Listenability as a Tool for Advancing Health Literacy

Abstract

Health literacy research and practice has focused mainly on the readability of written documents. Yet oral communication plays at least as important a role in the interpersonal ecology in which people make real decisions about their health. Moreover, the single-minded quest for short sentences and simple vocabulary inherent in the readability paradigm can subvert the effort to engage in patient- or consumer-centered communication. Listenability is the quality of discourse that eases the cognitive burden that aural processing imposes. Listenability is a function of oral-based language plus “considerate” rhetorical structures. The Listenability Style Guide presented in this article offers evidence-based recommendations for producing listenable discourse. A study testing the applicability of the Listenability Style Guide to postsurgical discharge instructions was conducted. College students either heard or read discharge instructions composed in either high or moderate listenability. Comprehension was higher for this population in reading than in listening. Across modalities, the high listenability version was easiest to comprehend. Incorporating listenability concerns in research and practice is consistent with emerging, broad conceptions of health literacy and with the dictates of the National Action Plan to Improve Health Literacy.

Improving the readability and usability of documents conveying health information to patients and consumers has led to vast improvements in health literacy (e.g., Davis, Holcombe, Berkel, Pramanik, & Divers, 1998; Philipson et al., 1999). Yet, in the highly interpersonal ecology in which patients and consumers make real decisions about their health (Fox & Jones, 2009; Peerson & Saunders, 2009), oral communication about health information is arguably at least as important as written communication (Rosenfeld et al., 2011; Roter, 2011; Rubin, Freimuth, Parmer, Kaley, & Okundaye, 2011; Schonlau, Martin, Haas, Derose, & Rudd 2011). The spoken word predominates in such crucial health situations as conversations with lay friends and relatives about the pros and cons of alternative medical treatments, listening to radio broadcasts about precautions to take in an environmental disaster, watching direct-to-consumer pharmaceutical ads—as well as in exam room consultations with healthcare providers. Yet, the field of health literacy lacks a construct analogous to readability that could help advance the quality of spoken health messages. Toward that end, this article introduces the notion of health message listenability. Though the listenability construct deliberately eschews the reductionistic formulae that characterize readability research and a practice, this article does offer a menu of language and discourse features generally associated with listenability. A pilot study that applies listenability analysis to surgical discharge instructions suggests the benefits of implementing listenability guidelines.

Many different discourse features determine a reader's capacity to comprehend, retain, and evaluate information appearing in written documents in general (Graesser, Millis, & Zwaan, 1997), and in written health messages in particular (Houts, C. C. Doak, L. G. Doak & Loscalzo, 2006; Neuhauser & Kala, 2011; Wilson & Wolf, 2009). These features include recognizable genre schemata, organizational signaling, use of white space, and integration of inference-relevant graphics. Of course, common and easily decoded words and simple sentence structures—the “plain language” components from which readability levels are calculated—also help predict reading comprehension in general (Davison, 1984; Fry, 1989), and comprehension of written health messages in particular (Dowe, Lawrence, Carlson & Keyserling, 1997; Friedman & Hoffman-Goetz, 2006).

In practice, however, composing materials to meet health literacy standards often devolves into a unidimensional pursuit for low grade-level readability scores. Thus, for example health literacy guidelines such as The Group Health Research Institute's PRISM (Ridpath, Greene & Wiese, 2007; similarly, see Covering Kids and Families, 2005) recommends that materials conform to eighth-grade or lower readability levels. Other authorities prescribe even lower readability-level targets (Badarudeen & Sabharwal, 2010). The U.S. Department of Health and Human Services' Simply Put guide enjoins health communicators to “[u]se words with one or two syllables when you can. Keep most sentences, if possible, between eight to ten words and limit paragraphs to three to five sentences” (2010b, p. 6).

This single-minded quest for short sentences and simple vocabulary has been criticized on a number of grounds for texts in general (Holland & Campbell, 1982) and for health messages in particular (Zarcadoolas, 2010). Because it is not difficult to manipulate readability levels to meet easily quantifiable benchmarks (“writing to formula;” Armbruster, Osborn & Davison, 1985), writers and editors can readily obsess about microlinguistic structures rather than attending to the broader, multidimensional process of helping audiences construe meaning in messages (Selzer, 1981).

The seductiveness of readability formulae is one factor leading the field of health literacy to relegate relatively little attention to spoken communication. (Another factor contributing to the neglect of health literacy in oral interaction is the dearth of convenient instruments for measuring that skill. See Rubin and colleagues, 2011.) Yet, it is clear that patients and consumers acquire (or fail to acquire) much of their health information not by reading, but by listening. Many people discard the myriads of printed health promotion brochures and newsletters with which they are inundated (Parvanta, Myers, DiCarlo, & Burgh, 2011). Even in the face of highly accessible health information, both in print and online, face-to-face interactions remain the most trusted source of health information (Hesse et al., 2007). Moreover, even when consumers do go to the Internet for health information, they typically filter the significance of that information by discussing it with health professionals and with members of their informal social networks (Fox et al., 2009). Health information seeking and utilization—the ubiquity of print and new technologies notwithstanding—remain highly social behaviors embedded in interpersonal interaction (Longo et al., 2010). Moreover, accessing health information using new technologies does not diminish communicators' reliance on the spoken word, given that multimedia platforms such as YouTube have become standard vehicles for health promotion messages (Hayanga & Kaiser, 2008). In short, health literacy requires a great deal of listening (Rosenfeld et al., 2011; Schonlau et al., 2011).

Notwithstanding some reading research that treats listening as merely reading by ear (Sticht & James, 1984), the listening process does differ from reading in terms of certain challenges as well as certain advantages. For example, listeners can take advantage of prosodic cues to help clarify the informational structure of discourse units (Wennerstrom, 2001). On the other hand, spoken language is “fast fading.” Thus listeners are generally at the mercy of the speaker's rate of production, unable to slow down to decode more complex syntax or to backtrack to earlier sections where they may have lost the thread of an argument.

Listenable discourse (Rubin, 1993; Rubin, Hafer & Arata, 2000; Rubin & Rafoth, 1986) is characterized by linguistic and rhetorical structures that ease the particular cognitive burdens listeners face, that is, the challenges of processing language in real time and without a visual trace. Rubin (1993) offers the dictum, LISTENABILITY = ORAL-BASED LANGUAGE+CONSIDERATENESS. Because speakers—unlike writers—generally must compose under the same psycholinguistic limitations that constrain the listeners who receive their messages (e.g., processing language serially in real time), many of the features of oral-based language correspond to the features that enhance listening comprehension. For example, compared to writing, it is hard to orally compose a sentence that contains multiple embedded clauses. (Writing-typical syntax with embedded clauses: “Because the patient was apparently allergic to albuterol, when he was administered treatment for an asthma episode that persisted, he tested high for acidosis, because his bronchial spasms were just exacerbated.” Speaking-typical syntax with coordinated clauses: “The patient was administered treatment for a persistent asthma episode, but he was apparently allergic to albuterol, and so his bronchial spasms were just exacerbated, and then he ended up testing high for acidosis.”) Thus when speakers produce oral language, their sentences gain heft by stringing together coordinated propositions more so than by enveloping clauses one within another. Correspondingly, listenable prose is marked by coordinating conjunctions (“and,” “so,” “then”, “but”) more than by subordination (“because,” “although,” “until”). The second element of the listenability equation, considerate prose (Arbruster & Anderson, 1985), is language that helps listeners predict what is coming up in the stream of discourse. For example, because everyone is familiar with how stories work, messages that utilize prototypical narrative structure tend to be considerate, and therefore listenable:

So there's this kid named Kevin who used to live on my block, right? And one afternoon around five he decides he's just going to ride his bike down to the school yard to see who's there, ya know? Like play some basketball or something. But he doesn't put on his bike helmet because I guess he figures it's just a few blocks and he doesn't want to look dorky…

“The Listenability Style Guide” reproduced in the Appendix is intended to assist communicators in crafting messages for audiences to hear rather than to read. It was developed by reviewing linguistic research on differences between oral-based and writing-based discourse (Biber & Conrad, 2009; Harris, 2009; Horowitz & Samuels, 1987; Lawrence & Snow, 2003; Rubin et al., 2000) The Listenability Guide functions more as a menu of stylistic resources rather than as a checklist of obligatory features. That is, by no means does a message require all of the elements listed in the guide in order to qualify as listenable. Nor does the guide by itself offer any easy rubric for quantifying listenability. It would not be reasonable, for example, to rank messages by listenability by counting the number of different listenable features each contains. In its present form, the Listenability Style Guide is a qualitative tool for conveying evidence-based recommendations for enhancing listenability.

Applying the Listenability Framework to Postsurgical Discharge Instructions

Using listenability guidelines similar to those appearing in the guide presented in the Appendix, Rubin and colleagues (2000) located naturally occurring samples of prototypical oral-based and literate-based expository prose (magazine articles and speeches on identical topics and ghostwritten by the same author). Subsequent computer-assisted linguistic analysis (Biber, Johansson, Leech, Conrad, & Finegan, 1999) confirmed that the oral-based samples, as expected, contained more markers of involvement, concreteness, and context-dependence. The oral-based and the literate-based passages were presented to college student listeners and readers in both speech and in writing. Results indicated that these competent readers could comprehend literate-based prose as well as oral based when reading. But when listening, oral-based versions led to at least marginally better comprehension. In short, this study found that message listenability matters for listeners, but less so for readers.

Principles of listenability have not previously been applied to health messages that are often delivered orally. Postprocedure discharge instructions constitute one context in which much concern has been raised about patients' comprehension and retention of orally conveyed information (Clarke et al., 2005; Crane, 1997). A typical finding is that at least one third of patients discharged from emergency departments fail to understand the prescribed aftercare and that most of those patients are unaware of their lack of understanding (Engel et al., 2009). Three days after discharge, fewer than half of older adult patients recall receiving any discharge instructions at all (Flacker, Park, & Sims, 2007). Health literacy is commonly linked to patient safety (Murphy-Knoll, 2007), and effective communication at time of hospital discharge is a key factor in improving ambulatory patient safety and reducing readmissions (Kripalani, Jackson, Schnipper & Coleman, 2007). Even when experts focus on problematic discharge instructions as a problem in written patient materials, they typically reach the conclusion that the remedy lies in improved oral communication at discharge or follow-up (e.g., Chugh, Williams, Grigsby & Coleman, 2009). Accordingly, the present study examines the effects of listenability modification on comprehension of discharge instructions. It is predicted that in oral presentation, highly listenable versions of a health message will be comprehended better, and with less expended cognitive effort, that an otherwise comparable message of moderate listenability. Further, it is predicted that when participants read printed versions of the messages, this advantage for listenability will evaporate.

Method

Participants

Fifty-seven undergraduate students at a large university in the Southeastern United States participated in this study to fulfill a course requirement to serve in a research pool. Sixty-eight percent were female. Average age was 20.23 years (SD = 3.92 years). Because these participants were young, well educated, and healthy, it was anticipated that any effects of message manipulations in this study would constitute a conservative estimate of their likely impact among a more typical patient population.

Stimulus Discharge Messages

The message manipulation process began with a written set of postsurgical instructions, obtained from a large cardiac and thoracic surgery practice group, for open heart surgery patients. The original document was edited for length (810 words) and to remove location-specific references. This version constituted the moderate listenability condition and included such passages as follows:

It is expected and normal to notice an occasional “clicking noise” or sensation in the chest cavity in the first days after surgery. This clicking should occur less often with time. It should dissipate completely within the first couple of weeks. If it gets worse, however, that is a cause for alarm. In that case, a doctor should be consulted.

The Flesch-Kinkaid readability level for the moderate listenability passage was Grade 8 (62 on the Flesch Reading Ease Scale). Thus, a deliberate effort was made to make sure that moderate listenability was not conflated with low readability. Similarly, by selecting a passage that in its original form was already rather patient-friendly, this study avoids extreme comparisons that would otherwise deny advances in health literacy that have already infused discharge instructions with plain language.

To construct the high listenability version of the discharge instructions, the listenability guidelines in the Appendix were applied as completely as possible, while still maintaining a natural sounding text. The resulting high listenability revision included passages such as the following:

Barry Thompson called us in the first days after his heart surgery to say he heard an occasional “clicking noise” and a clicking feeling in his chest. Barry's clicking noise got less frequent each day, and it went away completely within the first couple of weeks. Don't worry if you hear the same clicking noise Barry heard. That's normal. But let's say you start hearing more and more clicking. Then you better call us.

The resulting high listenability text was slightly longer than the low listenability version (870 words), and slightly more readable (Flesch-Kincaid grade 6.5; Flesch Reading Ease 73).

Both the high and the low listenable texts were audio recorded by a male speaker of Standard American English who has experience teaching pronunciation. The speaking rate for both versions was comparable (143 and 133 words per minute). Print copies of both the high and the low listenable versions were prepared using a 14-point print font.

Comprehension and Process Measures

Three dependent measures were analyzed in this study: First, comprehension was measured by a 10-item comprehension test developed for the stimulus discharge instructions. Second, a second measure of comprehension was provided by two cloze tests (Bormuth, 1967) corresponding to the high and low listenable versions of the discharge instructions. The cloze measures were constructed by removing every sixth word (if a content word) from print copies. Participants were to fill in the missing words. The same number of blanks appeared on both cloze tests. Only exact matches were counted as correct. Third, Amount of Invested Mental Effort (AIME; Salomon, 1984) is a 4-item self-report measure often used to indicate cognitive burden in processing messages (e.g., Greene, Krecmar, Rubin, Hale, & Walters, 2003). After testing for internal reliability, one item was removed, resulting in a 3-item measure with a Cronbach's alpha of .80. In addition to the three dependent measures, a self-report measure inquiring about prior knowledge of cardiac surgery was used as a covariate.

Analysis

Participants were nested in combinations of channel (listening, reading) and listenability (high, moderate). The covariate was prior knowledge of postcardiac surgery care. Separate 2 × 2 analyses of covariance were run for each of the three dependent variables (multiple choice test, cloze test, and AIME). Post hoc contrasts involving interaction means were conducted using the Student-Neuman-Keuls procedure.

Results

Cell means for all three dependent variables appear in Table 1.

Table 1. Adjusted cell means (and standard deviations) for three dependent variables

	Reading		Listening
	High listenable M (SD)	Moderate listenable M (SD)	High listenable M (SD)	Moderate listenable M (SD)
Multiple choice	8.85 (0.36)	7.94 (0.37)	6.87 (0.36)	6.69 (0.38)
Cloze	18.93 (0.97)	15.00 (0.99)	19.80 (0.96)	13.31 (1.03)
Mental effort	10.43 (0.54)	10.20 (0.55)	10.26 (0.53)	11.15 (0.57)

Cloze Test of Comprehension

The analysis of covariance for cloze test scores revealed a main effect for message listenability (F _1,52 = 27.55, p < .001; eta² = .35). The high listenability mean (M = 19.36) exceeded the mean for moderate listenability (M = 14.16). No other main, interaction, or covariate effects for this variable were statistically significant.

Multiple Choice Test of Comprehension

Channel of communication exerted a main effect on multiple-choice test scores (F _1,52 = 9.30, p < .01; eta² = .15). Reading comprehension (M = 7.90) exceeded listening comprehension (M = 6.78) for this population. No other main, interaction, or covariate effects for this variable were statistically significant.

AIME

No main, interaction, or covariate effects were statistically significant for this variable.

Discussion

Applying the Listenability Framework to Postsurgical Discharge Instructions

In previous research on expository prose comprehension (Rubin et al., 2000), listeners understood highly listenable passages better than they understood low listenability passages. They also expended less mental effort in understanding highly listenable prose. This study attempted to replicate those findings in the context of comprehension of health messages. As in previous research, reading offered advantages over listening in terms of comprehension. The college student participants in this study no doubt are skilled readers who are able to control the temporal and sequential flow of information so as to maximize the potential of the written word to support cognitive processing. At the same time, the highly listenable style did present an advantage over the moderately listenable style in the cloze measure of comprehension. Although the expected interaction between channel of communication and listenability did not prove statistically significant, it was marginally so (p < .15; eta² = .035), and the cell means were arrayed in the expected order. Inspection of cloze-test means in Table 1 indicates that participants who were required to listen to the passage of only moderate listenability were disadvantaged, especially in comparison to listeners who were exposed to the high listenability version. In other words, it's relatively difficult to listen to language that is not modified for listenability.

Reservations to the findings of this study include the fact that participants were (assumed) healthy college students rather than heterogeneous patients. Presumably, college students enjoy higher literacy skills than the general public, and especially so when they are not mired down by physical illness and attendant anxieties. By design, all of the participants selected for this study were native speakers of English. It remains for future studies to determine the degree to which implementing listenability principles in health communication offers advantages to English language learners. Moreover, the comparison used in this study was between moderate and high listenability rather than a more extreme comparison between low and high readability. No doubt very poorly constructed discharge instructions may be found, but it seemed counterproductive to fail to acknowledge advances already made in producing discharge instructions that are well informed by health literacy perspectives. Thus, the test here was a very conservative one.

Improving the quality and delivery of discharge instructions continues to be an area of great importance (Kripalani et al., 2007). Innovative strategies for improving patient use of discharge instructions—some involving intriguing technologies (e.g., Bickmore et al., 2010)—continue to evolve. So long as patients continue to receive discharge instructions through spoken interaction (Flacker et al., 2007), then no doubt patient safety could be further improved by incorporating listenability principles.

Conclusion

Conceptions of health literacy have evolved in recent years to encompass a rich portrait of health information exchange and decision-making processes as embedded in social contexts, affected by social determinants to be sure, yet promising individual agency for patients/consumers/citizens who can access tools and resources (e.g., Baker, 2006; Fox & Jones, 2009; Peerson & Saunders, 2009). One related trend of particular note recognizes the essential role of oral communication in health literacy processes (e.g., Rosenfeld et al., 2011; Roter, 2011; Rubin et al., 2011; Schonlau et al., 2011). Notwithstanding these widened horizons, health literacy practice remains highly dependent on traditional readability perspectives that encourage health communicators to focus on sentence and word length in order to meet certain quantitative benchmarks (Badarudeen & Sabharwal, 2010). Although the movement for document readability has certainly contributed much to reform health services (e.g., Davis et al., 1998; Philipson et al., 1999), the focus on readability detracts from the potential to bring health literacy practice more in line with emergent conceptions.

The listenability perspective can help redress that imbalance. It focuses attention on the particular cognitive burdens that attend everyday social interactions, media viewing, as well as health-care encounters in which the spoken word predominates. Messages that are best tailored to reduce those cognitive burdens include elements of oral-based language and considerate rhetorical structures (Rubin, 1993). The Listenability Guide included in this article (see Appendix) is an effort to encapsulate evidence-based recommendations for oral message construction. Yet, it is not amenable to simplistic benchmarking. One cannot easily “write to the formula” (Armbruster & Anderson, 1985) to achieve listenability. Rather, the listenability perspective encourages communicators to take into account the complexity of health information exchanges (Zarcadoolas 2010), including the complex distractions and preoccupations that challenge people as they attempt to extract meaning from health messages.

One important caveat to consider for health communicators seeking to employ the Listenability Guide pertains to the cultural and linguistic identity of the target audience. While listenable language may be considered a “universal precaution” that should improve the health literacy processes of all patients and consumers (DeWalt et al., 2011), it should in no way be construed as a substitute for adhering to culturally and linguistically appropriate standards of health services (U.S. Department of Health and Human Services, 2001). For example, enhancing the listenability of a message in English cannot take the place of professional interpreting for individuals with limited English proficiency. In addition, the research upon which the Listenability Guide is based pertains mainly to native speakers of mainstream American or British English. It may be that some particular recommendations in the Listenability Guide might prove dysfunctional for listeners from other cultural or linguistic backgrounds. For example, the Guide recommends using contractions (“You're halfway home”) to help construct shared context. However, for at least some English language learners, contracted forms may impede comprehension (Ito, 2001).

Health literacy is woven in a fabric of talk. Thus, listenability is a factor in a wide range of health contexts. This article explored the use of listenability guidelines in formulating postsurgical instructions. Postsurgical instructions are an interesting case, since much effort has gone into producing better written forms (e.g., Taylor & Cameron, 2000), whereas many patients recall—if they recall the discharge process at all—mostly the verbal interactions with hospital staff (Engel et al., 2009; Flacker et al., 2007). One might well question the disposition of all those print discharge sheets. In addition to the discharge process and other face-to-face provider-patient interactions, the listenability perspective might be fruitfully applied to mediated health-promotion encounters packaged in multimedia formats such as streaming video (Hayanga & Kaiser, 2008; Keelan, Pavri-Garcia, Tomlinson, & Wilson, 2007) or through telemedicine (Greenberg, 2009).

At present, the National Action Plan to Improve Health Literacy (U.S. Department of Health and Human Services, 2010a) is guiding the advancement of health literacy research and policy. The Action Plan contains scant explicit mention of the role of oral communication and message listenability, thought if one reads between the lines, it is apparent that message listenability is at least implicitly wired into the future of health literacy. For example, the Action Plan does highlight radio programming as an important medium for disseminating health information, particularly to certain cultural groups that continue to rely on radio as a preferred medium. Similarly, the Action Plan does call for developing instruments to measure listening in health literacy processes, but one can hardly develop such an instrument without attending to the listenability of the listening stimuli from which patient/consumer skill is inferred. Moreover, the extensive attention devoted to patient-centered provider communication in the Action Plan inevitably challenges us to consider the listenability of those providers' speech as they interact with their patients.

Appendix: The Listenability Style Guide

Table

Listenability domain	Recommended language or discourse practice	Example
Oral-based sentence structure	When you link clauses, use coordinating conjunctions when possible. Don't overuse subordination.	To help this medication to work better, and you should eat a low fat diet, and you should exercise, and you should not smoke.
		NOT: In addition to eating a proper diet (such as a low-cholesterol/low-fat diet), it is known that this medication will work better if life style changes include exercising, losing weight if overweight, and stopping smoking.
	When you do use subordinate clauses, put them at the ends of sentences.	People with hypertension should avoid fast foods, because fast foods contain a lot of salt.
		NOT: Fast foods, because they contain excessive salt, should be avoided by hypertension patients.
	Keep clauses to moderate length. Avoid many long phrases, which are really “reduced clauses.”	Don't eat grapefruit and don't drink grapefruit juice while you are taking this medicine. But listen to your doctor's instructions about this grapefruit rule.
		NOT: Absent your doctor's instructions to the contrary, avoid consumption of grapefruit products while under this medication.
	Make your sentences express actions by using verb forms. Avoid dense noun forms that derive from verbs.	Consume lots of liquids so you don't dehydrate.
		NOT: Consumption of plentiful liquids prevents dehydration.
Oral-based vocabulary	Use personal pronouns to evoke a face-to-face context. Imperative sentences (understood you) also work well.	The day after your surgery your knee and lower leg might swell. So keep your leg raised.
		NOT: Swelling in the knee and lower leg is common 24 hours after surgery. Raising the leg will help reduce that symptom.
	Repeat content nouns and verbs across sentences. Don't aim for diverse vocabulary for its own sake (low lexical diversity).	To see if your tumor contains malignant cells, your doctor will do a needle biopsy. The needle biopsy will draw out a small section of the tumor that your doctor can examine under a microscope.
		NOT: Your doctor will perform a needle biopsy on your tumor. This procedure allows microscopic examination of sampled cells to search for malignancies.
	Use everyday words (but don't be afraid of using specialized language if it is common to the patient/consumer community)	Your incision from the surgery may turn red for a day or two. It's okay so long as the redness does not spread.
		NOT: The surgical incision may become inflamed for a day or two. You need to become concerned only if the inflammation spreads to surrounding tissue.
Features of face-to-face conversation	Use contractions freely.	You've got to take it before breakfast. This medicine won't work if you don't take it on an empty stomach.
		NOTE: This medication will not function properly if administered with milk or milk products.
	Use questions to focus attention.	Now would you like to know some of the easiest ways to keep your blood sugar steady all day long?
		NOT: Several strategies are available to reduce fluctuations in blood sugar levels.
	Use conjunctions at the beginning of sentences to create conversational flow.	And so you've probably heard a lot about the importance of getting lots of calcium for women's bones to stay strong.
	Use tag questions.	There is so much conflicting advice about the value of back surgery, isn't there?
	Use simple and common idioms	When you spend too much time being a couch potato, it'll get you in the end.
		NOT: Sedentary life styles carry health risks.
	Periodically call your interlocutor by name	I'll tell you, Ms. Lawson, you really know a lot about how insulin works.
Considerateness toward listeners	Use internal summaries	So now we've talked about two different ways to keep your child's teeth from going bad. One way is for you to give your child more water to drink and less soda. The second way is to get your child's teeth sealed at the dentist's office. Now here's a third thing you can do to keep your child's teeth healthy.
	Give listeners “advance organizers” to help them predict what information will follow.	In this little talk I'm going to cover three simple things you can do to reduce the chance of getting a stroke. You can cut back the amount of red meat that you eat. You can walk 1000 steps every day. And you can take one baby aspirin at night. First let's talk about the meat in your diet.
	Explicitly signal transitions between topics	So that's what I wanted to tell you about causes of reflux. If you don't have any more questions, let's talk now about what you can do about it.
	Announce important topics.	Vitamin A. Vitamin A in vegetables like carrots can keep your eyes healthy.
	Convey information in little stories	I know this one lady, my neighbor Rosie, who didn't even want to walk up her driveway to get her mail her knees hurt her so much. At first she was just using ice on her knees, and it helped at first, but then icing stopped working for her. So she starting using heat. Finally she figured out that the best thing for her was to switch between ice packs and heating pads.
		NOT: Alternating heat and cold can sometimes relieve chronic knee pain.
	Be redundant. State and then paraphrase important points	Suppose you lose consciousness or you can't stay awake during times when you are usually awake. If that happens, you should go back to the emergency room. Abnormal sleepiness or blacking out are absolutely reasons to return to the hospital right away.
		NOT: Loss of consciousness and inability to remain awake are signs that it would be wise to return to the emergency room.
	Convey information with vivid analogies to everyday objects and events.	So you see, using an expired prescription is a little like eating spoiled food. It doesn't give you the benefit you need, and it can actually hurt you.
		NOT: Expired prescriptions often fail to deliver the necessary therapeutic effect and may cause additional harm.