## Abstract

In developing countries, education levels vary dramatically, and the number of years of schooling does not always correlate with the true level of educational competency. This study was designed to verify the accuracy of the Short-Test of Functional Health Literacy in Adults (S-TOFHLA) in individuals with mild cognitive impairment (MCI) and mild Alzheimer's disease (AD), when compared with healthy controls (HCs), in order to assess its utility as a measure of functional literacy. One hundred forty-eight subjects were divided into three groups: HC (n = 61), MCI patients (n = 42), and AD patients (n = 45). The S-TOFHLA does not seem to be suitable as an instrument to measure functional literacy for patients with advanced cognitive impairment, but proved to be appropriate in both the HC group and MCI patients in numeracy and prove to be useful as an adjuvant to estimate IQ, reading ability, and premorbid IQ, as an indicator of cognitive reserve.

## Introduction

Different levels of schooling are common among the elderly population, particularly in developing countries. Illiteracy is recognized as a world-wide issue and, according to United Nations (UN), the prevalence of illiteracy in adults around the world is around 17%; a number which increases when considering the number of functional illiterates (Unesco, 2011). Results of the National Adult Literacy Survey (NALS) showed that 21%–23% of American adults were at the lowest level of literacy, and one-third of these individuals were over the age of 65 (Kirsch, Jungeblut, Jenkins, & Kolstad, 2010). In Brazil, approximately 42.6% of the elderly population is considered functionally illiterate (defined as having less than 4 years of schooling; IBGE, 2010).

Functional literacy can be characterized as possessing a set of capabilities that allow the individual to acquire and use new information, be it through reading, writing, comprehension, or interpretation (Williams et al., 1995). Literacy can be improved through educational programs or can diminish when age and cognitive diseases become factors (Baker, 2006; Baker, Gazmararian, Sudano, & Patterson, 2000).

The S-TOFHLA proved to be an adequate tool to measure levels of functional health literacy in Brazilian individuals with normal cognitive function, correlating strongly with the number of years of formal education (Carthery-Goulart et al., 2009). In this sample, 23.4% were classified as inadequate (scoring between zero and 53 points), 9% as marginal range (scoring between 54 and 66 points), and 67.6% as adequate (scoring between 67 and 100 points). This correlated with the level of schooling, but even more strongly to the Mini-Mental State Examination (MMSE), suggesting that it could be used as an adjunct in confirming accurate scores on a cognitive test (Brucki, Mansur, Carthery-Goulart, & Nitrini, 2011).

In a longitudinal study, it was observed that patients with Alzheimer's disease (AD) maintained stable NART scores during the course of the disease and which was a good predictor of intellectual function (Paque & Warrington, 1995). Other studies showed minimal decline, with relatively preserved reading skills into the advanced stages of disease (Fromm, Holland, Nebes, & Oakley, 1991; Grober et al., 2008; Patterson, Graham, & Hodges, 1994). Tests of evaluation of premorbid intelligence quotient (IQ) based on irregular word reading and lexical decision showed no difference in outcomes between patients with subjective memory complaints, mild cognitive impairment (MCI), AD, and controls (Almkvist & Tallberg, 2009). Information about health literacy along the spectrum of cognitive impairment is less frequent in the literature.

Neuropsychological tests are markedly influenced by schooling, and in almost all cognitive areas, illiterate individuals and those with little education have lower scores than those with moderate to high educational experience (Ardila, Rosselli, & Rosas, 1989; Bertolucci, Okamoto, Neto, Ramos, & Brucki, 1998; Brucki & Rocha, 2004; Manly, Schupf, Tang, & Stern, 2005; Porto, Caramelli, & Nitrini, 2010; Rosselli, 1993; Yassuda et al., 2009).

### The Brazilian classification

In Brazil, education is classified into: Elementary school (1st grade to 8th grade), middle school (9th grade to 11th grade), and higher education (+12 years of education). The main problem when using the number of years of formal education as a factor is that, in countries where schools have different programs, number of hours. and efficacy of learning, we are faced with different performance among older adults of the same educational level. Thus, a measure of the degree of literacy becomes urgent, as an adjunct to the educational level in the assessment of neuropsychological tests.

Lower health literacy is associated with increased mortality and disability, generating higher costs. The low education, being part of cognitive decline even among non-demented people, has clear implications for public policy. Materials related to health should be redesigned to make them more understandable and accessible, especially for people with cognitive impairment. It is important to reduce the cognitive demands to the materials of health, peculiarly those that require high capacities of reasoning, memory, and with regard to health decisions and financial situation (Boyle et al., 2013).

There are no studies using the S-TOFHLA as a measure of literacy in dementia and it is a relatively simple test to adopt and a more useful measure of literacy than the number of years of formal schooling. This study aimed to verify the performance on Brazilian Portuguese version (Carthery-Goulart et al., 2009) of the S-TOFHLA in individuals with MCI and mild AD compared with healthy controls (HCs) in order to verify if it still can be used as a measure of functional literacy in patients.

## Method

### Participants

The sample consisted of 148 subjects who were divided into three groups: Health controls (HC) (n = 61), MCI patients (n = 42), and AD patients (n = 45; Table 1).

Table 1.

Demographic characteristics of HC, MCI, and AD groups

HC (n = 61)

MCI (n = 42)

p-value
Mean (SDMn Mx Mean (SDMn Mx Mean (SDMn Mx
Gender (F/M) 35/26   27/15   26/19   .75a
Age 70.66 (6.55)* 60 91 72.64 (7.71) 61 89 75.80 (4.81)* 65 86 <.01b
Education 8.72 (5.44) 21 7.14 (5.11) 20 6.96 (4.57) 19 .13 b
HC (n = 61)

MCI (n = 42)

p-value
Mean (SDMn Mx Mean (SDMn Mx Mean (SDMn Mx
Gender (F/M) 35/26   27/15   26/19   .75a
Age 70.66 (6.55)* 60 91 72.64 (7.71) 61 89 75.80 (4.81)* 65 86 <.01b
Education 8.72 (5.44) 21 7.14 (5.11) 20 6.96 (4.57) 19 .13 b

Notes: F = Female; M = Male; SD = standard deviation; HC = health control; MCI = mild cognitive impairment; AD = Alzheimer's disease; Mn = Minimum; Mx = Maximum.

aChi-square test.

bKruskal–Wallis test.

*Significant difference between control and AD groups (p < .005).

Significant difference between MCI and AD groups (p < .005).

HC subjects were healthy volunteers, recruited from the general population in São Paulo, Brazil. They had no history of neurological or psychiatric disease and showed no evidence of cognitive impairment.

Subjects with MCI and AD were recruited from the Cognitive Neurology Outpatient Clinic from de Hospital das Clínicas of the University of São Paulo School of Medicine and the Cognitive Neurology Outpatient Clinic from Santa Marcelina Hospital in São Paulo, Brazil.

The following inclusion criteria were used for the entire cohort: More than 1 year of education (formal or informal), at least 60 years of age, presence of an informant, and absence of depressive symptoms (<6 points, according to the 15-item Geriatric Depression Scale [15-GDS; Almeida & Almeida, 1999]). AD patients were included based on diagnosis of probable AD (Mckhann et al., 1984), with stable doses of cholinesterase inhibitors, absence of moderate or severe dementia, absence of dementia caused by other etiologies, absence of psychiatric disorders, and a Cornell Scale for Depression in Dementia score lower than 7 (Carthery-Goulart et al., 2007). For MCI patients, a score lower than 5 on the Functional Activities Questionnaire (Pfeffer et al., 1982) and a diagnosis of MCI (as proposed by Petersen et al., 2001) were used as inclusion criteria.

All patients were evaluated by members of the Behavioral and Cognitive Neurology Unit of the Department of Neurology at the University of São Paulo School of Medicine, from the Center of Cognitive Disorders (CEREDIC) and Santa Marcelina Hospital. All patients performed a comprehensive neuropsychological, functional evaluation (Functional Activities Questionnaire) and Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE; Bustamante et al., 2003).

HC groups were required to fulfill these criteria: The absence of cognitive decline according to education-adjusted cutoff scores of the MMSE (Brucki, Nitrini, Caramelli, Bertolucci, & Okamoto, 2003) and a score lower than 2 on the Functional Activities Questionnaire.

Subjects completed a structured interview, lasting approximately 1.5 h. The study was approved by Ethics Committee of Hospital das Clínicas from the University of São Paulo School of Medicine and Santa Marcelina Hospital. All subjects who agreed to participate signed written informed consent.

### Design

Subjects completed a structured interview with a neuropsychologist that included basic demographic information, socioeconomic status, and co-morbidity. They were then given health literacy evaluations and a cognitive battery to measure global status, crystallized and fluid intelligence, memory, visuoconstructive tasks, and estimated IQ.

## Material

We obtained permission from the author for use and application of the S-TOFHLA.

### Health Literacy

Health literacy was assessed by the S-TOFHLA (Baker et al., 1999), which includes both reading comprehension and numeracy competency sections. The reading comprehension text is comprised of two passages, with a total of 36 items. Each passage has every fifth or sixth word deleted and, for each blank space, the respondent must select the word that best completes the sentence from four given options. The total score of the reading comprehension section is 72 points. The numeracy test evaluates qualitative literacy needed in the healthcare setting. It comprises two medicine bottles and two cards containing information about medicine intake, date of appointments, and the result of laboratorial test. The numeric items total 28 points. The total score of the test is 100 points. Individuals scoring between zero and 53 points are considered in the inadequate range, between 54 and 66 points puts them in the marginal range, and those that score between 67 and 100 are considered to be adequate in health literacy.

### Cognitive Evaluation

All subjects were given the Brief Cognitive Screening Battery (BCSB; Nitrini et al., 2004, 2007) (using learning and recall of drawings), Dementia Rating Scale (DRS; Mattis, 1988; Porto, Fichman, Caramelli, Bahia, & Nitrini, 2003), Rey Auditory Verbal Learning Test (RAVLT; Malloy-Dizniz et al., 2007; Spreen, Strauss, & Sherman, 2006), Raven's Colored Matrices (Angelini, Alves, Custódio, Duarte, & Duarte, 1999), Clock Drawing (Sunderland et al., 1989), and Verbal Fluency (Spreen et al., 2006).

### Estimated IQ

Estimated IQ was calculated by adding the weighted scores of subtests, block design, and vocabulary on the Wechsler Adult Intelligence Scale-III (Ringe, Saine, Lacritz, Hynan, & Cullum, 2002).

## Analysis Plan

For correlations between education, age, the S-TOFHLA, and the neuropsychological test, we used Spearman's correlation. R-values of >.75 indicate strong correlation, a value between .75 and .5 indicates average correlation, and <.5 indicates a weak or absent correlation. The significance level used was .05.

Factors were compared using the Kruskal–Wallis test, followed by non-parametric multiple comparisons, when necessary. Nominal and ordinal qualitative measures were described in groups with absolute and relative frequencies. We utilized χ2 to determine association between measures and nominal groups, and ordinal measures were compared with the Kruskal–Wallis test, followed by non-parametric multiple comparisons. All statistical analysis was carried out using the Statistical Package for the Social Sciences (SPSS) program, version 15.0.

## Results

The sample is described in Table 1. There were no statistically significant differences in schooling (p = .13) or gender (p = .75) between groups. However, a statistical difference was observed in age (p < .001). Multiple comparisons showed that the difference in age between the groups is between AD and HC groups (p < .001) and between MCI and AD groups (p = .010).

In S-TOFHLA, non-parametric multiple comparisons indicated that a statistically significant difference in reading comprehension and the overall score occurred among the three groups (HC, MCI, and AD). In numeracy, the statistical difference occurred only between the AD and HC groups (Table 3).

Table 2.

Performance of HC, MCI, and AD groups on the testsa

Tests Mean (SD)

p-value
MMSE 28.36 (1.48)* 25.98 (2.68)* 23.96 (2.90)†‡ <.01
DRS-Total 136.49 (8.12)* 127.90(10.46)* 117.00 (11.64)†‡ <.01
BCSB-Incidental memory 5.84 (1.36)* 4.57 (1.67)* 3.51 (1.77)†‡ <.01
BCSB-Immediate recall 7.84 (1.19)* 6.88 (1.63)* 5.38 (1.70)†‡ <.01
BCSB-Learning 8.59 (1.20)* 7.81 (1.45)* 5.76 (1.76)†‡ <.01
BCSB-Delayed recall 7.92 (1.44)* 6.20 (2.39)* 3.09 (2.63)†‡ <.01
BCSB-Recognition 9.80 (0.44)* 9.24 (1.12)* 7.64 (3.23)†‡ <.01
RAVLT-Total 38.34 (8.68)* 31.17 (9.70)* 22.91 (6.42)†‡ <.01
RAVLT-30′ 6.49 (3.54)* 4.14 (3.39)* 0.82 (1.40)†‡ <.01
Verbal Fluency (animals) 16.38 (4.96)* 13.76 (4.33)* 10.60 (3.39)†‡ <.01
Clock Drawing 8.33 (2.07)* 7.20 (2.43)* 6.49 (2.83) <.01
Vocabulary 26.46 (9.81) 23.51 (9.56) 23.20 (9.16) .20
Block Design 25.43 (8.93)* 18.37 (8.42)* 14.18 (6.53) <.01
Raven's Colored Matrices 25.84 (6.24)* 20.52 (6.35)* 17.76 (5.99) <.01
Estimated IQ 98.77 (10.41)* 91.45 (9.92)* 87.95 (8.92)†‡ <.01
Tests Mean (SD)

p-value
MMSE 28.36 (1.48)* 25.98 (2.68)* 23.96 (2.90)†‡ <.01
DRS-Total 136.49 (8.12)* 127.90(10.46)* 117.00 (11.64)†‡ <.01
BCSB-Incidental memory 5.84 (1.36)* 4.57 (1.67)* 3.51 (1.77)†‡ <.01
BCSB-Immediate recall 7.84 (1.19)* 6.88 (1.63)* 5.38 (1.70)†‡ <.01
BCSB-Learning 8.59 (1.20)* 7.81 (1.45)* 5.76 (1.76)†‡ <.01
BCSB-Delayed recall 7.92 (1.44)* 6.20 (2.39)* 3.09 (2.63)†‡ <.01
BCSB-Recognition 9.80 (0.44)* 9.24 (1.12)* 7.64 (3.23)†‡ <.01
RAVLT-Total 38.34 (8.68)* 31.17 (9.70)* 22.91 (6.42)†‡ <.01
RAVLT-30′ 6.49 (3.54)* 4.14 (3.39)* 0.82 (1.40)†‡ <.01
Verbal Fluency (animals) 16.38 (4.96)* 13.76 (4.33)* 10.60 (3.39)†‡ <.01
Clock Drawing 8.33 (2.07)* 7.20 (2.43)* 6.49 (2.83) <.01
Vocabulary 26.46 (9.81) 23.51 (9.56) 23.20 (9.16) .20
Block Design 25.43 (8.93)* 18.37 (8.42)* 14.18 (6.53) <.01
Raven's Colored Matrices 25.84 (6.24)* 20.52 (6.35)* 17.76 (5.99) <.01
Estimated IQ 98.77 (10.41)* 91.45 (9.92)* 87.95 (8.92)†‡ <.01

Notes: SD = standard deviation; MCI = mild cognitive impairment; HC = health control; AD = Alzheimer's disease; MMSE = Mini-Mental State Exam; BCSB = Brief Cognitive Screening Battery; DRS = Dementia Rating Scale; RAVLT = Rey auditory Verbal Learning Test; IQ = intelligence quotient.

aKruskall–Wallis test.

*Significant difference between control and MCI groups (p < .005).

Significant difference between MCI and AD groups (p < .005).

Significant difference between control and AD groups (p < .005).

Table 3.

Performance of HC, MCI, and AD groups on the S-TOFHLAa

Tests Mean (SD)

p-value
S-TOFHLA Reading 46.79 (21.66)* 36.62 (19.66)* 20.58 (13.58)†‡ <.01
S-TOFHLA Numeracy 22.72 (6.47) 20.33 (7.22) 19.60 (7.25) .04
S-TOFHLA Total 69.41 (25.88)* 56.95 (24.47)* 40.18 (16.99)†‡ <.01
S-TOFHLA Time 365.85 (201.07) 451.00 (195.06) 361.02 (190.03) .08
Tests Mean (SD)

p-value
S-TOFHLA Reading 46.79 (21.66)* 36.62 (19.66)* 20.58 (13.58)†‡ <.01
S-TOFHLA Numeracy 22.72 (6.47) 20.33 (7.22) 19.60 (7.25) .04
S-TOFHLA Total 69.41 (25.88)* 56.95 (24.47)* 40.18 (16.99)†‡ <.01
S-TOFHLA Time 365.85 (201.07) 451.00 (195.06) 361.02 (190.03) .08

Notes: SD = standard deviation; MCI = mild cognitive impairment; HC = health control; AD = Alzheimer's disease; S-TOFHLA = Short Test of Functional Health Literacy in Adults.

aKruskall–Wallis test.

*Significant difference between control and MCI groups (p < .005).

Significant difference between MCI and AD groups (p < .005).

Significant difference between control and AD groups (p < .005).

Table 4 is showing the performance of the groups divided by education in the S-TOFHLA and demonstrates that there is an increase in the value of the S-TOFHLA with increasing of years of education.

Table 4.

Performance of HC, MCI, and AD groups on the S-TOFHLA divided by years of schooling

Schooling (years) Mean (SD)

HC

MCI

S-TOFHLA Reading S-TOFHLA Numeracy S-TOFHLA Total S-TOFHLA Reading S-TOFHLA Numeracy S-TOFHLA Total S-TOFHLA Reading S-TOFHLA Numeracy S-TOFHLA Total
≤4 32.58 (18.43) 19.25 (7.51) 51.83 (23.56) 22.45 (11,01) 16.55 (7.43) 39 (16.51) 15.13 (11.24) 19.25 (8.57) 34.38 (16.22)
5–8 40.67 (21.68) 23.92 (3.60) 64.58 (24.36) 49.14 (15.40) 25 (3.74) 74.14 (13.06) 24.25 (14.20) 21 (5.29) 45.25 (19.02)
≥9 63.12 (10.91) 25.48 (4.90) 88.60 (12.90) 53.85 (14.46) 24.23 (4.62) 78.08 (15.36) 28.38 (13.48) 20.46 (6.04) 48.85 (13.20)
Schooling (years) Mean (SD)

HC

MCI

S-TOFHLA Reading S-TOFHLA Numeracy S-TOFHLA Total S-TOFHLA Reading S-TOFHLA Numeracy S-TOFHLA Total S-TOFHLA Reading S-TOFHLA Numeracy S-TOFHLA Total
≤4 32.58 (18.43) 19.25 (7.51) 51.83 (23.56) 22.45 (11,01) 16.55 (7.43) 39 (16.51) 15.13 (11.24) 19.25 (8.57) 34.38 (16.22)
5–8 40.67 (21.68) 23.92 (3.60) 64.58 (24.36) 49.14 (15.40) 25 (3.74) 74.14 (13.06) 24.25 (14.20) 21 (5.29) 45.25 (19.02)
≥9 63.12 (10.91) 25.48 (4.90) 88.60 (12.90) 53.85 (14.46) 24.23 (4.62) 78.08 (15.36) 28.38 (13.48) 20.46 (6.04) 48.85 (13.20)

Notes: SD = standard deviation; MCI = mild cognitive impairment; HC = health control; AD = Alzheimer's disease; S-TOFHLA = Short Test of Functional Health Literacy in Adults;

The S-TOFHLA scores did not correlate with age in either group, showing that age does not influence the outcome. Correlations were observed between the neuropsychological tests, demographics, and scores on the S-TOFHLA as shown in Table 5.

Table 5.

Spearman's correlations among S-TOFHLA scores, schooling, age, and neuropsychological tests

Age Schooling S-TOFHLA R

MMSE HC .065 .506* .509* .472* .545* −.278*
MMSE MCI −.117 .419* .543* .358* .506* −.317*
MMSE AD .165 .327* .278 .176 .296* −.120
RCM HC −.026 .614* .663* .437* .663* −347*
RCM MCI −.133 .700* .681* .534* .666* −.374*
RCM AD −.219 .389* .373* .172 .416* −.187
Block Design HC −.122 .529* .606* .456* .598* −.553*
Block Design MCI −.153 .667* .707* .453* .672* −.249
Block Design AD −.200 .401* .561* .367* .630* −.156
Vocabulary HC .044 .616* .706* .618* .732* −.578*
Vocabulary MCI .222 .615* .588* .607* .640* −.381*
Vocabulary AD .089 .626* .490* .315* .569* −.169
Estimated IQ HC .035 .724* .741* .638* .758* −.545*
Estimated IQ MCI .084 .803* .797* .599* .793* −.389*
Estimated IQ AD −.115 .529* .535* .291 .585* .046
Age HC   −.082 −.087 −.072 .023
Age MCI   −.142 −116 −.156 .053
Age AD   −.076 −.148 −.110 .011
Schooling HC   .634* .477* −.401* .612*
Schooling MCI   .745* .536* .748* −.371
Schooling AD   −.084 .196 −.026 .341*
Age Schooling S-TOFHLA R

MMSE HC .065 .506* .509* .472* .545* −.278*
MMSE MCI −.117 .419* .543* .358* .506* −.317*
MMSE AD .165 .327* .278 .176 .296* −.120
RCM HC −.026 .614* .663* .437* .663* −347*
RCM MCI −.133 .700* .681* .534* .666* −.374*
RCM AD −.219 .389* .373* .172 .416* −.187
Block Design HC −.122 .529* .606* .456* .598* −.553*
Block Design MCI −.153 .667* .707* .453* .672* −.249
Block Design AD −.200 .401* .561* .367* .630* −.156
Vocabulary HC .044 .616* .706* .618* .732* −.578*
Vocabulary MCI .222 .615* .588* .607* .640* −.381*
Vocabulary AD .089 .626* .490* .315* .569* −.169
Estimated IQ HC .035 .724* .741* .638* .758* −.545*
Estimated IQ MCI .084 .803* .797* .599* .793* −.389*
Estimated IQ AD −.115 .529* .535* .291 .585* .046
Age HC   −.082 −.087 −.072 .023
Age MCI   −.142 −116 −.156 .053
Age AD   −.076 −.148 −.110 .011
Schooling HC   .634* .477* −.401* .612*
Schooling MCI   .745* .536* .748* −.371
Schooling AD   −.084 .196 −.026 .341*

Notes: R = coefficient of correlation; HC = Health Control; MCI = mild cognitive impairment; AD = Alzheimer's disease; S-TOFHLA = Short Test of Functional Health Literacy in Adults; MMSE = Mini-Mental State Exam; RCM = Raven's Colored Matrices; IQ = intelligence quotient.

*p < .05.

Fig. 1 shows the performance of the groups (control, MCI, and AD) in relation to the total score of the S-TOFHLA, the reading comprehension and numeracy items, showing decreased performance with disease progression in total score and reading comprehension. The numeracy score were less affected by disease severity than the other measures. The only difference was between HC and AD subjects.

Fig. 1.

Group performance (Control, MCI, and AD) in relation to the total score of the S-TOFHLA, Reading Comprehension, and Numeracy tests.

Fig. 1.

Group performance (Control, MCI, and AD) in relation to the total score of the S-TOFHLA, Reading Comprehension, and Numeracy tests.

## Discussion

The goal of the present study was to verify the performance on the S-TOFHLA in individuals with MCI and mild AD compared with HCs in order to verify if it still can be used as a measure of functional literacy in patients impaired and according to the results, the S-TOFHLA was not helpful as an instrument measuring literacy function in patients with AD. However, it proved to be adequated in both the control group and MCI patients in numeracy, showing the functionality of individuals with MCI.

A study that assessed healthy numeracy among older adults found that whereas scores on reading comprehension decreased with older age, age did not differentiate ability on numeracy, and it may reflect an emphasis on basic number recognition instead of using strategies and skills associated with the decline in age (Donelle, Hoffman-Goetz, & Arocha, 2007). The numeracy item correlates with schooling and does not seem to be affected by aging.

Our results showed that demographic factors, such as gender and age, did not influence the performance of the S-TOFHLA or the neuropsychological tests within groups. Even with the factor of age showing statistically significant differences between the groups, there were no significant correlations of age with the S-TOFHLA or neuropsychological tests, demonstrating that the tests are not affected by this variable.

There was a statistically significant correlation between the S-TOFHLA total and MMSE for the control group and the group diagnosed with MCI. In this study, the association between S-TOFHLA and MMSE for the control group showed a weaker correlation compared with a previous study with a sample of healthy subjects; this sample, however, was composed of young adults, adults and seniors, and the median age was 45 years with 11 years of schooling (Brucki et al., 2011).

Another more recent study by Chin and colleagues (2011) investigated measures that were associated with tools used to predict behaviors and health outcomes, such as the S-TOFHLA and REALM. The results showed that older adults who had higher levels of processing capacity or knowledge performed better on both measures (S-TOHFLA and REALM). Wolf and colleagues (2012) studied the degree to which cognitive abilities explain the association between health education, performance on common tasks relating to health and functional health status. In this study, conducted with adults between 55 and 74 years, measures of health literacy (REALM, TOFHLA, and Newest Vital Sign [NVS]) were strongly correlated with cognitive skills, both fluid and crystallized. Low functional literacy and poor fluid and crystallized cognitive function were both associated with poorer performance on health tasks. Associations between health literacy and performance of common tasks were explained by lack of background knowledge and the need to be able to apply new information to build a sound foundation of understanding, linking fluid and crystalized processes (Weiss et al., 2005).

In our study, the S-TOFHLA showed correlation with tests that assess both skills (fluid and crystallized). The S-TOFHLA was also shown to be correlated with estimated IQ, which is a sum of the measure of both types of intelligence, for both the control group and patients diagnosed with MCI.

Intellectual decline is the hallmark of dementia and AD, and a good measure of premorbid intelligence in relation to current abilities, which can be useful in assessing the magnitude of the decline (Dierckx et al., 2008; Powell, Brossart, & Reynolds, 2003).

In our study, the S-TOFHLA showed strong and significant correlation with estimated IQ. The estimated IQ is obtained through two measures: Vocabulary as a measure of crystallized intelligence and cubes as measures of fluid intelligence. S-TOFHLA could be used as a measure of premorbid level in health and MCI individuals.

Premorbid IQ is an index of cognitive reserve. Starr and Lonie (2008) worked with the hypothesis that cognitive reserve continues to influence cognition after AD diagnosis, having an indirect impact on activities of daily living.

The S-TOFHLA does not seem to be suitable as an instrument to measure functional literacy for patients with more advanced cognitive impairment (mild dementia), but proved to be useful as an adjuvant to estimated IQ, reading ability, and premorbid IQ, when used as an indicator of cognitive reserve.

Our findings suggested that S-TOFHLA could be used as a measure of relative independency for MCI controlling their own medications. There was no difference in performance between controls and MCI patients in the numeracy items that are related to schedule and control of medications. Individuals with low numeracy are less able to understand the health risks and how to use the prescribed medication, and the numerical ability seems to matter at judgments and decisions in important ways (Peters et al., 2006). According to this study, MCI patients seem to be preserved in this item and may be considered independent for these activities.

General limitations of our findings are that S-TOFHLA has not been validated in the Brazilian population; therefore, we have no parameters to affirm the level of functional health literacy, using the labels of adequate, borderline, and inadequate. We also included more fluid than crystallized intelligence tests. However, we do not have instruments to measure crystallized intelligence that have been adapted for the Brazilian population.

## Funding

This study received financial support from Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) (2009/06475-5).

None declared.

## Acknowledgements

We would like to thank Dr Fábio Porto, who assisted us in statistical analysis.

## References

Almeida
O. P.
Almeida
S. A.
Confiabilidade da versão brasileira da Escala de Depressão em Geriatria (GDS) versão reduzida
Arquivos de Neuro-Psiquiatria
,
1999
, vol.
57
(pg.
421
-
426
)
Almkvist
O.
Tallberg
I. M.
Cognitive decline from estimated premorbid status predicts neurodegeneration in Alzheimer's disease
Neuropsychology
,
2009
, vol.
23
(pg.
117
-
124
)
Angelini
A. L.
Alves
I. C. B.
Custódio
E. M.
Duarte
W. F.
Duarte
J. L. M.
Manual Matrizes Progressivas Coloridas de Raven: escala especial
,
1999
São Paulo
Centro Editor de Testes e Pesquisas em Psicologia
Apolinario
D.
Braga
R. C. O. P.
Magaldi
R. M.
Busse
A. L.
Campora
F.
Brucki
S.
, et al.  .
Short assessment of health literacy for Portuguese-speaking
Revista de Saúde Pública
,
2012
, vol.
4
(pg.
702
-
711
)
Ardila
A.
Rosselli
M.
Rosas
P.
Neuropsychological assessment of illiterates. Visuospatial and memory abilities
Brain Cognition
,
1989
, vol.
11
(pg.
147
-
166
)
Baker
D. W.
The meaning and the measure of health literacy
Journal of General Internal Medicine
,
2006
, vol.
21
(pg.
878
-
883
)
Baker
D. W.
Gazmararian
J. A.
Sudano
J.
Patterson
M.
The association between age and health literacy among elderly persons
The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences
,
2000
, vol.
55
(pg.
S368
-
S374
)
Baker
D. W.
Williams
M. V.
Parker
R. M.
Gazmararian
J. A.
J.
Development of a brief test to measure functional health literacy
Patient Education and Counseling
,
1999
, vol.
38
(pg.
33
-
42
)
Bertolucci
P. H. B.
Okamoto
I. H.
Neto
J. T.
Ramos
L. R.
Brucki
S. M. D.
Desempenho da população brasileira na bateria neuropsicológica do Consortium to Establish a Registry for Alzheimer's Disease (CERAD)
Revista de Psiquiatria Clínica
,
1998
, vol.
25
(pg.
80
-
83
)
Boyle
P. A.
Yu
L.
Wilson
R. S.
Segawa
E.
Buchman
A. S.
Bennett
D. A.
Cognitive decline impairs financial and health literacy among community-based older persons without dementia
Psychology and Aging
,
2013
, vol.
28

3
(pg.
614
-
624
)
Brucki
S. M. D.
Mansur
L. L.
Carthery-Goulart
M. T.
Nitrini
R.
Formal education, health literacy and Mini-Mental State Examination
Dementia and Neuropsychologia
,
2011
, vol.
5
(pg.
26
-
30
)
Brucki
S. M. D.
Nitrini
R.
Caramelli
P.
Bertolucci
P. H. F.
Okamoto
I. H.
Sugestões para o uso do Mini-Exame do Estado Mental no Brasil
Arquivos de Neuropsiquiatria
,
2003
, vol.
61

3-B
(pg.
777
-
781
)
Brucki
S. M. D.
Rocha
M. S. G.
Category fluency test: Effects of age, gender and education on total scores, clustering and switching in Brazilian Portuguese-speaking subjects
Brazilian Journal of Medicine Biological Research
,
2004
, vol.
37
(pg.
1771
-
1777
)
Bustamante
S. E. Z.
Bottino
C. M. C.
Lopes
M. A.
Azevedo
D.
Hototian
S. R.
Litvoc
J.
, et al.  .
Arquivos de Neuropsiquiatria
,
2003
, vol.
61

3-A
(pg.
601
-
606
)
Carthery-Goulart
M. T.
Anghinah
R.
Areza-Fegyveres
R.
Bahia
V.
Brucki
S.
Damin
A.
, et al.  .
Perfomance of a Brazilian population on the test of functional health literacy in adults
Revista de Saúde Pública
,
2009
, vol.
43
(pg.
631
-
638
)
Carthery-Goulart
M. T.
Areza-Fegyveres
R.
Schultz
R. R.
Okamoto
I.
Caramelli
P.
Bertolucci
P. H. F.
, et al.  .
Versão brasileira da escala Cornell de Depressão em Demência (Cornell Depression Scale in Dementia)
Arquivos de Neuro-Psiquiatria
,
2007
, vol.
65

3-B
(pg.
912
-
915
)
Chin
J.
Morrow
D. G.
Stine-Morrow
E. A. L.
Conner-Garcia
T.
Graumlich
J. F.
Murray
M. D.
The Process-Knowledge Model of Health Literacy: Evidence from a componential analysis of two commonly used measures
The Journal of Community Health
,
2011
, vol.
16
(pg.
222
-
241
2011
Davis
T. C.
Long
S. W.
Jackson
R. H.
Mayeaux
E. J.
George
R. B.
Murphy
P. W.
, et al.  .
Rapid estimate of adult literacy in medicine: A shortened screening instrument
Family Medicine
,
1993
, vol.
25
(pg.
391
-
395
)
Dierckx
E.
Engelborghs
S.
Raedt
R. D.
Van Buggenhout
M.
De Deyn
P. P.
Verleye
G.
, et al.  .
Differentiation between dementia and depression among older persons: Can the difference between actual and premorbid intelligence be useful?
Journal of Geriatric Psychiatry and Neurology
,
2008
, vol.
21
(pg.
242
-
249
)
Donelle
L.
Hoffman-Goetz
L.
Arocha
J. F.
Assessing health numeracy among community-dwelling older adults
Journal of Health Communication
,
2007
, vol.
12
(pg.
651
-
665
)
Fromm
D.
Holland
A.
Nebes
R. D.
Oakley
M. A.
Cortex
,
1991
, vol.
27
(pg.
367
-
376
)
Grober
E.
Charles
B. H.
Lipton
R. B.
Zonderman
A. B.
Resnick
S. M.
Kawas
C.
Memory impairment, executive dysfunction, and intellectual decline in preclinical Alzheimeŕs disease
Journal of the International Neuropsychological Society
,
2008
, vol.
14
(pg.
266
-
278
)
Instituto Brasileiro de Geografia e Estatística (IBGE)
2010

CENSO 2010. Retrieved July 29, 2012, from http://www.ibge.gov.br/home/.
Jastak
S.
Wilkinson
G. S.
Wide Range Achievement Rest-Revised 3
,
1993
Wilmington, DE
Jastak Associates
Kirsch
I. S.
Jungeblut
A.
Jenkins
L.
A.
An adult literacy in America: A first look at the results of the National Adult Literacy Survey
,
2010
3rd ed
Washington, DC
National Center for Education Statistics, U.S. Department of Education
Malloy-Diniz
L. F.
Lasmar
V. A. P.
Gazinelli
L. S. R.
Fuentes
D.
J. V.
The Rey Auditory-Verbal Learning Test: Applicability for the Brazilian elderly population
Revista Brasileira de Psiquiatria
,
2007
, vol.
29

4
(pg.
324
-
329
)
Manly
J. J.
Schupf
N.
Tang
M.
Stern
Y.
Cognitive decline and literacy among ethnically diverse elders
Journal of Geriatric Psychiatry and Neurology
,
2005
, vol.
18
(pg.
213
-
217
)
Mattis
S.
Dementia rating scale. Professional manual
,
1988
Florida
Psychological Assessment Resource
McKhann
G.
Drachman
D.
Folstein
M.
Katzman
R.
Price
D.
E. M.
Clinical diagnosis of Alzheimer's disease: Report of the NINCDS-ADRDA work group under the auspice of the Department of Health And Human Services Task Force on Alzheimer's disease
Neurology
,
1984
, vol.
34
(pg.
939
-
944
)
Nelson
H. E.
,
1982
UK
NFER Nelson
Nitrini
R.
Caramelli
P.
Herrera
H.
Porto
C. S.
Charchat-Fichman
H.
Carthery
M. T.
, et al.  .
Performance of illiterate and literate nondemented elderly subjects in two tests of long-term memory
Journal of the International Neuropsychological Society
,
2004
, vol.
10
(pg.
634
-
638
)
Nitrini
R.
Caramelli
P.
Porto
C. S.
Charchat-Fichman
H.
Formigoni
A. P.
Carthery-Goulart
M. T.
, et al.  .
Brief cognitive battery in the diagnosis of mild Alzheimers disease in subjects with medium and high levels of education
Dementia and Neuropsychologia
,
2007
, vol.
1
(pg.
32
-
36
)
Paque
L.
Warrington
E. K.
A longitudinal study of reading ability in patients suffering from dementia
Journal of the International Neuropsychological Society
,
1995
, vol.
1
(pg.
517
-
524
)
Patterson
K.
Graham
N.
Hodges
J. R.
Reading in dementia of the Alzheimer type: A preserved ability?
Neuropsychology
,
1994
, vol.
8
(pg.
395
-
407
)
Peters
E.
Vastfjall
D.
Slovic
P.
Mertz
C. K.
Mazzocco
K.
Dickert
S.
Numeracy and decision making
Psychological Science
,
2006
, vol.
17

5
(pg.
407
-
413
)
Petersen
R. C.
Doody
R.
Kurz
A.
Mohs
R. C.
Morris
J. C.
Rabins
P. V.
, et al.  .
Current concepts in Mild Cognitive Impairment
Archives of Neurology
,
2001
, vol.
58
(pg.
1985
-
1992
)
Pfeffer
R. I.
Pfeffer
R. I.
Kurosaki
T. T.
Harrah
C. H.
Chance
J. M.
Filos
S.
Measurement of functional activities in older adults in the community
The Journals of Gerontology
,
1982
, vol.
37
(pg.
323
-
329
)
Porto
C. S.
Caramelli
P.
Nitrini
R.
The influence of schooling on performance in the Mattis Dementia Rating Scale (DRS)
Dementia and Neuropsychologia
,
2010
, vol.
4
(pg.
126
-
130
)
Porto
C. S.
Fichman
H. C.
Caramelli
P.
Bahia
V. S.
Nitrini
R.
Brazilian version of the Mattis Dementia Rating Scale: Diagnosis of mild dementia in Alzheimers disease
Arquivos de Neuropsiquiatria
,
2003
, vol.
61

2-B
(pg.
339
-
345
)
Powell
B. D.
Brossart
D. F.
Reynolds
C. R.
Evaluation of the accuracy of two regression-based methods for estimating premorbid IQ
Archives of Clinical Neuropsychology
,
2003
, vol.
18
(pg.
277
-
292
)
Ringe
W. J.
Saine
K. C.
Lacritz
L. H.
Hynan
L. S.
Cullum
C. M.
Assessment
,
2002
, vol.
9
(pg.
254
-
260
)
Rosselli
M.
Neuropsychology of illiteracy
Behavioral Neurology
,
1993
, vol.
6
(pg.
107
-
112
)
Spreen
O.
Strauss
E.
Sherman
E. M. S.
A compendium of neuropsychological tests: Administration, norms, and commentary
,
2006
3rd ed.
New York
Oxford University Press
Starr
J. M.
Lonie
J.
Estimated pré-morbid IQ effects on cognitive and functional outcomes in Alzheimer disease: A longitudinal study a treated cohort
BMC Psychiatry
,
2008
, vol.
8
pg.
27

Sunderland
T.
Hill
J. L.
Melow
A. M.
Lawlor
B. A.
Gundersheimer
J.
Newhouse
P. A.
, et al.  .
Clock drawing in Alzheimer's disease: A novel measure of dementia severity
Journal of American Geriatrics Society
,
1989
, vol.
37
(pg.
725
-
729
)
United Nations Educational, Scientific and Cultural Organization (UNESCO) Global Monitoring Report on Education for All
2011

Retrieved November 16, 2012, from http://www.unesco.org
Weiss
B. D.
Mays
M. Z.
Martz
W.
Castro
K. M.
DeWalt
D. A.
Pignone
M. P.
, et al.  .
Quick assessment of literacy in primary care: The newest vital sign
The Annals of Family Medicine
,
2005
, vol.
3
pg.
514

Williams
M. V.
Parker
R.
Barker
D.
Parikh
N. S.
Pitkin
K.
Coates
W. C.
, et al.  .
Inadequate functional health literacy among patients at two public hospitals
Journal of the American Medical Association
,
1995
, vol.
274
(pg.
1677
-
1682
)
Wolf
M. S.
Curtis
L. M.
Wilson
E. A. H.
Revelle
W.
Waite
K. R.
Smith
S. G.
, et al.  .
Literacy, cognitive function, and health: Results of the LitCog study
Journal of General Internal Medicine
,
2012
, vol.
27

10
(pg.
1300
-
1307
)
Yassuda
M. S.
Diniz
B. S. O.
Flaks
M. K.
Viola
L. F.
Pereira
F. S.
Nunes
P. V.
, et al.  .
Neuropsychological profile of Brazilian older adults with heterogeneous educational backgrounds
Archives of Clinical Neuropsychology
,
2009
, vol.
24
(pg.
71
-
79
)