use of prior knowledge in order to understand new information (e.g., by including an advanced organizer and headers). The revised passages may have enabled older adults with more HK to obtain information from these passages more efficiently because they could use structural features such as the concept list to scaffold their use of knowledge to elaborate information into a situation model (also see Gao et al., 2011; Morrow et al., 2012). However, conclusions about knowledge-based moderation of passage revision effects are tentative because revision benefits for the passage summary were not influenced by knowledge. Nonetheless, it is possible that knowledge influenced revision effects for the question rather than summary measure because the multiple-choice questions provided additional support for knowledge use. At the same time, while HK may exaggerate the benefit of passage revision, even older adults with fewer resources (lower PC and knowledge) benefited from the revised passages, suggesting that passage revision did help older adults varying with abilities better remember and uptake the information.

A notable aspect of our findings is that the revised passages were better remembered than the typical ones, even though the two versions did not differ in average readability scores. Readability is an important aspect of health documents that can influence patient comprehension (Friedman & Hoffman-Goetz, 2007) and can be a barrier to accessing web-based health information because many web documents exceed recommended readability levels (McInnes & Haglund, 2011). Nonetheless, our study suggests that increasing document readability is insufficient for improving older patients’ comprehension. What is needed is a more comprehensive, theory-based, multi-faceted approach to design that targets features of document content, organization, medium as well as language, which jointly influence comprehension. An important challenge is to develop tools to help evaluate and revise online information at multiple levels, perhaps by building on existing multi-faceted measures such as the Suitability Assessment of Materials (Doak et al., 1995). An important challenge during the revision process was to determine an effective order for presenting the important ideas related to the passage topic. Some previous work has focused on the organization of procedural health texts such as medication instructions (e.g., Morrow, Leirer, Andrassy, Tanke, & Stine-Morrow, 1996), but less is known about effective organization for other types of self-care information. In our study, we relied on the guidance of medical experts as well as theories of text organization (e.g., Lorch et al., 2011). We also relied on these theories to decide how to effectively signal this organization, such as the use of advanced organizers to enhance the integration and retention of important ideas. Computational approaches to analyzing large corpora of web-based patient information may help identify effective structures that support patient comprehension (e.g., Ramesh, Houston, Brandt, Fang, & Yu, 2013). More generally, we note that this multi-faceted redesign is important for patient communication, and could, for example, apply to other written healthcare materials (e.g., post-visit patient summaries) in order to promote HK, decision-making, and outcomes among diverse patients.

There are some limitations of the study. The use of a multi-faceted approach to revise health information resulted in passages that differed from the typical passages in multiple ways. Therefore, we were unable to pinpoint which aspects of the revision were most important for improving learning. However, we note that multi-component interventions are often used to improve self-care behaviors and health outcomes, and high-intensity, multimodal interventions are more likely than less complex interventions to influence behaviors and improve outcomes (Roter et al., 1998; Sheridan et al., 2011). Nonetheless, identifying which facets of document redesign have the greatest affect on patient comprehension may improve the efficiency of interventions by focusing design efforts on the most critical characteristics. Also, the passages were not evaluated within a website context, so that our estimates of passage redesign benefits did not factor in issues such as website usability. We also note that, like other complex interventions, our multi-faceted approach to revising online documents may require more resources than approaches to writing typical online health documents. Another limitation was that the majority of the older adults in our sample had adequate health literacy, with a smaller proportion showing marginal and inadequate health literacy compared to our previous studies (Levinthal et al., 2008; Morrow et al., 2006). However, we note that similar patterns of association between health literacy and cognitive ability measures were found in the present study and in these previous studies. This finding bolsters confidence that the results relating cognitive ability and passage revision benefits in the current study would generalize beyond our sample of participants.

Despite these limitations, our study did find that a multi-faceted approach to redesigning health information guided by theories of text comprehension and document design improves older adults’ memory for self-care information typically found on credible websites. This approach could be broadly adopted by health information websites. In future research we hope to extend the approach to promote patient comprehension in other health contexts, such as personal health records (such as lab reports, treatment records and medication information) that can be accessed through patient portals to electronic health records systems, so that older adults can take a more active role in managing their illness and maintaining their health.

Funding

This research was supported by the National Institute of Aging (R01 AG031718). Any opinions, findings, and conclusions or recommendations expressed in this publication are those of the authors and do not necessarily reflect the views of the National Institute of Health.

Acknowledgments

We thank the research coordinators at the Saint Francis Medical Center for their assistance with data collection. We also thank undergraduate assistants in that Human Factors Lab at the University of Illinois at Urbana Champaign for their help with developing the materials and data coding. Portions of the data were presented at the 37th Annual Cognitive Science Society Meeting in Pasadena, California in 2015.

References

  1. Adams C. Labouvie-Vief G. Hobart C. J., & Dorosz M (1990). Adult age group differences in story recall style. Journal of Gerontology, 45, P17–P27. doi:10.1093/geronj/45.1.p17 [DOI] [PubMed] [Google Scholar]
  2. Baker D. W. (2006). The meaning and the measure of health literacy. Journal of General Internal Medicine, 21, 878–883. doi:10.1111/j.1525-1497.2006.00540.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Baker D. W. Gazmararian J. A. Sudano J., & Patterson M (2000). The association between age and health literacy among elderly persons. The Journals of Gerontology Series B, Psychological Sciences and Social Sciences, 55, S368–S374. doi:10.1093/geronb/55.6.S368 [DOI] [PubMed] [Google Scholar]
  4. Baker D. W. Williams M. V. Parker R. M. Gazmararian J. A., & Nurss J (1999). Development of a brief test to measure functional health literacy. Patient Education and Counseling, 38, 33–42. doi:10.1016/s0738-3991(98)00116-5 [DOI] [PubMed] [Google Scholar]
  5. Baltes P. B. (1997). On the incomplete architecture of human ontogeny. Selection, optimization, and compensation as foundation of developmental theory. The American Psychologist, 52, 366–380. doi:10.1037/0003-066X.52.4.366 [DOI] [PubMed] [Google Scholar]
  6. Beier M. E., & Ackerman P. L (2005). Age, ability, and the role of prior knowledge on the acquisition of new domain knowledge: promising results in a real-world learning environment. Psychology and Aging, 20, 341–355. doi:10.1037/0882-7974.20.2.341 [DOI] [PubMed] [Google Scholar]
  7. Berkman N. D. Davis T. C., & McCormack L (2010). Health literacy: what is it?Journal of Health Communication, 15, 9–19. doi:https://doi.org/10.1080/10810730.2010.499985 [DOI] [PubMed] [Google Scholar]
  8. 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, 97–107. doi:10.7326/0003-4819-155-2-201107190-00005 [DOI] [PubMed] [Google Scholar]
  9. Boudewyns V. O’Donoghue A. C. Kelly B. West S. L. Oguntimein O. Bann C. M., & McCormack L. A (2015). Influence of patient medication information format on comprehension and application of medication information: A randomized, controlled experiment. Patient Education and Counseling, 98, 1592–1599. doi:10.1016/j.pec.2015.07.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Brewer W. F., & Lichtenstein E. H (1981). Event schemas, story schemas, and story grammars (Vol. 9, pp. 363–379). Hillsdale, NJ: Erlbaum. [Google Scholar]
  11. Chin J., Fu W-T., & Kannampallil T. (2009). Adaptive information search: Age-dependent interactions between cognitive profiles and strategies. In Proceedings of the 27th ACM Conference on Human Factors in Computing Systems CHI’09 (pp. 1683–1692). Boston: ACM Press. doi:10.1145/1518701.1518961 [Google Scholar]
  12. Chin J., & Fu W.-T (2010). Interactive effects of age and interface differences on search strategies and performance. In Proceedings of the 28th international conference on Human factors in computing systems - CHI ‘10 (pp. 403–412). Atlanta, GA: ACM Press. doi:10.1145/1753326.1753387 [Google Scholar]
  13. Chin J. Madison A. Gao X. Graumlich J. F. Conner-Garcia T. Murray M. D.,…Morrow D. G (2015). Cognition and health literacy in older adults’ recall of self-care information. The Gerontologist. doi:10.1093/geront/gnv091 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Chin J. Morrow D. G. Stine-Morrow E. A. L. Conner-Garcia T. Graumlich J. F., & Murray M. D (2011). The Process-Knowledge Model of health literacy: Evidence from a componential analysis of two commonly used measures. Journal of Health Communication, 16, 222–241. doi:10.1080/10810730.2011.604702 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Chin J. Payne B. Gao X. Conner-Garcia T. Graumlich J. F. Murray M. D.,…Stine-Morrow E. A (2015). Memory and comprehension for health information among older adults: distinguishing the effects of domain-general and domain-specific knowledge. Memory, 23, 577–589. doi:10.1080/09658211.2014.912331 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. 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, 1228–1239. doi:10.1111/j.1525-1497.2004.40153.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Doak C. C. Doak L. G., & Root J. H (1995). Teaching patients with low literacy skills (2nd ed). Philadelphia: LWW. [Google Scholar]
  18. Ekstrom R. B. French J. W. Harman H. H., & Dermen D (1976). Manual for kit of factor-referenced cognitive tests. Princeton, NJ: Educational Testing Service. [Google Scholar]
  19. Friedman D. B., & Hoffman-Goetz L (2007). An exploratory study of older adults’ comprehension of printed cancer information: is readability a key factor?Journal of Health Communication, 12, 423–437. doi:10.1080/10810730701438658 [DOI] [PubMed] [Google Scholar]
  20. Gao X. Bertel S. Chin J. Madison A. Morrow D. G. Stine-Morrow E. A. L.,…Murray M (2011, October). The effects of domain knowledge and concept map on processing and memory for hypertension-related text. Poster presented at Health Literacy Annual Research Conference, Chicago. [Google Scholar]
  21. Gazmararian J. A. Williams M. V. Peel J., & Baker D. W (2003). Health literacy and knowledge of chronic disease. Patient Education and Counseling, 51, 267–275. doi:10.1016/s0738-3991(02)00239-2 [DOI] [PubMed] [Google Scholar]
  22. Go A. S. Mozaffarian D. Roger V. L. Benjamin E. J. Berry J. D. Borden W. B.,…Turner M. B; American Heart Association Statistics Committee and Stroke Statistics Subcommittee (2013). Heart disease and stroke statistics–2013 update: a report from the American Heart Association. Circulation, 127, e6–e245. doi:10.1161/CIR.0b013e31828124ad [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Hill-Briggs F. Schumann K. P., & Dike O (2012). Five-step methodology for evaluation and adaptation of print patient health information to meet the < 5th grade readability criterion. Medical Care, 50, 294–301. doi:10.1097/MLR.0b013e318249d6c8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Johnson R. E. (1970). Recall of prose as a function of the structural importance of the linguistic units. Journal of Verbal Learning and Verbal Behavior, 9, 12–20. doi:10.1016/S0022-5371(70)80003-2 [Google Scholar]
  25. Kintsch W. (1998). Comprehension: A paradigm for cognition. New York: Cambridge University Press. [Google Scholar]
  26. Levinthal B. R., Morrow D. G., Tu W., Wu J., & Murray M. D. (2008). Cognition and health literacy in patients with hypertension. Journal of General Internal Medicine, 23, 1172–1176. doi:10.1007/s11606-008-0612-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Lorch R. F. J. Lemarié J., & Grant R. A (2011). Three information functions of headings: A test of the SARA theory of signaling. Discourse Processes, 48, 139–160. doi:10.1080/0163853X.2010.503526 [Google Scholar]
  28. McInnes N., & Haglund B. J (2011). Readability of online health information: implications for health literacy. Informatics for Health & Social Care, 36, 173–189. doi:10.3109/17538157.2010.542529 [DOI] [PubMed] [Google Scholar]
  29. Meyer B. J. F. (2003). Text coherence and readability. Topics in Language Disorders, 23, 204–224. doi:10.1097/00011363-200307000-00007 [Google Scholar]
  30. Miller L. M. S. (2009). Age differences in the effects of domain knowledge on reading efficiency. Psychology and Aging, 24, 63–74. doi:10.1037/a0014586 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Morrow D., & Chin J (2015). Decision making and health literacy among older adults. In Hess T. Strough J., & Lockenhoff C. (eds.), Aging and decision making (pp. 261–282). San Diego, CA: Academic Press. doi:10.1016/b978-0-12-417148-0.00013-3 [Google Scholar]
  32. Morrow D. D’andrea L. Stine-Morrow E. A. L. Shake M. Bertel S. Chin J.,…Murray M (2012). Comprehension of multimedia health information among older adults with chronic illness. Visual Communication, 11, 347–362. doi:10.1177/1470357212446413 [Google Scholar]
  33. Morrow D. Clark D. Tu W. Wu J. Weiner M. Steinley D., & Murray M. D (2006). Correlates of health literacy in patients with chronic heart failure. The Gerontologist, 46, 669–676. doi:10.1093/geront/46.5.669 [DOI] [PubMed] [Google Scholar]
  34. Morrow D. G. Leirer V. O. Andrassy J. M. Tanke E. D., & Stine-Morrow E. A (1996). Medication instruction design: younger and older adult schemas for taking medication. Human Factors, 38, 556–573. doi:10.1518/001872096778827305 [DOI] [PubMed] [Google Scholar]
  35. Morrow D. G. Weiner M. Young J. Steinley D. Deer M., & Murray M. D (2005). Improving medication knowledge among older adults with heart failure: a patient-centered approach to instruction design. The Gerontologist, 45, 545–552. doi:10.1093/geront/45.4.545 [DOI] [PubMed] [Google Scholar]
  36. Nesbit J. C., & Adesope O. O (2006). Learning with concept and knowledge maps: A meta-analysis. Review of Educational Research, 76, 413–448. doi:10.3102/00346543076003413 [Google Scholar]
  37. Nielsen-Bohlman L. Panzer A. M., & Kindig D. A (2004). Health literacy: A prescription to end confusion. Washington, DC: The National Academies Press. [PubMed] [Google Scholar]
  38. Pander Maat H., & Lentz L (2010). Improving the usability of patient information leaflets. Patient Education and Counseling, 80, 113–119. doi:10.1016/j.pec.2009.09.030 [DOI] [PubMed] [Google Scholar]
  39. Ramesh P., Houston T., Brandt C., Fang H., & Yu H. (2013). Improving patients’ electronic health record comprehension with NoteAid. Studies in Health Technology and Informatics, 192, 714–718. [PubMed] [Google Scholar]
  40. Roter D. L. Hall J. A. Merisca R. Nordstrom B. Cretin D., & Svarstad B (1998). Effectiveness of interventions to improve patient compliance: a meta-analysis. Medical Care, 36, 1138–1161. doi:10.1097/00005650-199808000-00004 [DOI] [PubMed] [Google Scholar]
  41. Salthouse T. A. (1991). Mediation of adult age differences in cognition by reductions in working memory and speed of processing. Psychological Science, 2, 179–183. doi:10.1111/j.1467–9280.1991.tb00127.x [Google Scholar]
  42. Sharit J. Hernández M. A. Czaja S. J., & Pirolli P (2008). Investigating the roles of knowledge and cognitive abilities in older adult information seeking on the web. ACM Transactions on Computer-Human Interaction, 15, 3. doi:10.1145/1352782.1352785 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Sheridan S. L. Halpern D. J. Viera A. J. Berkman N. D. Donahue K. E., & Crotty K (2011). Interventions for individuals with low health literacy: a systematic review. Journal of Health Communication, 16, 30–54. doi:10.1080/10810730.2011.604391 [DOI] [PubMed] [Google Scholar]
  44. Stine E. A. L., & Hindman J (1994). Age differences in reading time allocation for propositionally dense sentences. Aging, Neuropsychology, and Cognition, 1, 2–16. doi:10.1080/09289919408251446 [Google Scholar]
  45. Stine-Morrow E. A. L., & Miller L. M. S (2009). Aging, self-regulation, and learning from text. In Ross B. (ed.), The psychology of leanring and motivation (Vol. 51, pp. 255–296). San Diego, CA: Academic Press. doi:10.1016/s0079-7421(09)51008-0 [Google Scholar]
  46. Stine-Morrow E. A. Soederberg Miller L. M. Gagne D. D., & Hertzog C (2008). Self-regulated reading in adulthood. Psychology and Aging, 23, 131–153. doi:10.1037/0882-7974.23.1.131 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Taha J. Czaja S. J. Sharit J., & Morrow D. G (2013). Factors affecting usage of a personal health record (PHR) to manage health. Psychology and Aging, 28, 1124–1139. doi:10.1037/a0033911 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Wolf M. S. Gazmararian J. A., & Baker D. W (2005). Health literacy and functional health status among older adults. Archives of Internal Medicine, 165, 1946–1952. doi:10.1001/archinte.165.17.1946 [DOI] [PubMed] [Google Scholar]
  49. Wright P. (1999). Writing and information design of healthcare materials. In Candlin C., Hyland K. (eds.), Writing: texts, processes and practices (pp. 85–98). London: Longman. [Google Scholar]
  50. Zickuhr K., & Madden M (2012). Older adults and internet use. Retrieved from http://www.pewinternet.org/2012/06/06/older-adults-and-internet-use/ [Google Scholar]

Articles from The Gerontologist are provided here courtesy of Oxford University Press

RESOURCES