Abstract
Background and objectives: Healthcare providers communicate the risks and benefits of treatments using frequencies, percentages, or proportions. However, many patients lack the numerical skills needed to interpret this information accurately to make informed choices.
Design, setting, participants, and measurements: We assessed numeracy, the capacity to use and comprehend numbers, in a prospective cohort study of 187 patients with stage 4 to 5 chronic kidney disease or ESRD. Patients completed a three-item numeracy test and were assessed for global mental status, cognitive function, type of hemodialysis access, and kidney transplant use. We examined the association of numeracy with healthcare use and other cognitive and sociodemographic variables.
Results: Over 50% of patients answered one or fewer numeracy questions correctly. Although African Americans (P = 0.0001), women (P = 0.05), and the unemployed (P = 0.0004) demonstrated lower numeracy skills, numeracy deficits were prevalent in every subgroup. In analyses adjusted for demographics and length of follow-up, higher numeracy was significantly associated with receipt of a transplant or active waiting list status. Numeracy was not associated with dialysis modality or hemodialysis vascular access.
Conclusion: Similar to prior findings in the general population, these findings indicate that poor numeracy skills are very common in patients with advanced chronic kidney disease and end-stage renal disease. Additional research is needed to further explore whether poor numeracy is a barrier to receipt of a kidney transplant. Clinicians caring for patients with kidney disease should consider using tools to enhance communication and overcome limited numeracy skills.
Numeracy, also referred to as quantitative literacy, can be defined as the capacity to use and comprehend numbers in daily life (1,2). In the United States, an estimated 110 million adults lack the quantitative skills needed to complete more than simple, everyday activities (3). Although deficient numeracy skills can coexist with deficient literacy skills, recent findings indicate that many patients have even poorer quantitative skills (1,2). Inadequate numeracy skills appear to be more prevalent in African American, Hispanic American, and elderly populations and may contribute to health disparities (3–5). Numeracy may be particularly important in the setting of chronic kidney disease (CKD) given the disproportionate representation of populations with poor numeracy skills coupled with the integral role that advanced planning plays in optimal CKD care (e.g., pre-emptive transplantation or arteriovenous fistula [AVF] placement).
Health numeracy as defined by Golbeck, refers to the degree that individuals can apply numerical, graphical, and statistical skills to understand and act on “health information needed to make effective health decisions” (6). Health numeracy can be classified into categories. A few nonexhaustive examples of these include simple number or quantity recognition (basic health numeracy), the ability to perform one-step manipulations on numbers or to understand proportions (computational health numeracy), the ability to make approximations and deductions (analytical health numeracy), and the ability to convert and meaningfully compare between percentages, proportions, and probabilities (statistical health numeracy) (6). Analytical and statistical health numeracy require a higher-level skill set than other categories of health numeracy.
Since healthcare providers often communicate with patients by using numerical information, deficient numeracy skills may be a barrier to care in patients with chronic health conditions. Deficient numeracy skills may limit patient involvement in the decision-making process or result in choices that are not in accordance with the patient’s beliefs and values (7,8). Poorer patient involvement, understanding, or adherence to the medical regimen may contribute to suboptimal outcomes (1,9–11).
There are few data characterizing numeracy skills in the CKD and end-stage renal disease (ESRD) populations (11). Yet, these patients regularly interact with the healthcare system due to their frequent coexisting comorbidities and the systemic manifestations of kidney disease. Additionally, many interventions that can significantly improve survival and quality of life in ESRD or advanced CKD are invasive and require weighing risks and benefits (e.g., kidney transplantation and AVF placement). Recent evidence suggests that patients with kidney disease who lack adequate health literacy skills may be disadvantaged by inferior access to kidney transplantation (12). This may play a role in the well-documented disparities in transplantation rates among African Americans, women, and those of lower socioeconomic status (12–16). Preliminary findings also suggest that numeracy deficits may be prevalent in the kidney transplant population (17).
Characterization of the relationship between numeracy and cognitive function in the setting of CKD may improve our understanding of factors that contribute to numeracy. While multiple studies have documented the presence of cognitive deficits in CKD and ESRD (18–21) and previous findings in patients with hypertension and congestive heart failure have established an association between cognitive performance and health literacy (22,23), there is a gap in our understanding of cognitive performance and numeracy. In this report, we describe the numeracy skills of an outpatient cohort of ESRD and stage 4 to 5 CKD patients assessed using a measure of analytical and statistical numeracy. We also explore the sociodemographic characteristics associated with numeracy. Second, we examine the relationship between patient numeracy and kidney disease-specific healthcare utilization. Third, we explore the association of numeracy with cognitive function.
Materials and Methods
Study Setting and Design
As part of a previously described (24,25) larger prospective cohort study investigating memory, sleep, and quality of life in adult, English-speaking patients with advanced CKD, we approached patients during their routine CKD clinic visit, dialysis clinic visit, or their initial evaluation at a kidney transplantation clinic between March 2004 and October 2007. Exclusion criteria included extremes of age and active, severe medical or psychiatric illness as has been previously described (24,25).
As of October 2007, 269 patients provided written informed consent. Of these, 23 either met exclusion criteria or withdrew consent and 54 did not complete the cognitive assessment for the following reasons: 46 because of time constraints or loss of interest, four due to kidney transplantation before protocol initiation, three due to patient death, and one because of inability to tolerate polysomnography as required for the parent study. Of the remaining 192 patients, five had missing numeracy test data and were excluded from our analysis. The University of Pittsburgh Institutional Review Board approved this study.
The baseline data collection session included a brief standardized health interview and questionnaire, assessment of current medication use, and BP and anthropometric measurements. Age, sex, race, education, and employment status were self-reported. We obtained a history of physician-diagnosed medical illnesses, previous surgical treatments, or medical procedures from the health interview. Additionally, we abstracted standard serum laboratory tests, current and previous hemodialysis access, and current and previous kidney transplantation status from the medical record.
Health Access Utilization
Dialysis modality was categorized as hemodialysis (HD) or peritoneal dialysis (PD). Vascular access was categorized as an AVF/arteriovenous graft (AVG) or dialysis catheter in patients actively receiving HD treatment. To examine patient medical decision-making, study participants with any history of an AVF/AVG placement were given credit for an AVF/AVG regardless of its current functional status, because functional status may be related to factors out of the patient’s control. Kidney transplant utilization was categorized as active/transplanted or inactive. To examine patient medical decision-making, participants were classified as active/transplanted if they had a prior kidney transplant, regardless of its functional status, or had received a new kidney transplant after study enrollment, or were active on the University of Pittsburgh’s kidney transplant waiting list at any point after study enrollment. Active status on the kidney transplant waiting list was determined by local transplant center medical record abstraction and included an assessment of wait list status at least 1 year following the patient enrollment date. Duration of follow-up was calculated as the number of elapsed days from the date of study enrollment to the last assessment of kidney transplant wait list status.
Measures of Numeracy and Cognitive Function
Numeracy was assessed using a validated three-question instrument (26). The three items were: (1) the number of expected heads in 1000 coin flips; (2) to convert from a percentage (1%) to a proportion (‘_ in 1000′); and (3) to convert a proportion (1 in 1000) to a percentage (‘_%’). The score is the total number of correct responses. This instrument and modified versions have been widely used in prior studies assessing numeracy skills (9,26–28).
Concomitant with the numeracy assessment, participants completed a neuropsychological examination with a battery of measures selected because they have been established as reliable and valid measures of the domains that they assess and, in particular, sensitive to deficits typically found in middle-aged to older adults with chronic illnesses (29). This examination was conducted by a trained research coordinator at a time when the participant felt they were at their best. This cognitive assessment was typically performed in the morning of a nondialysis day at the patient’s home. The cognitive examination assessed a range of functions including global mental status (Modified Mini-Mental State Examination [3MSE]) (30), intelligence quotient (Wechsler Abbreviated Scale of Intelligence [WASI]) (31), language (Controlled Oral Word Association [COWA] test) (32), and visual-spatial/constructional abilities (CLOX 1 and 2) (33–35).
Statistical Analysis
Baseline demographics, comorbidities, laboratory values, and cognitive measures were described using means and SD for continuous variables and percentages for categorical variables. Univariate relationships between numeracy scores and baseline demographic, clinical, and laboratory variables were assessed using ANOVA for continuous variables and the χ2 statistic or Fisher exact test for categorical variables. We used an ordinal logistic regression model to identify demographic and clinical factors independently associated with numeracy. Model selection included major demographic variables and those variables found to have a relationship (P < 0.10) in univariate analysis. With the major demographic factors forced into the model, we then used an iterative backwards selection process examining P values to derive a parsimonious model. The final model included age, sex, race, education, working status, diabetes, and hemoglobin.
The unadjusted association between numeracy and health access utilization variables was assessed using the χ2 statistic or Fisher exact test. We examined adjusted associations between clinical variables and kidney transplant utilization using a logistic regression model. Model selection followed the above outlined stepwise criteria. The final model included age, sex, race, numeracy score, and duration of study follow-up. The association between cognitive measures and numeracy was assessed using ANOVA and Pearson correlation coefficients. For all analyses, P values <0.05 were considered significant. Analyses were performed using SAS (version 8.1; SAS Institute, Cary, NC).
Results
Baseline Characteristics
The associations of participant demographic and baseline clinical characteristics with numeracy are shown in Table 1. More than half the participants answered one or fewer numeracy questions correctly, and less than 20% of the sample answered all three items correctly. Participant age was not associated with numeracy. Women, African Americans, and the unemployed had significantly lower numeracy than their counterparts. Participants without a 12th grade education also tended to have lower numeracy, although this difference was not statistically significant. Patients with nondialysis-dependent CKD and ESRD demonstrated similar numeracy. Higher hemoglobin and albumin levels were both associated with higher numeracy. Among nondialysis-dependent CKD patients, a higher estimated GFR was associated with higher numeracy. Among ESRD patients, a shorter dialysis vintage tended to be associated with higher numeracy, although this did not achieve statistical significance.
Table 1.
Demographic characteristics and numeracy
| Totaln = 187 | Numeracy Score |
Pa | ||||
|---|---|---|---|---|---|---|
| 0n = 35 | 1n = 61 | 2n = 57 | 3n = 34 | |||
| Demographics | ||||||
| Age | 52.0 (14.8) | 55.3 (11.9) | 53.4 (15.0) | 49.3 (15.1) | 50.7 (16.2) | 0.2 |
| Gender | 0.05 | |||||
| Female | 71 | 16 (22.5) | 27 (38.0) | 22 (31.0) | 6 (8.5) | |
| Male | 116 | 19 (16.4) | 34 (29.3) | 35 (30.2) | 28 (24.1) | |
| Race | <0.001 | |||||
| African American | 68 | 24 (35.3) | 20 (29.4) | 17 (25.0) | 7 (10.3) | |
| Non-African American | 119 | 11 (9.2) | 41 (34.5) | 40 (33.6) | 27 (22.7) | |
| Employment Status | <0.001 | |||||
| Student/employed | 52 | 3 (5.8) | 13 (25.0) | 18 (34.6) | 18 (34.6) | |
| Unemployed | 135 | 32 (23.7) | 48 (35.6) | 39 (28.9) | 16 (11.9) | |
| Educational Level | 0.06b | |||||
| 12th grade education or greater | 157 | 26 (16.6) | 49 (31.2) | 50 (31.9) | 32 (20.4) | |
| Less than 12th grade education | 30 | 9 (30.0) | 12 (40.0) | 7 (23.3) | 2 (6.7) | |
| Tobacco Use | 0.6b | |||||
| Active smoker | 25 | 7 (28.0) | 7 (28.0) | 8 (32.0) | 3 (12.0) | |
| Former/nonsmoker | 162 | 28 (17.3) | 54 (33.3) | 49 (30.2) | 31 (19.1) | |
| Medical Comorbidities | ||||||
| Diabetes | 74 | 19 (25.7) | 26 (35.1) | 20 (27.0) | 9 (12.2) | 0.07 |
| Nondiabetic | 113 | 16 (14.2) | 35 (31.0) | 37 (32.7) | 25 (22.1) | |
| History of stroke | 18 | 5 (27.8) | 5 (27.8) | 6 (33.3) | 2 (11.1) | 0.7b |
| No history of stroke | 169 | 30 (17.8) | 56 (33.1) | 51 (30.2) | 32 (18.9) | |
| Coronary artery disease | 44 | 8 (18.2) | 15 (34.1) | 13 (29.5) | 8 (18.2) | 0.9 |
| No history of coronary artery disease | 143 | 27 (18.9) | 46 (32.2) | 44 (30.8) | 26 (18.2) | |
| History of depression | 26 | 7 (26.9) | 9 (34.6) | 7 (26.9) | 3 (11.5) | 0.4 |
| No history of depression | 161 | 28 (17.4) | 52 (32.3) | 50 (31.1) | 31 (19.3) | |
| Renal Disease | 0.4 | |||||
| CKD stage 4 to 5 | 75 | 13 (17.3) | 24 (32.0) | 20 (26.7) | 18 (24.0) | |
| ESRD | 112 | 22 (19.6) | 37 (33.0) | 37 (33.0) | 16 (14.3) | |
| Duration of follow-upc (years) n = 187 | 2.88 (1.03) | 2.87 (1.01) | 2.71 (1.04) | 3.01 (1.06) | 2.98 (1.00) | 0.4 |
| Dialysis vintaged (years) n = 112 | 2.26 (2.06) | 2.94 (2.51) | 2.47 (2.03) | 2.08 (1.90) | 1.27 (1.48) | 0.09 |
| Labs | ||||||
| Hemoglobin (g/dl) | 11.75 (1.3) | 11.3 (1.3) | 11.7 (1.4) | 11.8 (1.3) | 12.3 (1.2) | 0.03 |
| Albumin (g/dl) | 3.80 (0.46) | 3.55 (0.58) | 3.79 (0.43) | 3.94 (0.40) | 3.85 (0.39) | 0.002 |
| Phosphorus (mg/dl) | 5.05 (1.3) | 5.28 (1.15) | 4.94 (1.14) | 5.27 (1.48) | 4.67 (1.25) | 0.1 |
| Creatininee (mg/dl) n = 75 | 4.3 (1.6) | 5.1 (1.3) | 3.9 (1.6) | 4.8 (1.6) | 3.9 (1.6) | 0.052 |
| estimated GFRe (ml/min per 1.73 m2) n = 75 | 17.6 (8.3) | 13.0 (4.0) | 19.8 (10.1) | 15.1 (5.3) | 21.1 (8.7) | 0.01 |
Continuous variables are presented as means with standard deviations. Categorical variables are expressed as frequencies and percentages.
Chi-square or analysis of variance were used as appropriate.
Fisher exact test.
Duration of follow-up: Elapsed time from date of study enrollment to last assessment of kidney transplant status.
For ESRD patients only.
For nondialysis-dependent CKD patients.
Numeracy score represents the total number of correct answers on the three-item test.
Hemoglobin in g/dl can be converted to g/L by multiplying by 10.
Albumin in g/dl can be converted to g/L by multiplying by 10.
Phosphorus in mg/dl can be converted to mmol/L by multiplying by 0.3229.
All HD patients either had a Kt/V > 1.2 or a urea reduction ratio >65%, and neither of these were associated with numeracy (data not shown). Patients on PD versus HD did not differ in numeracy (data not shown). Thirty-eight patients had assessments pre- and approximately 6 months post-kidney transplant; numeracy was unchanged in these patients (data not shown). The 82 patients who were not included in this report were similar in age, sex, and race to the study participants. However, patients excluded from the study were more likely to be on HD (63% versus 46%, respectively) and less likely to have CKD stage 4 to 5 (18% versus 40%, P = 0.002).
Independent Associations with Numeracy
In the adjusted ordinal logistic regression model shown in Table 2, African-American race, less than a 12th grade education, nonstudent/unemployed work status, and lower hemoglobin were associated with lower numeracy. We tested for significant interactions between education and race as well as education and sex, however neither interaction was independently associated with numeracy, and their inclusion did not affect the model.
Table 2.
Factors independently associated with higher numeracy using ordinal logistic regressiona
| Adjusted OR (95% CI) | |
|---|---|
| Age | 0.99 (0.97 to 1.01) |
| Gender (female versus male) | 0.68 (0.37 to 1.23) |
| Race (AA versus non-AA) | 0.35 (0.19 to 0.67) |
| Education (12th grade versus less than 12th grade) | 3.55 (1.18 to 10.74) |
| Working status (student/employed versus nonstudent/unemployed) | 2.44 (1.25 to 4.77) |
| Diabetes (diabetic versus nondiabetic) | 0.61 (0.33 to 1.11) |
| Hemoglobin (per g/dl increase) | 1.28 (1.02 to 1.61) |
n = 187.
AA, African American.
Healthcare Use
The relationship of numeracy to CKD-specific healthcare use is shown in Table 3. Numeracy was not associated with dialysis modality or HD access. However, higher numeracy was associated with kidney transplantation or active status on the wait list (P = 0.01). After adjusting for age, sex, race, and duration of follow-up, this relationship remained significant as shown in Table 4. Excluding study patients enrolled primarily at their initial kidney transplant evaluation and patients with early stage 4 CKD (documented eGFR >20 ml/min per 1.73 m2) strengthened the association between numeracy and kidney transplant status (n = 83, odds ratio [OR] 1.89, 95% confidence interval [CI] 1.10 to 3.25). However, in the subset of patients with an eGFR <20 ml/min per 1.73 m2 and recruited primarily at their initial kidney transplant evaluation, numeracy was not independently associated with transplant utilization (n = 75, OR 1.13, 95% CI 0.56 to 2.25).
Table 3.
The association of numeracy with health care utilization
| Numeracy Score |
Pa | ||||
|---|---|---|---|---|---|
| 0 | 1 | 2 | 3 | ||
| ESRD (n = 112) | |||||
| PD | 1 (4.8) | 6 (28.6) | 9 (42.9) | 5 (23.8) | 0.1b |
| HD | 21 (23.1) | 31 (34.1) | 28 (30.8) | 11 (12.1) | |
| HD Accessc (n = 86) | |||||
| AV fistula or graft | 18 (24.3) | 25 (33.8) | 24 (32.4) | 7 (9.5) | 0.6b |
| Dialysis catheter | 3 (25.0) | 4 (33.3) | 3 (25.0) | 2 (16.7) | |
| Transplant Status (n = 187) | |||||
| Actively listed or ever transplanted | 18 (15.3) | 34 (28.8) | 37 (31.4) | 29 (24.6) | 0.01 |
| Not actively listed nor transplanted | 17 (24.6) | 27 (39.1) | 20 (29.0) | 5 (7.2) | |
Categorical variables expressed as frequency with percentage.
Chi-square used as appropriate.
Fisher exact test.
Only HD patients were included in this analysis.
Table 4.
Factors independently associated with kidney transplantation or active wait list status using logistic regressiona
| Adjusted OR (95% CI) | |
|---|---|
| Age | 0.95 (0.93 to 0.98) |
| Gender (female versus male) | 0.50 (0.25 to 1.01) |
| Race (AA versus non-AA) | 1.02 (0.48 to 2.20) |
| Numeracy | 1.44 (1.01 to 2.07) |
| Duration of follow-upb (per increase of 1 day) | 1.00 (1.00 to 1.01) |
n = 187.
Duration of follow-up: Elapsed days from date of study enrollment to last assessment of kidney transplant status.
Cognitive Function
Cognitive performance and numeracy associations are displayed in Table 5. Global mental status (as assessed by the 3MSE), intelligence (as assessed by WASI), and language (as assessed by the COWA) were each associated with numeracy (P < 0.0001 for each). Higher scores (indicating better performance) on each of these cognitive measures correlated with higher numeracy scores. Visual-spatial constructional abilities as assessed by CLOX-1 and CLOX-2 did not exhibit statistically reliable associations with numeracy (P = 0.06 and P = 0.08, respectively).
Table 5.
The association of numeracy with cognitive performance
| Cognitive measure | Total | Numeracy Score |
Associationa | |||
|---|---|---|---|---|---|---|
| 0 | 1 | 2 | 3 | |||
| 3MSEb | 92.0 (6.8) | 86.8 (9.5) | 91.4 (5.7) | 93.6 (5.4) | 95.65 (3.7) | r = 0.41 |
| P < 0.001 | ||||||
| IQc | 98.7 (15.8) | 83.8 (9.8) | 96.6 (13.9) | 101.1 (13.9) | 113.3 (12.6) | r = 0.56 |
| P < 0.001 | ||||||
| COWAd | 33.3 (13.2) | 25.5 (11.6) | 33.8 (12.5) | 32.4 (11.2) | 41.8 (14.4) | r = 0.33 |
| P < 0.001 | ||||||
| CLOX-1 ≥14e | 91 (49.2%) | 23 (25.3%) | 29 (31.9%) | 27 (29.7%) | 12 (13.2%) | r = −0.18 |
| P = 0.06 | ||||||
| CLOX-2 ≥15f | 72 (38.9%) | 18 (25.0%) | 26 (36.1%) | 20 (27.8%) | 8 (11.1%) | r = −0.19 |
| P = 0.08 | ||||||
Continuous variables are presented as means with SD. Categorical variables are expressed as frequencies and percentages.
Pearson correlation coefficients.
3MSE, higher score indicates better global mental status.
Intelligence quotient measured by WASI, higher score indicates better intellectual function.
COWA test, higher score indicates better language ability.
Clock drawing task score ≥14, higher score indicates better visuospatial/constructional/executive abilities.
Clock copying task score ≥15, higher score indicates better visuospatial/constructional abilities.
Discussion
The majority of patients with advanced CKD or ESRD exhibited low numeracy as assessed by a three-item numeracy measure. Over 50% of our kidney disease sample answered one or fewer of the three numeracy questions correctly. This suggests that the presence of low numeracy skills in the CKD and ESRD populations is quite common (and our findings may underestimate the true prevalence of low numeracy within the kidney disease population, as patients with severe comorbidities were excluded). African-American race, less than a 12th grade education, nonstudent/unemployed status, and lower hemoglobin were independently associated with lower numeracy. Nevertheless, even after controlling for these correlates of numeracy, greater numeracy was independently associated with receipt of a kidney transplant or active wait list status. Numeracy also significantly correlated with cognitive measures of global mental status, intellectual functioning, and language.
Our findings of widespread poor numeracy skills are consistent with Schwartz’s findings in a sample of approximately 300 middle-aged to elderly predominantly Caucasian women veterans, in which 58% of respondents answered one or fewer items correctly (26). Other studies have reported comparable results (27,36). For example, one study examining consecutive adult patients presenting to emergency departments in Massachusetts found that 43% of participants answered zero to one (of four) numeracy questions correctly. Approximately 45% of women and 63% of African Americans answered zero to one question correctly (similar to our findings of 60% of women, 65% of African Americans) (36). Additionally, studies using alternative measures of numeracy (e.g., Wide Range Achievement Test, 3rd edition) have documented widespread deficiencies, including nearly 70% of diabetics and general primary care patients with numeracy skills below the ninth grade level (1,37,38). Furthermore, a nationally representative survey also documented significantly lower numeracy skills in the general United States population compared with other developed countries, as well as disparities in scores between Caucasians and African Americans/Hispanic Americans (39).
Although our kidney disease sample demonstrated qualitatively similar numeracy deficiencies as previously reported general population samples, this should not provide false comfort to nephrologists. Together, these findings suggest that low numeracy is a very common problem for patients. However, the prevalence of low numeracy has important implications for CKD and ESRD patients. CKD and ESRD patients may not understand information presented using percentages and proportions; formats that are commonly used to communicate risks and benefits (4,7,8,26). Consequently, the potential for misguided interpretation and poor decision-making is significant. Given the frequency with which kidney disease patients make health-related decisions (e.g., daily medication adherence, daily dietary restriction adherence, thrice-weekly to daily dialysis treatment compliance, advanced care planning, etc.), the potential for numeracy to affect treatments and outcomes is substantial. To diminish the effects of inadequate patient numeracy, physicians should consider using visual aids to facilitate effective communication (4,7,40) and adhere to recent recommendations for conveying health risk information (7).
The findings of low numeracy in CKD and ESRD patients are also qualitatively similar to those from a pilot study suggesting that kidney transplant patients may exhibit low to marginal numeracy skills (17). Our study also complements the ESRD findings of Grubbs and colleagues who recently noted that approximately one-third of their dialysis patients exhibited inadequate health literacy (12). The CKD and ESRD population appear to demonstrate numeracy deficits as well.
Our findings also confirm that numeracy deficits may be prevalent even among those with a high school or college education. Just as in the Schwartz study, more than 90% of our sample had at least a 12th grade education, highlighting the prevalence of numeracy deficiencies despite a high school education (26). Our findings suggest that poor numeracy skills may be common even among educated CKD or ESRD patients. Healthcare providers should be aware of these deficiencies and avoid assuming that educated patients with kidney disease possess the requisite skills to understand numerical information.
Our findings also suggest that numeracy skills may play a role in the kidney transplant process independent of factors known to influence kidney transplant access including age, sex, and race. While these findings are limited by the lack of documentation of patients’ medical suitability for transplantation, these hypothesis generating results parallel recent findings that inadequate health literacy is associated with fewer referrals for kidney transplant evaluation (12). Numeracy could affect a patient’s transplant status if patients with poor numeracy skills: (1) fail to fully appreciate the morbidity and mortality benefits of kidney transplantation as communicated by their healthcare providers or (2) fail to understand the benefits and risks of kidney transplantation as explained in common media outlets and periodicals and therefore broach the topic with their physician less frequently. This may cause patients with lower numeracy to pursue or inquire about a kidney transplant workup less frequently. The independent association of numeracy with kidney transplantation was not present in patients recruited at their initial kidney transplant evaluation, suggesting that numeracy may be important in the earlier stages of the transplant process.
In addition to being a potential barrier to kidney transplant access, inadequate numeracy may also contribute to the documented race (14), gender (13), and socioeconomic (15) disparities in kidney transplantation. Analogous links between numeracy and racial disparities in glycemic control have been established (5). Alternatively, physician bias regarding race, gender, education, socioeconomic status, or other confounding variables may result in the less frequent referral of patients with poor numeracy skills for transplant evaluation. Larger confirmatory studies that assess patient numeracy in advanced CKD and provide longitudinal follow-up along with documentation of the reason for nonactive transplant status are needed.
The association of global mental status, intellectual functioning, and language with numeracy is a novel finding in adult patients with chronic disease. To date, numeracy has not been associated with intelligence in adults (41), and its correlation with other cognitive domains in adults with chronic illnesses has not been well explored. One of the strengths of the numeracy assessment used in this report is that it is brief and does not require a neuropsychologist to administer or interpret it. Previous studies have identified cognitive deficiencies in the CKD and ESRD population (18–21), and declining cognitive function has been implicated as a potential mediating factor of the lower health literacy documented among the elderly (22,23). Our study suggests a potential relationship between cognitive impairments and numeracy deficiencies in the CKD and ESRD population. The association between cognitive function and numeracy warrants further exploration to determine whether numeracy acts as one potential mediator of the poor outcomes associated with cognitive decline.
Our findings should be interpreted in light of several limitations. First, we used a brief numeracy assessment that has been widely utilized in various forms (8,9,26–28,41,42). However, this measure tests analytical and statistical numeracy including the ability to convert between percentages and proportions, a task often considered to require more advanced numerical skills. The instrument did not directly assess patients’ capacity to interpret and apply numerical health information accurately. Future studies in CKD patients should consider using an instrument that measures a more encompassing range of numerical skills to help better define barriers in healthcare provider-patient communication. Second, although the 82 patients who did not complete the study were demographically similar to study participants, their exclusion may have influenced our results. Third, we used local institutional medical records to ascertain patients’ transplant waiting list status. While the local medical center is the largest kidney transplant provider in the region, patients may have been active on waiting lists at other transplant centers. However, this misclassification should bias toward finding no relationship between numeracy and kidney transplantation. Fourth, there are additional factors that contribute to kidney transplant utilization that were not controlled for in this study (e.g., insurance status and comorbid illnesses). Although, patients with active severe comorbidities such as unstable angina were excluded from the study, we did not ascertain whether study participants were medically eligible for transplantation. Similarly, patients with poor numeracy may not have been considered for transplants due to deficits in cognitive function rather than numeracy per se. Finally, we excluded patients with severe comorbid illnesses who are likely to interact frequently with healthcare providers while making major decisions regarding care. However, we do not believe that CKD and ESRD patients with more severe comorbidities are likely to demonstrate significantly better numeracy skills than the studied sample.
Conclusion
Poor numeracy skills are common in the ESRD and advanced CKD population. While African Americans, those without a high-school education, and the unemployed may have a higher prevalence of deficient numeracy skills, all patient subgroups are substantially affected. Lower numeracy was associated with decreased kidney transplantation or active wait list status independent of age, sex, or race. Renal providers should consider using well-designed visual aids to enhance patient communication, especially when discussing important health decisions including transplantation. Further studies are necessary to reveal whether numeracy deficits may contribute to documented transplant disparities and whether well-designed patient communication aides can improve transplant access.
Disclosures
None.
Acknowledgments
We sincerely want to thank Carol Stilley and the participants and staff that have supported this project. This work was supported by Fresenius National Kidney Foundation Young Investigator Grant, Paul Teschan Research Grant, NIH DK66006 and DK77785 (Unruh); National Kidney Foundation Research Fellowship, Ruth L. Kirschstein National Research Service Award Institutional Research Training Grants T32-DK061296 and Individual Postdoctoral Training Grant F32-DK084676 (Abdel-Kader).
Footnotes
Published online ahead of print. Publication date available at www.cjasn.org.
References
- 1.Cavanaugh K, Huizinga MM, Wallston KA, Gebretsadik T, Shintani A, Davis D, Gregory RP, Fuchs L, Malone R, Cherrington A, Pignone M, DeWalt DA, Elasy TA, Rothman RL: Association of numeracy and diabetes control. Ann Intern Med 148: 737–746, 2008 [DOI] [PubMed] [Google Scholar]
- 2.Rothman RL, Housam R, Weiss H, Davis D, Gregory R, Gebretsadik T, Shintani A, Elasy TA: Patient understanding of food labels: The role of literacy and numeracy. Am J Prev Med 31: 391–398, 2006 [DOI] [PubMed] [Google Scholar]
- 3.Kutner M, Greenberg E, Baer J: A First Look at the Literacy of America’s Adults in the 21st Century, Washington, DC, National Center for Education Statistics, December2005. NCES publication number 93275 [Google Scholar]
- 4.Nelson W, Reyna VF, Fagerlin A, Lipkus I, Peters E: Clinical implications of numeracy: Theory and practice. Ann Behav Med 35: 261–274, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Osborn CY, Cavanaugh K, Wallston KA, White RO, Rothman RL: Diabetes numeracy: An overlooked factor in understanding racial disparities in glycemic control. Diabetes Care 32: 1614–1619, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Golbeck AL, Ahlers-Schmidt CR, Paschal AM, Dismuke SE: A definition and operational framework for health numeracy. Am J Prev Med 29: 375–376, 2005 [DOI] [PubMed] [Google Scholar]
- 7.Lipkus IM: Numeric, verbal, and visual formats of conveying health risks: Suggested best practices and future recommendations. Med Decis Making 27: 696–713, 2007 [DOI] [PubMed] [Google Scholar]
- 8.Woloshin S, Schwartz LM, Moncur M, Gabriel S, Tosteson AN: Assessing values for health: Numeracy matters. Med Decis Making 21: 382–390, 2001 [DOI] [PubMed] [Google Scholar]
- 9.Estrada CA, Martin-Hryniewicz M, Peek BT, Collins C, Byrd JC: Literacy and numeracy skills and anticoagulation control. Am J Med Sci 328: 88–93, 2004 [DOI] [PubMed] [Google Scholar]
- 10.Apter AJ, Cheng J, Small D, Bennett IM, Albert C, Fein DG, George M, Van Horne S: Asthma numeracy skill and health literacy. J Asthma 43: 705–710, 2006 [DOI] [PubMed] [Google Scholar]
- 11.Devraj R, Gordon EJ: Health literacy and kidney disease: Toward a new line of research. Am J Kidney Dis 53: 884–889, 2009 [DOI] [PubMed] [Google Scholar]
- 12.Grubbs V, Gregorich SE, Perez-Stable EJ, Hsu CY: Health literacy and access to kidney transplantation. Clin J Am Soc Nephrol 4: 195–200, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Jindal RM, Ryan JJ, Sajjad I, Murthy MH, Baines LS: Kidney transplantation and gender disparity. Am J Nephrol 25: 474–483, 2005 [DOI] [PubMed] [Google Scholar]
- 14.Navaneethan SD, Singh S: A systematic review of barriers in access to renal transplantation among African Americans in the United States. Clin Transplant 20: 769–775, 2006 [DOI] [PubMed] [Google Scholar]
- 15.Patzer RE, Amaral S, Wasse H, Volkova N, Kleinbaum D, McClellan WM: Neighborhood poverty and racial disparities in kidney transplant waitlisting. J Am Soc Nephrol 20: 1333–1340, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Alexander GC, Sehgal AR: Barriers to cadaveric renal transplantation among blacks, women, and the poor. JAMA 280: 1148–1152, 1998 [DOI] [PubMed] [Google Scholar]
- 17.Chisholm MA, Fair J, Spivey CA: Health literacy and transplant patients and practitioners. Public Health 121: 800–803, 2007 [DOI] [PubMed] [Google Scholar]
- 18.Kurella M, Chertow GM, Fried LF, Cummings SR, Harris T, Simonsick E, Satterfield S, Ayonayon H, Yaffe K: Chronic kidney disease and cognitive impairment in the elderly: The health, aging, and body composition study. J Am Soc Nephrol 16: 2127–2133, 2005 [DOI] [PubMed] [Google Scholar]
- 19.Kurella M, Chertow GM, Luan J, Yaffe K: Cognitive impairment in chronic kidney disease. J Am Geriatr Soc 52: 1863–1869, 2004 [DOI] [PubMed] [Google Scholar]
- 20.Seliger SL, Siscovick DS, Stehman-Breen CO, Gillen DL, Fitzpatrick A, Bleyer A, Kuller LH: Moderate renal impairment and risk of dementia among older adults: The Cardiovascular Health Cognition Study. J Am Soc Nephrol 15: 1904–1911, 2004 [DOI] [PubMed] [Google Scholar]
- 21.Slinin Y, Paudel ML, Ishani A, Taylor BC, Yaffe K, Murray AM, Fink HA, Orwoll ES, Cummings SR, Barrett-Connor E, Jassal S, Ensrud KE: Kidney function and cognitive performance and decline in older men. J Am Geriatr Soc 56: 2082–2088, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Levinthal BR, Morrow DG, Tu W, Wu J, Murray MD: Cognition and health literacy in patients with hypertension. J Gen Intern Med 23: 1172–1176, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Morrow D, Clark D, Tu W, Wu J, Weiner M, Steinley D, Murray MD: Correlates of health literacy in patients with chronic heart failure. Gerontologist 46: 669–676, 2006 [DOI] [PubMed] [Google Scholar]
- 24.Abdel-Kader K, Myaskovsky L, Karpov I, Shah J, Hess R, Dew MA, Unruh M: Individual quality of life in chronic kidney disease: influence of age and dialysis modality. Clin J Am Soc Nephrol 4: 711–718, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Abdel-Kader K, Unruh ML, Weisbord SD: Symptom burden, depression, and quality of life in chronic and end-stage kidney disease. Clin J Am Soc Nephrol 4: 1057–1064, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Schwartz LM, Woloshin S, Black WC, Welch HG: The role of numeracy in understanding the benefit of screening mammography. Ann Intern Med 127: 966–972, 1997 [DOI] [PubMed] [Google Scholar]
- 27.Lipkus IM, Samsa G, Rimer BK: General performance on a numeracy scale among highly educated samples. Med Decis Making 21: 37–44, 2001 [DOI] [PubMed] [Google Scholar]
- 28.Sheridan SL, Pignone MP, Lewis CL: A randomized comparison of patients’ understanding of number needed to treat and other common risk reduction formats. J Gen Intern Med 18: 884–892, 2003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Lezak MD, Lezak MD: Neuropsychological Assessment, Oxford; New York, Oxford University Press, 2004 [Google Scholar]
- 30.Teng EL, Chui HC: The Modified Mini-Mental State (3MS) examination. J Clin Psychiatry 48: 314–318, 1987 [PubMed] [Google Scholar]
- 31.Psychological Corporation Wechsler Abbreviated Scale of Intelligence (WASI) manual. San Antonio, Harcourt Brace, 1999 [Google Scholar]
- 32.Benton A, Hamsher K: Multilingual Aphasia Examination, Iowa City, University of Iowa Press, 1976 [Google Scholar]
- 33.Royall DR, Chiodo LK, Polk MJ: Correlates of disability among elderly retirees with “subclinical” cognitive impairment. J Gerontol A Biol Sci Med Sci 55: M541–M546, 2000 [DOI] [PubMed] [Google Scholar]
- 34.Royall DR, Cordes JA, Polk M: CLOX: An executive clock drawing task. J Neurol Neurosurg Psychiatry 64: 588–594, 1998 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Royall DR, Mulroy AR, Chiodo LK, Polk MJ: Clock drawing is sensitive to executive control: A comparison of six methods. J Gerontol B Psychol Sci Soc Sci 54: P328–P333, 1999 [DOI] [PubMed] [Google Scholar]
- 36.Ginde AA, Clark S, Goldstein JN, Camargo CA, Jr: Demographic disparities in numeracy among emergency department patients: Evidence from two multicenter studies. Patient Educ Couns 72: 350–356, 2008 [DOI] [PubMed] [Google Scholar]
- 37.Huizinga MM, Beech BM, Cavanaugh KL, Elasy TA, Rothman RL: Low numeracy skills are associated with higher BMI. Obesity (Silver Spring) 16: 1966–1968, 2008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Huizinga MM, Carlisle AJ, Cavanaugh KL, Davis DL, Gregory RP, Schlundt DG, Rothman RL: Literacy, numeracy, and portion-size estimation skills. Am J Prev Med 36: 324–328, 2009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.National Center for Education Statistics Highlights From the 2003 International Adult Literacy and Lifeskills Survey (ALL). Available at: http://nces.ed.gov/pubs2005/2005117.pdf Accessed February 15, 2010
- 40.Ancker JS, Kaufman D: Rethinking health numeracy: A multidisciplinary literature review. J Am Med Inform Assoc 14: 713–721, 2007 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Peters E, Vastfjall D, Slovic P, Mertz CK, Mazzocco K, Dickert S: Numeracy and decision making. Psychol Sci 17: 407–413, 2006 [DOI] [PubMed] [Google Scholar]
- 42.Schwartz LM, Woloshin S, Welch HG: Can patients interpret health information? An assessment of the medical data interpretation test. Med Decis Making 25: 290–300, 2005 [DOI] [PubMed] [Google Scholar]