Skip to main content
NCBI home page
Journal of the American Medical Informatics Association: JAMIA logoLink to Journal of the American Medical Informatics Association: JAMIA
. 2016 May 27;23(6):1136–1142. doi: 10.1093/jamia/ocw019

Automated pictographic illustration of discharge instructions with Glyph: impact on patient recall and satisfaction

Brent Hill 1,✉, Seneca Perri-Moore 1, Jinqiu Kuang 1, Bruce E Bray 1, Long Ngo 2, Alexa Doig 1, Qing Zeng-Treitler 1
PMCID: PMC5070521  PMID: 27234601

Abstract

Objectives First, to evaluate the effect of standard vs pictograph-enhanced discharge instructions on patients’ immediate and delayed recall of and satisfaction with their discharge instructions. Second, to evaluate the effect of automated pictograph enhancement on patient satisfaction with their discharge instructions.

Materials and Methods Glyph, an automated healthcare informatics system, was used to automatically enhance patient discharge instructions with pictographs. Glyph was developed at the University of Utah by our research team. Patients in a cardiovascular medical unit were randomized to receive pictograph-enhanced or standard discharge instructions. Measures of immediate and delayed recall and satisfaction with discharge instructions were compared between two randomized groups: pictograph (n = 71) and standard (n = 73).

Results Study participants who received pictograph-enhanced discharge instructions recalled 35% more of their instructions at discharge than those who received standard discharge instructions. The ratio of instructions at discharge was: standard = 0.04 ± 0.03 and pictograph-enhanced = 0.06 ± 0.03. The ratio of instructions at 1 week post discharge was: standard = 0.04 ± 0.02 and pictograph-enhanced 0.04 ± 0.02. Additionally, study participants who received pictograph-enhanced discharge instructions were more satisfied with the understandability of their instructions at 1 week post-discharge than those who received standard discharge instructions.

Discussion Pictograph-enhanced discharge instructions have the potential to increase patient understanding of and satisfaction with discharge instructions.

Conclusion It is feasible to automatically illustrate discharge instructions and provide them to patients in a timely manner without interfering with clinical work. Illustrations in discharge instructions were found to improve patients’ short-term recall of discharge instructions and delayed satisfaction (1-week post hospitalization) with the instructions. Therefore, it is likely that patients’ understanding of and interaction with their discharge instructions is improved by the addition of illustrations.

Keywords: informatics, consumer, patient education, health literacy, pictograph

Background and Significance

The responsibility for the majority of a patient’s care after the patient is discharged from the hospital is relegated to the patient themselves and patient’s family.1 For this reason, it is essential that patients and their caregivers understand critical elements of the patient’s discharge instructions, such as activity restrictions, dietary guidelines, medication management, wound care, follow-up instructions, signs and symptoms of potential problems, and emergency contact information, in order to successfully negotiate the recovery period.2 Discharge instructions are typically given during the hospital discharge process.3 These instructions can be in verbal and/or written format and are prepared and explained to patients/caregivers by a nurse. If successful, the instructions will give the patients and/or caregivers the knowledge and skills necessary for the patient’s post-hospitalization care.4 The patient and their caregivers must be able to understand the patient’s discharge instructions so that the patient can recall aftercare instructions or recognize that the information they require for their post-discharge care can be found in their discharge instructions.5,6 Patients who cannot understand health information, and therefore cannot follow instructions regarding their care have higher rates of hospitalization, hospital readmission, costly and unnecessary complications, use of emergency services, and death.7,8 Should patients leave the hospital with an inadequate understanding of their medical condition and plan of care, they may not recognize the importance of adhering to their aftercare instructions.4–6,9

Unfortunately, prior studies have revealed that patients, regardless of health literacy and education level, commonly have problems understanding and recalling their discharge instructions.8,10,11 Many barriers to a patient’s ability to understand discharge instructions are inherent in the hospital environment and are difficult to ameliorate. Examples include side effects of medications, poor sleep quality while hospitalized, and effects associated with the patient’s morbidity.1 In order to address patients’ inability to understand and follow discharge instructions, it is essential to re-evaluate the efficacy of discharge instructions and employ strategies that increase the comprehensibility of discharge instruction for patients.12

One intervention that has the potential to improve patients’ comprehension of discharge instructions is the use of illustrations and pictures to depict important health and self-care information.1,7,13–15 Pictographs, such as the Wong-Baker Faces of Pain Scale, have been found to help patients with low health literacy express their experience better than numbers, rankings, or potentially unfamiliar words.16,17 Other studies have demonstrated the effectiveness of including pictographs in patient instructions.14,15,18–26 However, illustrating patient instructions is a time-consuming task.

In our research, we have developed an automated system called Glyph to enhance hospital discharge instructions. A team of two informaticians, two clinicians, a programmer, and a graphic designer developed Glyph. Glyph is a Java application with a web-based interface that accepts free text input and generates illustrations for that text. It has a few components: several preprocessing and processing modules that parse and annotate the instruction text, a look-up module that maps the annotated text to images in a library we created for the project, a rule engine that combines images based on semantic grammar, and a rendering module that integrates images with text.27 In preliminary studies, Glyph has demonstrated the ability to illustrate discharge instructions.28 In this study, we deployed and tested the Glyph system in a real-life clinical setting in a randomized controlled trial.

Objectives

The aims of this study were twofold. The first was to evaluate the effect of automated pictograph enhancement on patients’ immediate and delayed recall of discharge instructions. The second was to evaluate the effect of automated pictograph enhancement on patients’ satisfaction with discharge instructions.

Materials and Methods

Glyph System

The Glyph system illustrates patient discharge instructions via the following steps: (1) preprocessing free text; (2) annotating illustratable terms from the text; (3) removing nonillustratable text from further processing; (4) composing images based on the illustratable terms and grammar patterns; and (5) rendering images for the corresponding text.27 In a prior study, we used Glyph to illustrate 49 patient instructions representing 10 different discharge templates from the University of Utah Cardiology Service.28 Healthy participants were asked to review and then recall a set of discharge instructions that contained five pictograph-enhanced and five nonpictograph-enhanced items. The presence of Glyph pictographs in the instructions improved patients’ discharge instruction recall (P < .001).28 Table 1 shows examples of Glyph-illustrated instructions.

Table 1:

Discharge Instructions and Their Associated Illustrations

Instruction Illustration
graphic file with name ocw019ilf1.jpg
graphic file with name ocw019ilf2.jpg
Open in a new tab

Standard Discharge Instruction Processes

The study site, the Cardiovascular Medical Unit (CVMU), is within the University of Utah Health Care System which is a large academic medical center that serves Utah and five surrounding states.29 The CVMU has 35 beds and treats approximately 2382 patients annually. The division of cardiology has 45 primary faculty and 20 advanced-practice clinicians in general cardiology, interventional cardiology, heart failure/transplant, electrophysiology, and genetic cardiology.30 UUHC CVMU patients receive discharge instructions that are composed of two documents. One is a document created using a standard set of diagnosis-/procedure-specific templates for atrial fibrillation ablation, heart surgery, heart failure, minimally invasive chest surgery, electrophysiology device, and post-catheterization (Appendix A). A discharge nurse sometimes modifies the templates. For example, the nurse would remove information about a defibrillator from the electrophysiology device template if the patient had a pacemaker implant. Another document, the Patient Care Summary, is created by the patient’s physician and is combined with the diagnosis-/procedure-specific template and provided to the patient by the discharging nurse immediately prior to the patient’s discharge and in conjunction with discharge teaching. The patient care summary often includes personalized instructions about diet and activity as well as activity restrictions.

Sample and Setting

Study participants were recruited from the UUHC CVMU. This site was selected because the clinical members of our research team include a UUHC cardiologist and a cardiology research nurse. Additionally, the UUHC CVMU has a dedicated team of nurses that provide discharge teaching and instructions to patients, which facilitated the implementation of the study. Participants that met all the inclusion criteria and no exclusion criteria were approached by study nurses in their hospital room for consent to participate in the study. Criteria for inclusion in the study were as follows: ability to speak, read, and write in English; being discharged to home; ≥21 years old; and having a discharge plan that included complete discharge instructions. Criteria for exclusion from the study were as follows: having cognitive or physical impairments that prevented participation in the study; being employed as a nurse, physician, or pharmacist or works with discharge instructions; and being on the heart transplant or left ventricular assist device waiting list or being a recipient of either of these (because discharge instructions for these patients are more detailed, complex, and are closely monitored by their clinical team, unlike most patients). The diagnosis or procedure received by study participants included ablation, atrial fibrillation ablation, angiogram, atrial septal defect/patent foramen ovale, coronary artery disease/stent/heart attack, electrophysiology device, heart failure, heart surgery, and minimally invasive heart surgery.

All the patients who met the inclusion criteria and no exclusion criteria were approached for study participation until 144 patients consented and completed all the study procedures. Data was not collected for those patients who declined to participate in the study. Data collection was completed between September 10, 2013 and January 22, 2014. The sample size calculation was conducted using G*Power.31 Baseline immediate recall rate estimates of 0.44 for standard discharge instructions and 0.54 for pictograph-enhanced discharge instructions were based on Dr Zeng-Treitler’s (2009)32 preliminary research; she also estimated the standard deviation to be 0.20. Using the means for each group, 0.44 and 0.54, and a standard deviation (SD) of 0.20 yielded a Cohen’s d effect size of 0.5 with a two-tailed test, an α = 0.05 and power = 0.80, a sample size of 128 would have been sufficient to detect the large effect between groups, with 64 participants randomized to the control group and 64 participants randomized to the intervention group. The study was over-enrolled to account for possible attrition.

Instruments and Measures

Paivio’s additivity hypothesis posits that text paired with images is additive and results in superior recall rates than text alone.33–37 This foundational research on verbal and imaginal processing was applied in this study for instrument development. Additionally, recall has been used to measure a patient’s memory and understanding of his or her discharge instructions; some of these studies also used illustrations or cartoons paired with discharge and other health instruction text.5,6,14,16,25,38–43 The measure of patients’ free recall of instructions for this study was collected using prompt questions developed by the study investigators (Appendix B). The questions were based on key elements of discharge instructions, eg, “What wound care instructions were listed on your discharge instructions?,” and “What activity restrictions were given in your instructions?” A question about the patient’s medications was initially included in this instrument but was excluded because the CVMU protocol for medications was in transition to become a separate discharge task conducted by a clinical pharmacist. Recall rates were calculated by dividing the number of words a study participant remembered by the total word count of the document.5,44 Given that each patient’s discharge instructions differed in length and content based on the patient’s diagnosis or procedure and the amount of personalization, this calculation allowed for a comparison of the recall of different instructions through the creation of a ratio of words the patient remembered.

Patient satisfaction is strongly correlated with satisfaction with their hospital care, and discharge preparation has previously been rated lowest among all aspects of hospital care.45–48 Most publications that describe the measurement of patient satisfaction in the clinical environment do not publish their tools because they are proprietary, such as Krames On-Demand and the Press Ganey Inpatient Survey.49,50 The study team used simple patient satisfaction questions with a Likert-type response that ranged from 1 (completely dissatisfied) to 7 (completely satisfied) (Appendix C), similar to those used in the proprietary satisfaction surveys.

Data Collection Procedures

Prospective study participants’ names were provided to our study nurses by CVMU nurses after a daily CVMU discharge planning meeting, and study nurses were apprised of patients who received discharge orders throughout the day. Patients were screened, and those who qualified based on the inclusion/exclusion criteria were approached for consent. Patients who consented to participate in the study were randomized to receive standard vs pictograph-enhanced discharge instructions. Intervention group discharge instructions were prepared by study nurses, who obtained an electronic copy of the participant’s Patient Care Summary from the electronic health record and processed it with the Glyph system to illustrate diet, activity, and activity restriction sections. Study nurses reviewed the newly illustrated material for accuracy, readability, and patient safety, and no issues related to the automated insertion of pictograph illustrations were identified. The Patient Care Summary was then appended to include a diagnosis-/procedure-specific template that was illustrated with Glyph and then modified to include personalization alterations that were made to the templates by CVMU preventive cardiology nurses. Control group discharge instructions were prepared by study nurses, who printed an electronic copy of the participant’s Patient Care Summary from the electronic health record and appended it with a copy of the diagnosis-/procedure-specific template that included personalization alterations, if any. Modifications to the instructions by the discharge nurse for both pictograph-enhanced and standard discharge instructions were handled the same way, by either inclusion or exclusion of instructions. Study nurses were not blinded for the intervention, because they prepared the study documents and assessed recall in study participants.

Study procedures for data collection began after the patient received all standard hospital discharge teaching by their nurse, using standard discharge instructions. After completing the traditional final discharge teaching, study nurses were notified that the patient was ready to begin the study. The study nurse then went to the patient’s room and presented the participant with the version of study discharge instructions that they were randomized to receive and advised the participants to review the document for up to 15 min. When the participant indicated that they were finished reviewing their discharge instructions, they were asked a series of free recall questions (Appendix B) to assess their immediate recall of the instructions. Study nurses used a nonillustrated copy of each patient’s instructions and highlighted each word, or a close approximation, which was verbalized by each patient. Additionally, participants were asked two questions about their satisfaction with their discharge instructions (Appendix C). In order to assess delayed recall, participants were called 1 week after discharge and asked the same free recall and satisfaction questions by a member of the study team who was blinded to participant group assignment. Study data were collected and managed using Research Electronic Data Capture electronic data capture tools hosted at the University of University of Utah (CTSA 5UL1RR025764-02).51,52

Data Analysis

First, we examined the distributions of participant demographics and baseline clinical information to assess the performance of the randomization procedure. We expected that these distributions would be similar between the two groups (ie, those participants who received standard discharge instructions and those who received pictograph-enhanced discharge instructions). There were 11 variables used for this analysis. For those variables, which were categorical, we used the chi-square test or Fisher’s exact test, when the expected count for at least 25% of the cells in the contingency table fell below 5. For continuous variables, we used the two sample t-test when the distributions were nearly normally distributed; otherwise, we used the non-parametric Wilcoxon test. Variables that were statistically significant in this analysis were then used as confounding variables in the multivariate models. We used the type-I error of 0.05 for declaring a statistically significant difference between the two groups.

Second, we obtained the estimate of the primary endpoint of this study – the estimated adjusted mean difference of normalized pre-recall illustrated ratio and the normalized post-recall illustrated ratio between the two groups. Recall ratios were normalized by dividing the number of words a patient remembered by the total number of words in the patient’s discharge instructions. This was necessary because the lengths of individual discharge instructions varied based on the procedure or diagnosis the patient had received and on any personalization that was included in the instructions by the patient’s doctor or nurse. We assessed the distributions of the normalized recall ratios and found them to be reasonably symmetric and nearly normally distributed. We structured the data so that each subject had the pre-recall and post-recall set up for longitudinal data analysis using the linear mixed-effects models, which could capture the within-subject correlation of the two recall ratios. We tested different variance-covariance structures and found compound-symmetry (CS) to yield the lowest Akaike Information Criterion (AIC). Therefore, we used CS structure in the linear mixed effects model. In this model, we treated education level (below or above a 12th grade level) as a confounder and included time (pre- and post-recall), group, as well as the interaction between time and group. We then used linear contrasts to estimate the adjusted mean difference between the two groups (pictograph and standard) and also reported the linear contrast P-values.

Third, we examined the effect of the intervention on the patient satisfaction items (ie, understanding the discharge instructions, having enough information to manage health after leaving the hospital). We used a generalized linear model with log link and binary error to obtain the estimated relative risks, 95% confidence interval (CI), and P-values.

We used the SAS software version 9.3 (SAS Institute, Inc.) for all of the analyses.

Results

Of the 11 variables we examined for the randomization procedure (Table 2), we found that education (initially ≤4th grade, 5–8th grade, 9–12th grade, and >12th grade, but, because there were only five participants who reported having below 9th grade education, we used binary categories – ≤12th grade and >12th grade) to be the only variable that is a potential confounder (P = .012). The pictograph group included a substantially larger proportion of subjects who had above a 12th grade education (82%) compared with the standard group (63%). Our total sample included more males than females (70% vs 30%), and was largely white (Caucasian) (92% in the standard group and 99% in the pictograph group). The average age of the participants was about 60 years old, and the average number of medications the participants were taking was between 11 and 12. Both groups had a median length of hospital stay of 2 days and a median Charlson Comorbidity Index score of 2.0. The Charlson Comorbidity Index is a method of categorizing comorbidities based on International Classification of Disease (ICD) diagnosis codes, individually weighted based on the risk of mortality or resource use and then summed.

Table 2:

Distribution of Demographics and Baseline Clinical Information

Characteristic Standard Pictograph-Enhanced P-Value*
Gender
Male 71% 69% 0.771
Female 29% 31%
Race
White 92% 99% 0.171
African American 5% 1%
Others 3% 0%
Education
Below 12th grade 37% 18% 0.012
Above 12th grade 63% 82%
Language
English 93% 97% 0.442
Confidence managing health 88% 97% 0.056
Age (years) (mean ± SD, median) 59 ± 13 (62) 61 ± 13 (63) 0.199
Length of hospital stay (days) (mean ± SD, median) 3.7 ± 4.0 (2) 4.6 ± 5.3 (2) 0.577
Charlson Comorbidity Index (mean ± SD, median) 2.9 ± 3.1 (2) 2.9 ± 2.7 (2) 0.763
Number of hospitalizations in previous year (mean ± SD, median) 1 ± 0 (1) 2 ± 0 (2) 0.553
Number of medications from patient care summary (mean ± SD, median) 11 ± 5 (10) 12 ± 5 (12) 0.142
Open in a new tab

SD, standard deviation. *P-value obtained from Chi-square, Fisher's exact test, two-sample t-test, or Wilcoxon Rank Sum test.

For the primary outcome of interest, the normalized recall ratio (Table 3), we examined both the pre- and post-normalized recall ratio and found that there was a statistically significant difference for the normalized pre-recall illustrated ratio (P = .001) with an adjusted estimated mean difference of 0.014 (95% CI, 0.006-0.023) in favor of the pictograph-enhanced instructions. For the normalized post-recall ratio, we found no difference between the two groups (P = .852; estimated mean difference of 0.004; 95% CI, -0.008-0.009).

Table 3:

Distribution of Pre-Recall and Post-Recall Illustrated Ratio for Each Treatment Group

Ratio Standard (S, 95% CI) Pictograph-Enhanced (PE, 95% CI) Difference Between Treatments [PE S]* (95% CI)* P-Value*
Pre-Recall Ratio 0.04 ± 0.03 (0.04, 0.01–0.12) 0.06 ± 0.03 (0.05, 0.01–0.14) 0.014 (0.006–0.023) 0.001
Post-Recall Ratio 0.04 ± 0.02 (0.03, 0.0–0.12) 0.04 ± 0.02 (0.03, 0.0–0.12) 0.004 (-0.008, 0.009) 0.852
Open in a new tab

*The estimated mean differences and the P-values were obtained from the linear mixed effects model, which includes the interaction between time and treatment group controlling for education, with variance-covariance type compound symmetry.

For the satisfaction items regarding understanding the discharge instructions (Table 4), we found that all four items (two for pre-recall and two for post-recall) yielded a consistent trend in favor of the pictograph-enhanced instructions, both in the unadjusted proportions, and the adjusted relative risk estimates (all point estimates were larger than 1.0) from the generalized linear models adjusting for education. The biggest difference came from the post-recall understanding discharge instructions model, with a relative risk of 1.13 and a 95% CI of 0.99-1.29 (P = .070).

Table 4:

Distribution of Pre-Recall and Post-Recall Satisfaction Assessment for Each Treatment Group

Pre/Post Assessments Standard (S) (%) Pictograph-Enhanced (PE) (%) Risk Ratio Between Treatments [PE/S]* (95% CI)* P-Value*
Pre-recall understanding discharge 89 90 1.02 (0.91, 1.14) 0.757
Pre-recall have enough information 95 97 1.04 (-0.95, 1.28) 0.711
Post-recall understanding discharge 85 99 1.13 (0.99, 1.29) 0.07
Post-recall have enough information 92 97 1.06 (-0.98, 1.15) 0.142
Open in a new tab

*The risk ratios and P-values were obtained from the generalized linear models with log link and binary error controlling for education.

Discussion

The results of this study demonstrate that an informatics intervention to improve patient recall of and satisfaction with discharge instructions through automatic illustration of discharge instructions shows promise. Specifically, pictographic enhancement of the discharge instructions improved patient’s immediate recall of the instructions by 35%, which is clinically significant. However, the increase in immediate recall of pictograph-enhanced vs standard discharge instructions did not carry over to patients’ delayed recall 1-week post-discharge. According to Houts et al., (2006), most patients read written instructions once and then rely on memory to take health actions based on those instructions. Therefore, theoretically, the patients who were given pictograph-enhanced discharge instructions had more information to draw on when taking health actions. Although it is not clear whether patients that received pictograph-enhanced discharge instructions had better outcomes, the results hold sufficient promise to continue with research to explore the impact of illustrated discharge instructions on patient outcomes.

Patients were asked to recall the content of their discharge instructions to the best of their ability; however, normalized recall rates appeared to be extremely low. The highest normalized recall rate was ∼15%, and the lowest recall rate was about ∼1%. This value should not be interpreted as the percentage of information that patients recall. Our recall questions prompted patients to recall key sections of their discharge instructions, but did not require the patients to recall everything from their instructions. This is due, in part, to the nature of the content covered by discharge instructions; not all the content is critical to patients’ immediate post-discharge care (eg, statements such as “You are the most important person in your in care.”). As such, our questions focused mostly on the content that was most relevant for patients’ immediate post-discharge care. In an effort to estimate the percentage of content covered by the recall questions, a cardiologist on the study team reviewed a random sample of 10 discharge instructions used in this study and highlighted the content he considered to be immediate post-hospitalization instructions vs general health education. This analysis revealed that approximately 10–18% of the Patient Care Summary template content and 20–51% of the preventive cardiology template content are immediate post-hospitalization instructions. This finding leads to several implications for both the hospital discharge process as well as the metric used for this study. Patients receive an overwhelming amount of information at discharge and might benefit from simplified discharge instructions.1,11,14,53,54 It may be beneficial to limit inpatient discharge instructions to aftercare needs, such as symptom identification and wound care, and address health education information, such as dietary recommendations, in a separate document and educational setting.

Another implication is that, should a patient lose his or her discharge instructions or otherwise not be able to identify them, they may not have retained sufficient knowledge to adequately adhere to the post-hospitalization instructions. To mitigate this situation, the care team could provide an electronic copy of the discharge instructions to the patient and/or their caregivers via e-mail or the patient’s personal health record. Finally, free recall may not be the best metric to assess patients’ comprehension of discharge instructions. Reading for comprehension is a different process than reading for memorization. Therefore, exploring different methods of measuring a patient’s comprehension of the activities and concepts in discharge instructions is incumbent for future research in this area. Currently, structured interviews, questionnaires, measures of patient health literacy, and patient self-reporting are used to assess patient comprehension of discharge instructions.1,2,7,11,12,53,54,55 However, because we sought to measure the impact of illustrations paired with text on patient recall of discharge instructions, we sought to reveal patient comprehension at the instruction level rather than with concept identification.

Patient satisfaction with the ease of use of their discharge instructions was similar at discharge for both groups, but, after 1 week, during which the study participants had the opportunity to review their instructions at home, those who received the pictograph-enhanced instructions reported greater levels of satisfaction with the ease of use of their instructions than those that received standard instructions. Illustrated discharge instructions were considerably longer than the standard instructions, due to the amount of space pictographs occupy on the page. It is intriguing that patients found the instruction sets that included illustrations easier to use than standard instructions, even though the illustrated instructions were approximately three times longer. Previous research on the use of pictographs in health communication provides some insight into how pictographs increase patient satisfaction with ease of use. Pictures paired with text helps readers visualize concepts in the text, which can increase comprehension of the material; this may translate to increased satisfaction with ease of use because text instructions alone require more cognitive resources to make a mental model of the information presented in the text.14,25 For example, a study on asthma inhaler instructions found that patients who received picture-enhanced instructions performed more instruction steps faster and more accurately while expressing fewer doubts than those who received text-only instructions.25

Limitations of this study were that the sample was mostly white (Caucasian), and the vast majority had completed a high school education and beyond, which matches the demographics of Utah, where the study was conducted.56 This lack of diversity limits the generalizability of the results to other populations. Additional research with diverse patient populations would reveal more about the utility of pictograph-enhanced discharge instructions for patients with less education and lower health literacy as well as individuals from racial and ethnic minority groups, who may interpret pictographs in dramatically different ways. Another limitation is that, because comprehension is the translation of information into meaningful ideas, recall is not equivalent to comprehension. Similarly, comprehension does not always translate to adherence or satisfaction. On the other hand, although perfect recall of instructions is not necessary for adherence to those instructions, a very low level of recall is likely to hinder adherence.

Even with the inclusion of illustrations, most patients’ immediate recall rate of their hospital discharge instructions was relatively low. As mentioned previously, in hospital discharge instructions, information that is of immediate clinical significance is mixed with advice about long-term behavior changes and general patient education. To make discharge instructions more comprehensible, memorable, and usable, we must improve the content, presentation, formatting, and delivery of discharge instructions.

Conclusion

The inclusion of pictographs is one method to improve patients’ recall of and satisfaction with hospital discharge instructions, but is not a solution unto itself. Discharge instructions for patients hospitalized with cardiovascular diseases are already complex and extensive, and, when enhanced by pictographs, become considerably longer. Other methods of providing information to patients should be explored to capitalize on several avenues of communication, because it is unlikely that one intervention, however efficacious, is sufficient to address the multifactorial issues that impact a patient’s ability to understand and follow his or her hospital discharge instructions. Some possible avenues of providing discharge instructions include via a tethered, electronic personal health record, via e-mail, or even via an infographic that can be placed in a strategic location in the patient’s home.

This study also demonstrated the feasibility of the Glyph system for use in a busy clinical environment to automatically enhance patient education materials. This process was completed in a very busy cardiovascular medical unit and did not interrupt the flow of patient treatment and care. The addition of pictographs to discharge instructions has the potential to increase the efficiency of patient education, because the use of pictographs in health communication focuses patients’ attention to the materials, aids in their comprehension and performance of instructions, is not time consuming, and does not interfere with the work of the clinical environment.14,25

Acknowledgments

We would like to express our gratitude for the efforts of Duy Bui, Rebecca Morris, Katherine Doyon and Carrie Christensen for the many hours of work they put into this project.

SUPPLEMENTARY MATERIAL

Supplementary material is available online at http://jamia.oxfordjournals.org/.

Contributors

B.H. participated in the study implementation design, led the data collection, contributed to the data analysis and interpretation, and drafted and revised the article. S.P. participated in the study implementation design, the data collection, and drafting and design of the article. J.K. participated in the study implementation design, the data collection, and the data analysis. B.B. contributed to the study implementation design. L.G. led the statistical analysis. A.D. participated in the data interpretation and the drafting and revision of the article. Q.Z. was the principal investigator of the study, designed the study, led the study implementation design, oversaw the data collection, and contributed to the data analysis and drafting and revision of the article.

Funding

This work was supported by the National Institutes of Health (R01 LM07222).

Competing interests

None.

REFERENCES

  • 1.Chugh A, Williams MV, Grigsby J, Coleman EA. Better transitions: improving comprehension of discharge instructions. Frontiers Health Services Manag. 2009;25(3):11–32. [PubMed] [Google Scholar]
  • 2.Jacott WE. JACHO Advocates Improved Communication During Discharge. The Hospitalist. 2007; Accessed January 14, 2013. http://www.the-hospitalist.org/details/article/234091/New_Moves.html.
  • 3.Pfuntner AW, Wier LM, Elixhauser A. Overview of hospital stays in the United States, 2010. Agency for Healthcare Research and Quality. Location: Rockville, MD; 2012. [PubMed] [Google Scholar]
  • 4.Clark KD, Duco J, Lattimer C. Transitions of care: the need for a more effective approach to continuing patient care. The Joint Commission. Location: Oak Brook, IL; 2012. [Google Scholar]
  • 5.Linn AJ, van Dijk L, Smit EG, Jansen J, van Weert JCM. May you never forget what is worth remembering: the relation between recall of medical information and medication adherence in patients with inflammatory bowel disease. J Crohn's Colitis. 2013;7(11):e543–e550. [DOI] [PubMed] [Google Scholar]
  • 6.Kessels RPC. Patients' memory for medical information. J Royal Soc Med. 2003;96(5):219–222. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Boodman SG. Helping patients understand their medical treatment. 2011. http://www.kaiserhealthnews.org/stories/2011/march/01/health-literacy-understanding-medical-treatment.aspx. Accessed June 2, 2013. [Google Scholar]
  • 8.Institute of Medicine. Health Literacy: a Prescription to End Confusion. Institute of Medicine, Washington D.C.: National Academies Press; 2004. [PubMed]
  • 9.Pavlik V, Brown AE, Nash S, Gossey JT. Association of patient recall, satisfaction, and adherence to content of an electronic health record (EHR)-generated after visit summary: a randomized clinical trial. J Am Board Fam Med. 2014;27(2):209–218. [DOI] [PubMed] [Google Scholar]
  • 10.Sawyer TG, McBroom K, Granger B, Bride W, Harper M. 28. Making the difference: a shared position to address patients understanding of discharge instructions and post discharge adherence. Heart Lung. 2011;40(4):386. [Google Scholar]
  • 11.Engel KG, Buckley BA, Forth VE, et al. Patient understanding of emergency department discharge instructions: where are knowledge deficits greatest? Acad Emerg Med. 2012;19(9):E1035–E1044 [DOI] [PubMed] [Google Scholar]
  • 12.Dartmouth Atlas Project. The Revolving Door: A report on U.S. hospital readmissions. Healthcare Purchasing News. 2013;37(4):10. [Google Scholar]
  • 13.Choi J. Development and pilot test of pictograph-enhanced breast health-care instructions for community-residing immigrant women. Int J Nursing Practice. 2012;18(4):373–378. [DOI] [PubMed] [Google Scholar]
  • 14.Houts PS, Doak CC, Doak LG, Loscalzo MJ. The role of pictures in improving health communication: a review of research on attention, comprehension, recall, and adherence. Patient Educ Counsel. 2006;61(2):173–190. [DOI] [PubMed] [Google Scholar]
  • 15.Jeungok C. Pictograph-based discharge instructions for low-literate older adults after hip replacement surgery. J Gerontol Nursing. 2011;37(11):47–56. [DOI] [PubMed] [Google Scholar]
  • 16.Mayer GG, Villaire M. Health Literacy in Primary Care: A Clinician's Guide. New York, NY, USA: Springer Publishing Company; 2007. [Google Scholar]
  • 17.Wong-Baker Faces Foundation. 1983. http://www.wongbakerfaces.org. Accessed April 26, 2013.
  • 18.Austin PE, Matlack R, Dunn KA, Kesler C, Brown CK. Discharge instructions: do illustrations help our patients understand them? Ann Emer Med. 1995;25(3):317–320. [DOI] [PubMed] [Google Scholar]
  • 19.Cherry KE, Dokey DK, Reese CM, Brigman S. Pictorial illustrations enhance memory for sentences in younger and older adults. Exp Aging Res. 2003;29(3):353–370. [DOI] [PubMed] [Google Scholar]
  • 20.Choi J. Literature review: using pictographs in discharge instructions for older adults with low-literacy skills. J Clin Nursing. 2011;20(21-22):2984–2996. [DOI] [PubMed] [Google Scholar]
  • 21.Choi J. Improving discharge education using pictographs. Rehabilitation Nursing. 2013;38(5):240–246. [DOI] [PubMed] [Google Scholar]
  • 22.Dowse R, Ehlers M. Pictograms for conveying medicine instructions: comprehension in various South African language groups. South African J Sci. 2004;(100):687–693. [Google Scholar]
  • 23.Hickman MS, White WL, White WA. Illustrations as a patient education tool to improve recall of postoperative cataract medication regimens in the developing world. Hawaii Med J. 2010;69(9):212–215. [PMC free article] [PubMed] [Google Scholar]
  • 24.Katz MG, Kripalani S, Weiss BD. Use of pictorial aids in medication instructions: a review of the literature. Am J Health-System Pharm. 2006;63(23):2391–2397. [DOI] [PubMed] [Google Scholar]
  • 25.Kools M, van de Wiel MWJ, Ruiter RAC, Kok G. Pictures and text in instructions for medical devices: Effects on recall and actual performance. Patient Educ Counsel. 2006;64(1–3):104–111. [DOI] [PubMed] [Google Scholar]
  • 26.Zikmund-Fisher BJ, Witteman HO, Dickson M, et al. Blocks, ovals, or people? Icon type affects risk perceptions and recall of pictographs. Med Decis Mak. 2013;34:443–453. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Bui D, Nakamura C, Bray BE, Zeng-Treitler Q. Automated illustration of patients instructions. AMIA Annual Symposium Proceedings/AMIA Symposium AMIA Symposium. 2012;2012:1158-1167. [PMC free article] [PubMed] [Google Scholar]
  • 28.Zeng-Treitler Q, Perri S, Nakamura C, et al. Evaluation of a pictograph enhancement system for patient instruction: a recall study. J Am Med Inform Assoc. 2014;21(6):1026–1031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.University of Utah Healthcare, ND. http://healthcare.utah.edu/about. Accessed September 3, 2015.
  • 30.University of Utah School of Medicine Division of Cardiovascular Medicine, ND. http://medicine.utah.edu/internalmedicine/cardiology/index.php. Accessed September 3, 2015.
  • 31.Faul F, Erdfelder E, Lang A-G, Buchner A. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods. 2007;41:1149–1160. [DOI] [PubMed] [Google Scholar]
  • 32.Kim H, Nakamura C, Zeng-Treitler Q. Assessment of Pictographs Developed Through a Participatory Design Process Using an Online Survey Tool. Eysenbach G, ed. Journal of Medical Internet Research. 2009;11(1): e5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Paivio A, Rogers TB, Smythe PC. Why are pictures easier to recall than words? Psychonomic Sci. 1968;11(4):137–138. [Google Scholar]
  • 34.Paivio A. Mental Representations: A Dual Coding Approach. New York: Oxford University Press; 1986. [Google Scholar]
  • 35.Paivio A, Csapo K. Picture superiority in free recall: imagery or dual coding? Cognitive Psychol. 1973;5(2):176–206. [Google Scholar]
  • 36.Paivio A. Dual coding theory: retrospect and current status. Canadian J Psychology/Revue canadienne de psychologie. 1991;45(3):255–287. [Google Scholar]
  • 37.Thomas N. 2014. http://plato.stanford.edu/entries/mental-imagery/theories-memory.html. Accessed April 27, 2013. [Google Scholar]
  • 38.Sanderson BK, Thompson J, Brown TM, Tucker MJ, Bittner V. Assessing patient recall of discharge instructions for acute myocardial infarction. J Healthcare Qual. 2009;31(6):25–34. [DOI] [PubMed] [Google Scholar]
  • 39.Delp C, Jones J. Communicating information to patients: the use of cartoon illustrations to improve comprehension of instructions. Acad Emerg Med. 1996;3(3):264–270. [DOI] [PubMed] [Google Scholar]
  • 40.Houts PS, Bachrach R, Witmer JT, et al. Using pictographs to enhance recall of spoken medical instructions. Patient Educ Counsel. 1998;35(2):83–88. [DOI] [PubMed] [Google Scholar]
  • 41.Houts PS, Witmer JT, Egeth HE, Loscalzo MJ, Zabora JR. Using pictographs to enhance recall of spoken medical instructions II. Patient Educ Counsel. 2001;43(3):231–242. [DOI] [PubMed] [Google Scholar]
  • 42.Jolly BT, Scott JL, Sanford SM. Simplification of emergency department discharge instructions improves patient comprehension. Ann Emerg Med. 1995;26(4):443–446. [DOI] [PubMed] [Google Scholar]
  • 43.Zeng-Treitler Q, Kim H, Hunter M. Improving patient comprehension and recall of discharge instructions by supplementing free texts with pictographs. AMIA Annual Symposium Proceedings/AMIA Symposium AMIA Symposium. 2008:849–853. [PMC free article] [PubMed] [Google Scholar]
  • 44.Jansen J, van Weert J, van der Meulen N, van Dulmen S, Heeren T, Bensing J. Recall in older cancer patients: measuring memory for medical information. Gerontologist. 2008;48(2):149–157. [DOI] [PubMed] [Google Scholar]
  • 45.Clarke C, Friedman SM, Shi K, Arenovich T, Monzon J, Culligan C. Emergency department discharge instructions comprehension and compliance study. Can J Emerg Med. 2005;7(1):5–11. [DOI] [PubMed] [Google Scholar]
  • 46.Perez-Carceles MD, Gironda JL, Osuna E, Falcon M, Luna A. Is the right to information fulfilled in an emergency department? Patients' perceptions of the care provided. J Eval Clin Pract. 2010;16(3):456–463. [DOI] [PubMed] [Google Scholar]
  • 47.Topacoglu H, Karcioglu O, Ozucelik N, et al. Analysis of factors affecting satisfaction in the emergency department: a survey of 1019 patients. Adv Ther. 2004;21(6):380–388. [DOI] [PubMed] [Google Scholar]
  • 48.Huang J-A, Lai C-S, Tsai W-C, Weng R-H, Hu W-H, Yang D-Y. Determining factors of patient satisfaction for frequent users of emergency services in a medical center. J Chinese Med Assoc. 2004;67(8):403–410. [PubMed] [Google Scholar]
  • 49.Lo S, Stuenkel DL, Rodriguez L. The impact of diagnosis-specific discharge instructions on patient satisfaction. J PeriAnesthesia Nurs. 2009;24(3):156–162. [DOI] [PubMed] [Google Scholar]
  • 50.Clark PA, Drain M, Gesell SB, et al. Patient perceptions of quality in discharge instruction. Patient Educ Counsel. 2005;59(1):56–68. [DOI] [PubMed] [Google Scholar]
  • 51.Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap) – a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377–381. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.McCray AT. Promoting health literacy. J Am Med Inform Assoc. 2005;12(2):152–163. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Engel KG, Heisler M, Smith DM, Robinson CH, Forman JH, Ubel PA. Patient comprehension of emergency department care and instructions: are patients aware of when they do not understand? Ann Emerg Med. 2009;53(4):454–461.e15. [DOI] [PubMed] [Google Scholar]
  • 54.Kadakia RJ, Tsahakis JM, Issar NM, et al. Health literacy in an orthopedic trauma patient population: a cross-sectional survey of patient comprehension. J Orthopaedic Trauma. 2013;27(8):467–471. [DOI] [PubMed] [Google Scholar]
  • 55.Ginde AA, Weiner SG, Pallin DJ, Camargo CA., Jr Multicenter study of limited health literacy in emergency department patients. Acad Emerg Med 2008;15(6):577–580. [DOI] [PubMed] [Google Scholar]
  • 56.Bureau USC. State and County Quick Facts. 2015. http://quickfacts.census.gov/qfd/states/49000.html. Accessed September 8, 2015.

Articles from Journal of the American Medical Informatics Association : JAMIA are provided here courtesy of Oxford University Press

RESOURCES