The Association of Health Literacy Domains With Hospitalizations and Mortality

May 13, 2020

Despite previous research evidence, this study did not reveal an overall association of health literacy, numeracy, and graph literacy with all-cause hospitalizations or mortality.

ABSTRACT

Objectives: To determine whether health literacy, numeracy, and graph literacy are associated with all-cause hospitalizations or mortality in community-dwelling veterans.

Study Design: Retrospective cohort study.

Methods: A total of 470 community-dwelling veterans underwent evaluations of health literacy, numeracy, and graph literacy with validated instruments in 2012 and were followed until 2018. At the end of follow-up, the associations with all-cause hospitalizations and mortality were determined with the Andersen-Gill model and Cox regression multivariate analysis, respectively.

Results: There were no associations of health literacy, numeracy, or graph literacy with all-cause hospitalization or mortality after multivariate adjustment. In subgroup analysis, subjective numeracy was associated with hospitalizations in African Americans. Higher objective and subjective numeracy were associated with future hospitalizations only for those with a history of hospitalization. Higher graph literacy was associated with lower mortality in those with a history of hospitalization.

Conclusions: This study did not show associations of health literacy, numeracy, or graph literacy scores with lower risk of all-cause hospitalization or mortality. Further research is needed with random sampling in a broader spectrum of healthcare settings to better understand what roles health literacy, numeracy, and graph literacy might play in healthcare utilization and clinical outcomes.

Am J Manag Care. 2020;26(5):200-206. https://doi.org/10.37765/ajmc.2020.43152

Takeaway Points

Despite previous research evidence, this study did not reveal an overall association of health literacy, numeracy, and graph literacy with all-cause hospitalizations or mortality.

  • Higher subjective numeracy in African Americans and both objective and subjective numeracy in individuals with a history of hospitalization were associated with lower all-cause hospitalizations.
  • Higher graph literacy was associated with lower all-cause mortality in those individuals with a history of hospitalization.

Health literacy is the capacity to obtain, process, and use basic health information and services needed to make healthcare decisions.1 Health literacy is a multidimensional process, including system demands and complexities, as well as individuals’ skills and abilities,2 and may encompass numeracy and graph literacy. Numeracy is a set of quantitative abilities needed for comprehending, managing, and manipulating numerical expressions of probability about health information,3 and graph literacy is the ability to understand basic graphical representations used to present quantitative health-related information.4 Health literacy has been linked to self-management behaviors, healthcare costs, hospitalizations, mortality, and overall health status.5 Less is known about the specific roles of numeracy and graph literacy in health.

Patients receiving care at US Veterans Health Administration (VA) facilities tend to be older, have more comorbidities, and have lower levels of function, education, employment, and socioeconomic status than patients in other facilities.6-9 Prior work has also suggested concerningly low levels of health literacy, numeracy, and graph literacy in this population, particularly among African American veterans.10 Many risk factors for poor health outcomes among VA patients might be compounded by these health literacy concerns. On the other hand, veterans have access to an integrated healthcare system that provides a range of supports for patient-provider communication, self-management, and care coordination,11,12 which might mitigate the risks implied by lower levels of health literacy, numeracy, and graph literacy. Future studies could explore the specific aspects of numeracy and graph literacy that predict mortality in high-risk, high-need individuals. Optimal numeracy and graph literacy skills may influence patient decision-making and health behaviors.13 Patients with adequate numeracy and graph literacy skills may be empowered to safely and effectively use healthcare information contained in web-based patient portals.14,15 Patient engagement in their own healthcare may in turn lead to better self-management and potentially lower risk for hospitalizations and mortality.

The aim was to determine in a sample of veterans the association of health literacy, numeracy, and graph literacy with future all-cause hospitalizations and mortality. We predicted that, after adjustment for known covariates, higher levels of health literacy, numeracy, and graph literacy would be negatively associated with all-cause hospitalizations and mortality.

METHODS

Design and Participants

The present research is based on a cross-sectional study conducted from January to February 2012 on male veterans 20 years and older receiving outpatient care at a VA medical center.10 Participants were conveniently recruited at outpatient clinics and met the following inclusion criteria: enrollment in a VA clinic, sufficient cognitive function (Mini-Cog score >3), not depressed (Patient Health Questionnaire score <3), and a minimum education level of eighth grade.10 The participants underwent evaluations of health literacy, numeracy, and graph literacy. After obtaining exempted review status from our institutional review board in 2018, we proceeded to conduct a retrospective cohort study of the veterans who had participated in the 2012 study. Data on all-cause hospitalizations and mortality from 2011 until September 30, 2018, were retrieved from the VA electronic health record.

Measures

Health literacy, numeracy, and graph literacy. The following measures were obtained in the 2012 study and were described in detail in our previous publication.10 Health literacy was assessed with the Newest Vital Sign (NVS),16 which consists of a nutritional label and 6 associated questions. Scores of 0 and 1 suggest high likelihood (≥50%) of limited literacy, 2 and 3 suggest limited literacy, and 4 to 6 indicate adequate literacy. The instrument has shown indications of reliability and internal consistency. Objective numeracy was measured with 13 items assessing the ability to compare risk magnitude, convert percentages to proportions, convert proportions to percentages, convert probabilities to proportions, and compute probabilities.17-19 The Subjective Numeracy Scale is an 8-item self-report measure of perceived ability to perform various mathematical tasks and preference for the use of numerical or prose information.20 Graph literacy was measured with a scale consisting of 13 items that measure abilities to find specific information in a graph, find relationships in the data shown on the graph, and make inferences and predictions from the data.4

Outcome variables. The 2 primary study outcomes were all-cause hospitalizations and mortality.

Information on patients’ VA all-cause hospitalizations from 1 year prior to the initial health literacy assessments in 2012 and prospectively until December 2018 was obtained from the VA Corporate Data Warehouse (CDW). The primary reason for hospitalization was assessed using International Classification of Diseases, Ninth Revision and Tenth Revision codes. These had been assigned and recorded by trained staff after discharge. The primary reason for hospitalization was grouped by condition.

All-cause mortality was identified through official sources including VHA facilities, death certificates, and National Cemetery Administration data available from the VA CDW. There is high agreement (91%-99%) between dates of death recorded in the CDW and dates of death recorded in external sources that feed the VHA Vital Status File.21 The last day of follow-up was September 30, 2018.

Data Analysis

Baseline characteristics are presented as frequency (percentage) for categorical variables, as mean (SD) for normally distributed continuous variables, and as median (interquartile range [IQR]) for continuous variables with skewed distributions. We report descriptive statistics of sociodemographic data, number of medications, body mass index, and Charlson Comorbidity Index (CCI) score. We used the 5-digit zip code tabulation area and the median household income in the past 12 months (in 2011 inflation-adjusted dollars) by racial group from the US Census Bureau (2007-2011) to determine differences in median household income. For purposes of the analyses, the health literacy, objective and subjective numeracy, and graph literacy measures were handled as continuous variables. The association of health literacy, numeracy, and graph literacy with all-cause hospitalizations was determined with the Andersen-Gill model, accounting for repeated hospitalizations. Univariate and multivariate analyses were conducted adjusting for age, race, ethnicity, highest level of education, median household income, age-adjusted CCI score, and all-cause hospitalizations in the previous year. Patients were censored if they died without having a hospital admission. We built 4 models to assess the role of the covariates in the association between health literacy, numeracy, and graph literacy and all-cause hospitalization: Model 1 was adjusted for age, race, and ethnicity. Model 2 was adjusted for the covariates in model 1 and for median household income and educational level. Model 3 was adjusted for the covariates in models 1 and 2 and for CCI score. Model 4 was adjusted for the covariates in the previous models and for hospitalizations in the previous year.

The proportional hazard assumption was tested using scaled Schoenfeld residuals and found to be valid. Cox regression analysis was performed to calculate the hazard ratios (HRs) and 95% CIs of survival for scores of health literacy, numeracy, and graph literacy. Known factors associated with mortality (age, race, ethnicity, educational level, median household income, multimorbidity) were considered for inclusion in the analysis. We also built 4 models to assess the role of the covariates in the association between health literacy, numeracy, and graph literacy and mortality. Model 1 was adjusted for age, race, and ethnicity. Model 2 was adjusted for the confounders in model 1 and for educational level and median household income. Model 3 was adjusted for the covariates in models 1 and 2 and for CCI score. Model 4 was adjusted for the covariates in the previous models and for hospitalizations in the previous year. These covariates were selected on a theoretical basis a priori due to previously documented associations between each of these factors and hospitalizations or mortality in veterans.22,23 Veterans’ race was classified as Caucasian, black or African American, Asian, American Indian, or other, whereas ethnicity was considered either (1) Hispanic or Latino or (2) non-Hispanic and non-Latino. Race and ethnicity were self-reported by the veterans, and all the information was retrieved from the VA electronic health record. To assess the robustness of our results, sensitivity analyses were performed in which we dichotomized groups by age, race, and history of hospitalizations. A Pearson correlation was run to assess the relationship among health literacy, objective numeracy, subjective numeracy, and graph literacy. Associations were considered significant if P <.05. Follow-up duration was calculated as follows: (September 30, 2018 — literacy assessment administration date)/365. All analyses were performed using SPSS 24.0 for Macintosh (SPSS Inc; Chicago, Illinois) and SAS for Windows version 3.71 (SAS Institute Inc; Cary, North Carolina). All statistical tests were 2-tailed, and statistical significance was assumed for P <.05.

RESULTS

Participant Characteristics

Table 1 shows the characteristics of the 470 participants who had enough data for follow-up (of the original 502). Patients were all men, 40% were white and 83% were non-Hispanic, and the mean (SD) age was 56.8 (9.6) years. According to the NVS categories, 135 (28.7%) had a high likelihood of limited literacy (0-1 points), 123 (26.2%) had a possibility of limited literacy (2-3 points), and 212 (45.1%) showed adequate literacy (4-6 points).

Correlations

There was a strong, positive correlation between the NVS and the objective numeracy scale (r = 0.510; P <.0005). Objective and subjective numeracy were moderately and positively correlated (r = 0.475; P <.0005). The correlations between the NVS and the subjective numeracy and graph literacy scale scores were also positive but moderate (r = 0.352; P <.0005; and r = 0.497; P <.0005, respectively).

Hospitalizations

There were 1081 hospitalizations over a median (IQR) follow-up period of 2434 (17) days, with a range of 0 to 74 hospital admissions. In the year before evaluation of health literacy, numeracy, and graph literacy, 115 patients (24.5%) had at least 1 hospitalization and 355 (75.5%) did not have any hospitalizations. Over the follow-up period, 206 participants (43.8%) did not have any hospitalizations, whereas 264 (56.0%) had at least 1 hospitalization. Among the 1080 hospitalizations in this cohort, the leading causes of hospital admission were mental health issues including substance abuse (239; 22%), followed by pulmonary (181; 17%), cardiovascular (167; 15%), and gastrointestinal (106; 10%) conditions. Among the pulmonary group, acute exacerbations of chronic obstructive pulmonary disease (COPD) were the most common cause of hospitalization, and acute on chronic congestive heart failure exacerbations were the most frequent diagnosis among the cardiovascular causes. Unspecified abdominal pain was the most common gastrointestinal diagnosis.

As shown in Table 2, using the Andersen-Gill model fully adjusted for covariates (model 4), health literacy (HR, 0.98; 95% CI, 0.91-1.01; P = .637), objective numeracy (HR, 0.96; 95% CI, 0.91-1.02; P = .178), subjective numeracy (HR, 0.99; 95% CI, 0.97-1.01; P = .186), and graph literacy (HR, 0.97; 95% CI, 0.93-1.02; P = .199) scores were not associated with all-cause hospitalizations. However, some differences appeared after conducting sensitivity analysis. In terms of age, there were no associations between health literacy, numeracy, or graph literacy and all-cause hospitalizations in participants younger than 60 years or those 60 years and older after adjustment. There were significant associations of higher subjective numeracy with lower risk for all-cause hospitalizations in African Americans after adjustment for covariates (HR, 0.98; 95% CI, 0.95-1.00; P = .048) but no association of health literacy (HR, 1.02; 95% CI, 0.92-1.14; P = .721), objective numeracy (HR, 0.94; 95% CI, 0.87-1.02; P = .137), or graphical literacy (HR, 0.96; 95% CI, 0.91-1.02; P = .226) with all-cause hospitalizations in African Americans. There were no significant associations of health literacy, numeracy, and graph literacy with all-cause hospitalizations in whites. After dividing the groups into those with and those without hospitalizations in the previous year, differences emerged only in those participants with prior hospitalizations. Higher objective (HR, 0.89; 95% CI, 0.83-0.97; P = .005) and subjective (HR, 0.96; 95% CI, 0.93-0.99; P = .044) numeracy scores were associated with a lower risk of all-cause hospitalizations after adjustment. Neither health literacy nor graph literacy showed any associations with all-cause hospitalizations in those individuals having at least 1 hospitalization in the previous year.

Mortality

Over a median (IQR) follow-up period of 2434 (7) days, 63 deaths occurred. As shown in Table 3, using a Cox regression analysis fully adjusted for covariates (model 4), higher scores of health literacy (HR, 0.93; 95% CI, 0.81-1.06; P = .249), objective numeracy (HR, 0.99; 95% CI, 0.89-1.11; P = .910), and subjective numeracy (HR, 1.01; 95% CI, 0.98-1.04; P = .537) were not associated with all-cause mortality during follow-up. Although the univariate and multivariate Cox regression analyses in model 1 show that higher graph literacy scores were associated with a higher risk of all-cause mortality, subsequent adjusted models showed no association with all-cause mortality (model 4: HR, 0.93; 95% CI, 0.85-1.01; P = .084) (Table 3). In sensitivity analyses, there were no associations between health literacy, numeracy, or graph literacy and all-cause mortality after covariate adjustment according to age groups or race. Although there were no associations of any of the literacy domains with all-cause mortality in those participants without prior hospitalizations, there was a significant association between higher graph literacy scores and lower mortality in those individuals with at least 1 hospitalization in the previous year (unadjusted HR, 0.86; 95% CI, 0.77-0.97; P = .017; adjusted HR, 0.85; 95% CI, 0.74-0.97; P = .018). Neither health literacy nor numeracy was associated with all-cause mortality in those with or without hospitalizations in the previous year.

DISCUSSION

Contrary to our main hypothesis, the study did not show any associations of health literacy, numeracy, or graph literacy scores with lower risk of all-cause hospitalization or mortality after adjustment. There were, however, group differences between literacy domains and risk of all-cause hospitalization or mortality. The association of subjective numeracy with all-cause hospitalizations was seen in African Americans. Both higher objective and subjective numeracy were associated with future all-cause hospitalizations only for those individuals with a history of hospitalization in the previous year. Higher graph literacy was associated with lower all-cause mortality risk in those individuals with a history of hospitalization.

Health literacy has been associated with an increased risk of hospitalizations in prior cross-sectional and cohort studies.5,24,25 In contrast to those studies, our sample was younger and predominantly male, and our sample size is relatively small and may limit the interpretation of our results. Health literacy was assessed with the NVS, an instrument that tends to overestimate the proportion of patients with inadequate health literacy.16 It is also worth noting that adjustment for hospitalizations in the previous year was a factor that was not considered in the analysis of previous widely cited studies.5,24,25 Individuals with a history of previous hospitalization represent a group at higher risk of future hospitalizations.26,27 Failure to adjust for this covariate may potentially lead to an overestimation of the general effects of health literacy on future hospitalizations. Future studies would do well to account for a history of hospitalizations in the analysis. Also, as suggested earlier, a possible explanation for our findings is that the structure of the VA healthcare system helps mitigate the health risks, including risk of hospitalization, that might otherwise be implied by lower scores on the literacy and numeracy assessments. Factors in this could include emphases in the VA system on clear and multimodal communication, support of patient self-management skills, care coordination, and ongoing quality improvement.

The prior research evidence for the association between health literacy and mortality in older adults has been robust and consistent across studies.28-35 The absence of an association between health literacy scores and all-cause mortality in our study may be explained in part by the younger age of our participants and smaller sample size compared with the cited studies. The duration of the follow-up period may have underestimated the effects of health literacy on mortality risk.

Although there were no significant overall associations with future all-cause hospitalizations or mortality, there were subgroup differences. For patients with a history of hospitalization, numeracy showed an association with risk of further hospitalizations. The results of the subgroup analysis suggest specific hypotheses that can be tested in future research. Subjective numeracy may be associated with all-cause hospitalizations in African Americans. Both higher objective and subjective numeracy show an association with future all-cause hospitalizations only for those individuals with a history of hospitalizations in the previous year. Higher graph literacy may be associated with lower all-cause mortality risk in those individuals with a history of hospitalization. Most research on health numeracy has focused on patient decision making and risk communication, but little is known regarding the association of numeracy with clinical and utilization outcomes.36 Two studies reported an association of low health numeracy with higher risk of hospitalizations, one a small cross-sectional study consisting of a mostly African American group of female patients with asthma37 and the other a prospective cohort of patients with heart failure.38 No previous studies looked at the association of graph literacy with hospitalizations. Explanations for how lower numeracy may contribute to an increased risk of hospitalizations may be related to its role in patient self-management. Growing evidence shows the benefits of effective patient self-management programs in reducing healthcare utilization.39 Health numeracy is considered by some investigators to be an even more important ability than health literacy in enabling effective patient self-management.36,40,41 Patients with previous hospitalizations may experience more frequent interactions with the healthcare system that would demand from them more responsibility for self-management activities.42 Numeracy deficits may hinder these self-care abilities while patients experience the stress and complexity that characterize hospital discharge. As the direction of causation in the associations seen is not known, it should also be considered that frequent hospitalizations, which may be associated with lower overall health status and greater cognitive dysfunction,43,44 may undermine patients’ numeracy abilities and/or their confidence. The potential erosion of confidence in numeracy abilities may be particularly relevant to the finding of an association of subjective but not objective numeracy with all-cause hospitalizations in African Americans. Although there were no significant overall associations with future all-cause hospitalizations or mortality, exploratory subgroup analyses reveal some avenues for future research. Future research should examine in detail the relationship between health numeracy and healthcare utilization outcomes in longitudinal studies that include more diverse populations and varied healthcare systems.

The nonsignificant trend for lower all-cause mortality in individuals with higher graph literacy and the significant association with all-cause mortality in the subgroup of individuals with at least 1 hospitalization in the previous year deserve explanation. The research evidence on these domains is sparse. Lower numeracy was independently associated with increased all-cause mortality after an emergency department visit in one study,45 but no such evidence exists for graph literacy. Graphical abilities may be necessary when building self-management skills that lead to better health outcomes. Individuals with inadequate graph literacy may experience problems using and processing the rapidly growing amount of health information available in the graphic formats of the internet.46,47 Understanding graphs demands complex visuospatial and executive function abilities,47,48 which may be impaired in patients with early cognitive impairment, a known risk factor for mortality.49,50 As mentioned above, cognitive impairment is also a prevalent condition in individuals with the multiple hospital admissions that can accompany increased risk of mortality.43 The higher all-cause mortality of the subgroup of patients with previous hospitalizations is congruent with the expected higher risk nature of this group. Hospitalizations are linked to a multitude of adverse events that contribute to decreased survival, and these effects are probably cumulative.51,52 It follows that patients with a history of hospitalization would be more likely to benefit from interventions that support self-management,53,54 as these activities may subsequently affect mortality.

Another possible explanation for the negative results in our study may be related to the higher rate of hospitalizations due to mental health conditions. The relationship between health literacy and mental health illness is complex and still not well understood.55,56 In the context of an acute psychiatric decompensation, the effects of mental health issues may trounce any effects that health literacy domains may have in these patients. The second most common cause of hospitalizations in our study was secondary to COPD exacerbations. A recent meta-analysis showed that self-management interventions did not have an effect on hospitalizations in patients with COPD,57 another potential reason for the lack of an association of health literacy and hospitalizations in our findings. On the other hand, it seems reasonable to assume that, similar to our claims regarding the independent effects of literacy constructs on future hospitalizations, a history of previous hospitalization may just reflect the effects of undiscovered literacy constructs. In the latter case, adjustment for previous hospitalizations may not be necessary. However, even after excluding for the effects of prior hospitalizations on future hospitalizations, the results from multivariate analyses (model 4) (Table 2) still did not reach significance. A relatively small sample size may have resulted in underpowered models. Therefore, our findings should be considered exploratory. Future studies could investigate the specific aspects of numeracy and graph literacy that predict mortality in high-risk, high-need individuals.

Strengths and Limitations

Strengths of this study include the complete evaluation of health literacy, numeracy, and graph literacy domains; inclusion of data from electronic health records; a relatively long period of follow-up; and exclusion of participants with cognitive impairment or depression. Limitations of this study include the use of a convenience sample of male veterans at 1 medical center, which may not be representative of the older adult veteran population as a whole. Veterans with better overall health status and higher levels of health literacy may be more likely to participate. The ethnic, racial, educational, and socioeconomic composition and the structure of the healthcare system may be different from those of other settings. The Veterans Affairs healthcare system provides acute care services to many US veterans as part of a comprehensive package of healthcare services. However, previous research in the VA has shown that more than two-thirds of veterans enrolled in the VA have access to private insurance or other government healthcare programs, such as Medicaid and Medicare, and therefore may have received acute care at non-VA facilities. A limitation of this study is the lack of information regarding hospitalizations that may have occurred outside the VA. Another limitation is the inclusion of a sample of predominantly male veterans, which may limit the generalizability of the results to the general population of female patients and nonveterans with differing levels of health literacy. Future research should increase the participant sample size, include more women, and examine participants with more diverse backgrounds to better understand what roles health literacy, numeracy, and graph literacy might play in healthcare utilization. Future studies may also benefit from accounting for history of hospitalizations in the analysis.

CONCLUSIONS

This study did not show any overall associations of health literacy, numeracy, or graph literacy scores with all-cause hospitalization or mortality. Higher subjective numeracy in African Americans and both objective and subjective numeracy in individuals with past history of hospitalizations were associated with a lower risk for all-cause hospitalizations. Higher graph literacy was associated with a lower all-cause mortality risk only in those individuals with a past history of hospitalization.Author Affiliations: Geriatric Research, Education and Clinical Center, Veterans Successful Aging for Frail Elders, Miami VA Healthcare System (JF-G, YNM, RA-U, DS, AS, DB, MD, JGR), Miami, FL; University of Miami Miller School of Medicine (JF-G, YNM, RA-U, DS, AS, MD, JGR), Miami, FL.

Source of Funding: None.

Author Disclosures: The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (MD, JGR); acquisition of data (JF-G, YNM, DS, AS, DB, JGR); analysis and interpretation of data (JF-G, YNM, RA-U, DS, DB, JGR); drafting of the manuscript (JF-G, RA-U, JGR); critical revision of the manuscript for important intellectual content (RA-U, MD, JGR); statistical analysis (JF-G, YNM, RA-U, DS, AS, DB, JGR); provision of patients or study materials (AS, MD, JGR); administrative, technical, or logistic support (MD, JGR); and supervision (RA-U, MD, JGR).

Address Correspondence to: Jorge G. Ruiz, MD, Geriatric Research, Education and Clinical Center, Bruce W. Carter Miami VA Medical Center, 1201 NW 16th St, Miami, FL 33125. Email: j.ruiz@miami.edu.REFERENCES

1. Institute of Medicine. Health Literacy: A Prescription to End Confusion. Washington, DC: National Academies Press; 2004.

2. Pleasant A, Rudd RE, O’Leary C, et al. Considerations for a new definition of health literacy: National Academy of Medicine website. nam.edu/considerations-for-a-new-definition-of-health-literacy. Published April 4, 2016. Accessed June 15, 2019.

3. Peters E. Beyond comprehension: the role of numeracy in judgments and decisions. Curr Dir Psychol Sci. 2012;21(1):31-35. doi: 10.1177/0963721411429960.

4. Galesic M, Garcia-Retamero R. Graph literacy: a cross-cultural comparison. Med Decis Making. 2010;31(3):444-457. doi: 10.1177/0272989X10373805.

5. Berkman ND, Sheridan SL, Donahue KE, Halpern DJ, Crotty K. Low health literacy and health outcomes: an updated systematic review. Ann Intern Med. 2011;155(2):97-107. doi: 10.7326/0003-4819-155-2-201107190-00005.

6. Agha Z, Lofgren RP, VanRuiswyk JV, Layde PM. Are patients at Veterans Affairs medical centers sicker? a comparative analysis of health status and medical resource use. Arch Intern Med. 2000;160(21):3252-3257. doi: 10.1001/archinte.160.21.3252.

7. Steinman MA, Lee SJ, Boscardin WJ, et al. Patterns of multimorbidity in elderly veterans. J Am Geriatr Soc. 2012;60(10):1872-1880. doi: 10.1111/j.1532-5415.2012.04158.x.

8. Hastings SN, Smith VA, Weinberger M, Oddone EZ, Olsen MK, Schmader KE. Health services use of older veterans treated and released from Veterans Affairs medical center emergency departments. J Am Geriatr Soc. 2013;61(9):1515-1521. doi: 10.1111/jgs.12417.

9. Zulman DM, Pal Chee C, Wagner TH, et al. Multimorbidity and healthcare utilisation among high-cost patients in the US Veterans Affairs Health Care System. BMJ Open. 2015;5(4):e007771. doi: 10.1136/bmjopen-2015-007771.

10. Rodriguez V, Andrade AD, Garcia-Retamero R, et al. Health literacy, numeracy, and graphical literacy among veterans in primary care and their effect on shared decision making and trust in physicians. J Health Commun. 2013;18(suppl 1):273-289. doi: 10.1080/10810730.2013.829137.

11. Rosland AM, Nelson K, Sun H, et al. The patient-centered medical home in the Veterans Health Administration. Am J Manag Care. 2013;19(7):e263-e272.

12. Nelson KM, Helfrich C, Sun H, et al. Implementation of the patient-centered medical home in the Veterans Health Administration: associations with patient satisfaction, quality of care, staff burnout, and hospital and emergency department use. JAMA Intern Med. 2014;174(8):1350-1358. doi: 10.1001/jamainternmed.2014.2488.

13. Lipkus IM, Peters E. Understanding the role of numeracy in health: proposed theoretical framework and practical insights. Health Educ Behav. 2009;36(6):1065-1081. doi: 10.1177/1090198109341533.

14. Sharit J, Lisigurski M, Andrade AD, et al. The roles of health literacy, numeracy, and graph literacy on the usability of the VA’s personal health record by veterans. J Usability Stud. 2014;9(4):173-193.

15. Irizarry T, DeVito Dabbs A, Curran CR. Patient portals and patient engagement: a state of the science review. J Med Internet Res. 2015;17(6):e148. doi: 10.2196/jmir.4255.

16. Weiss BD, Mays MZ, Martz W, et al. Quick assessment of literacy in primary care: the Newest Vital Sign [erratum in Ann Fam Med. 2006;4(1):83. doi: 10.1370/afm.506]. Ann Fam Med. 2005;3(6):514-522. doi: 10.1370/afm.405.

17. Schwartz LM, Woloshin S, Black WC, Welch HG. The role of numeracy in understanding the benefit of screening mammography. Ann Intern Med. 1997;127(11):966-972. doi: 10.7326/0003-4819-127-11-199712010-00003.

18. Lipkus IM, Samsa G, Rimer BK. General performance on a numeracy scale among highly educated samples. Med Decis Making. 2001;21(1):37-44. doi: 10.1177/0272989X0102100105.

19. Cokely ET, Galesic M, Schulz E, Ghazal S, Garcia-Retamero R. Measuring risk literacy: the Berlin Numeracy Test. Judgm Decis Mak. 2012;7(1):25-47.

20. Fagerlin A, Zikmund-Fisher BJ, Ubel PA, Jankovic A, Derry HA, Smith DM. Measuring numeracy without a math test: development of the Subjective Numeracy Scale. Med Decis Making. 2007;27(5):672-680. doi: 10.1177/0272989X07304449.

21. Maynard C. Ascertaining veterans’ vital status: VA data sources for mortality ascertainment and cause of death. US Department of Veterans Affairs website. hsrd.research.va.gov/for_researchers/cyber_seminars/archives/video_archive.cfm?SessionID=1242. Published March 6, 2017. Accessed June 21, 2019.

22. Gao J, Moran E, Li YF, Almenoff PL. Predicting potentially avoidable hospitalizations. Med Care. 2014;52(2):164-171. doi: 10.1097/MLR.0000000000000041.

23. Wang L, Porter B, Maynard C, et al. Predicting risk of hospitalization or death among patients receiving primary care in the Veterans Health Administration. Med Care. 2013;51(4):368-373. doi: 10.1097/MLR.0b013e31827da95a.

24. Baker DW, Parker RM, Williams MV, Clark WS. Health literacy and the risk of hospital admission. J Gen Intern Med. 1998;13(12):791-798. doi: 10.1046/j.1525-1497.1998.00242.x.

25. Baker DW, Gazmararian JA, Williams MV, et al. Functional health literacy and the risk of hospital admission among Medicare managed care enrollees. Am J Public Health. 2002;92(8):1278-1283. doi: 10.2105/ajph.92.8.1278.

26. Khalid JM, Raluy-Callado M, Curtis BH, Boye KS, Maguire A, Reaney M. Rates and risk of hospitalisation among patients with type 2 diabetes: retrospective cohort study using the UK General Practice Research Database linked to English Hospital Episode Statistics. Int J Clin Pract. 2014;68(1):40-48. doi: 10.1111/ijcp.12265.

27. Lohman MC, Scherer EA, Whiteman KL, Greenberg RL, Bruce ML. Factors associated with accelerated hospitalization and re-hospitalization among Medicare home health patients. J Gerontol A Biol Sci Med Sci. 2018;73(9):1280-1286. doi: 10.1093/gerona/glw335.

28. Sudore RL, Yaffe K, Satterfield S, et al. Limited literacy and mortality in the elderly: the health, aging, and body composition study. J Gen Intern Med. 2006;21(8):806-812. doi: 10.1111/j.1525-1497.2006.00539.x.

29. Baker DW, Wolf MS, Feinglass J, Thompson JA, Gazmararian JA, Huang J. Health literacy and mortality among elderly persons. Arch Intern Med. 2007;167(14):1503-1509. doi: 10.1001/archinte.167.14.1503.

30. Baker DW, Wolf MS, Feinglass J, Thompson JA. Health literacy, cognitive abilities, and mortality among elderly persons. J Gen Intern Med. 2008;23(6):723-726. doi: 10.1007/s11606-008-0566-4.

31. Wolf MS, Feinglass J, Thompson J, Baker DW. In search of ‘low health literacy’: threshold vs. gradient effect of literacy on health status and mortality. Soc Sci Med. 2010;70(9):1335-1341. doi: 10.1016/j.socscimed.2009.12.013.

32. Bostock S, Steptoe A. Association between low functional health literacy and mortality in older adults: longitudinal cohort study. BMJ. 2012;344:e1602. doi: 10.1136/bmj.e1602.

33. McNaughton CD, Cawthon C, Kripalani S, Liu D, Storrow AB, Roumie CL. Health literacy and mortality: a cohort study of patients hospitalized for acute heart failure [erratum in J Am Heart Assoc. 2015;4(6):e000682. doi: 10.1161/JAHA.115.000682]. J Am Heart Assoc. 2015;4(5):e001799. doi: 10.1161/JAHA.115.001799.

34. Moser DK, Robinson S, Biddle MJ, et al. Health literacy predicts morbidity and mortality in rural patients with heart failure. J Card Fail. 2015;21(8):612-618. doi: 10.1016/j.cardfail.2015.04.004.

35. Smith SG, Jackson SE, Kobayashi LC, Steptoe A. Social isolation, health literacy, and mortality risk: findings from the English Longitudinal Study of Ageing. Health Psychol. 2018;37(2):160-169. doi: 10.1037/hea0000541.

36. Zikmund-Fisher BJ, Mayman G, Fagerlin A. Patient numeracy: what do patients need to recognize, think or do with health numbers? In: Anderson BL, Schulkin J, eds. Numerical Reasoning in Judgments and Decision Making About Health. Cambridge, United Kingdom: Cambridge University Press; 2014:80-104.

37. Apter AJ, Cheng J, Small D, et al. Asthma numeracy skill and health literacy. J Asthma. 2006;43(9):705-710. doi: 10.1080/02770900600925585.

38. McNaughton CD, Collins SP, Kripalani S, et al. Low numeracy is associated with increased odds of 30-day emergency department or hospital recidivism for patients with acute heart failure. Circ Heart Fail. 2013;6(1):40-46. doi: 10.1161/CIRCHEARTFAILURE.112.969477.

39. Panagioti M, Richardson G, Small N, et al. Self-management support interventions to reduce health care utilisation without compromising outcomes: a systematic review and meta-analysis. BMC Health Serv Res. 2014;14:356. doi: 10.1186/1472-6963-14-356.

40. Rothman RL, Montori VM, Cherrington A, Pignone MP. Perspective: the role of numeracy in health care. J Health Commun. 2008;13(6):583-595. doi: 10.1080/10810730802281791.

41. Osborn CY, Cavanaugh K, Wallston KA, Rothman RL. Self-efficacy links health literacy and numeracy to glycemic control. J Health Commun. 2010;15(suppl 2):146-158. doi: 10.1080/10810730.2010.499980.

42. Jack BW, Chetty VK, Anthony D, et al. A reengineered hospital discharge program to decrease rehospitalization: a randomized trial. Ann Intern Med. 2009;150(3):178-187. doi: 10.7326/0003-4819-150-3-200902030-00007.

43. Mathews SB, Arnold SE, Epperson CN. Hospitalization and cognitive decline: can the nature of the relationship be deciphered? Am J Geriatr Psychiatry. 2014;22(5):465-480. doi: 10.1016/j.jagp.2012.08.012.

44. Marzec LN, Carey EP, Lambert-Kerzner AC, et al. Cognitive dysfunction and poor health literacy are common in veterans presenting with acute coronary syndrome: insights from the MEDICATION study. Patient Prefer Adherence. 2015;9:745-751. doi: 10.2147/PPA.S75110.

45. McNaughton C, Rothman R, Roumie CL. Associations between health literacy/numeracy and mortality in emergency department patients. Med Encount. 2016;30(1):84.

46. Ruiz JG, Andrade AD, Hogue C, et al. The association of graph literacy with use of and skills using an online personal health record in outpatient veterans. J Health Commun. 2016;21(suppl 2):83-90. doi: 10.1080/10810730.2016.1193915.

47. Sharit J, Idrees T, Andrade AD, et al. Use of an online personal health record’s Track Health function to promote positive lifestyle behaviors in veterans with prediabetes. J Health Psychol. 2018;23(5):681-690. doi: 10.1177/1359105316681065.

48. Shah P. Graph comprehension: the role of format, content and individual differences. In: Anderson M, Meyer B, Olivier P, eds. Diagrammatic Representation and Reasoning. London, United Kingdom: Springer London; 2002:173-185.

49. Bassuk SS, Wypij D, Berkman LF. Cognitive impairment and mortality in the community-dwelling elderly. Am J Epidemiol. 2000;151(7):676-688. doi: 10.1093/oxfordjournals.aje.a010262.

50. Gale CR, Martyn CN, Cooper C. Cognitive impairment and mortality in a cohort of elderly people. BMJ. 1996;312(7031):608-611. doi: 10.1136/bmj.312.7031.608.

51. Gillick MR, Serrell NA, Gillick LS. Adverse consequences of hospitalization in the elderly. Soc Sci Med. 1982;16(10):1033-1038. doi: 10.1016/0277-9536(82)90175-7.

52. Covinsky KE, Pierluissi E, Johnston CB. Hospitalization-associated disability: “she was probably able to ambulate, but I’m not sure.” JAMA. 2011;306(16):1782-1793. doi: 10.1001/jama.2011.1556.

53. Heisler M, Bouknight RR, Hayward RA, Smith DM, Kerr EA. The relative importance of physician communication, participatory decision making, and patient understanding in diabetes self-management. J Gen Intern Med. 2002;17(4):243-252. doi: 10.1046/j.1525-1497.2002.10905.x.

54. Ditewig JB, Blok H, Havers J, van Veenendaal H. Effectiveness of self-management interventions on mortality, hospital readmissions, chronic heart failure hospitalization rate and quality of life in patients with chronic heart failure: a systematic review. Patient Educ Couns. 2010;78(3):297-315. doi: 10.1016/j.pec.2010.01.016.

55. Sentell TL, Shumway MA. Low literacy and mental illness in a nationally representative sample. J Nerv Ment Dis. 2003;191(8):549-552. doi: 10.1097/01.nmd.0000082185.26868.dc.

56. Lincoln A, Espejo D, Johnson P, et al. Limited literacy and psychiatric disorders among users of an urban safety-net hospital’s mental health outpatient clinic. J Nerv Ment Dis. 2008;196(9):687-693. doi: 10.1097/NMD.0b013e31817d0181.

57. Jolly K, Majothi S, Sitch AJ, et al. Self-management of health care behaviors for COPD: a systematic review and meta-analysis. Int J Chron Obstruct Pulmon Dis. 2016;11:305-326. doi: 10.2147/COPD.S90812.