Abstract

The main goal of this study was to produce normative data for the Portuguese population on the Trail Making Test (TMT). The study included 1,038 community-dwelling individuals aged between 18 and 93 years, who had educational backgrounds ranging from 3 to 22 years. The results showed that sex, age, and education were significantly associated with TMT performance. These demographic characteristics accounted for 57% of the performance variance at part A and 50% at part B. The normative data are presented as regression-based algorithms to adjust direct and derived test scores for sex, age, and education. The adjusted scores' percentile distributions and their correspondence with scaled scores are also provided.

Introduction

The Trail Making Test (TMT) was developed as part of the Army Individual Test Battery (1944) and was used in 1946 by Armitage to assess the effects of brain injury in soldiers. Later, this test was integrated in the Halstead-Reitan neuropsychological battery (Reitan & Wolfson, 1993). Since then, administration and scoring procedures have varied significantly (e.g., scoring of errors; discontinuation criteria) and alternate forms have been proposed (Atkinson, Ryan, Kryza & Charette, 2011; Salthouse et al., 2000). The TMT is currently one of the most widely used instruments in clinical and experimental neuropsychology (Rabin, Barr, & Burton, 2005; Strauss, Sherman, & Spreen, 2006).

Trail Making Test part A measures attention, visual scanning, and speed of eye-hand coordination and information processing. Part B, additionally, assesses working memory and executive functions, particularly, the ability to switch between sets of stimuli (Lezak, Howieson, & Loring, 2004; Mitrushina, Boone, & D'Elia, 1999; Sanchez-Cubillo et al., 2009). Trail Making Test part B is known to be sensitive to frontal lobe damage (Demakis, 2004; Gouveia, Brucki, Malheiros, & Bueno, 2007; Kaleita et al., 2004; McDonald, Delis, Norman, Tecoma, & Iragui-Madozi, 2005; Stuss, Bisschop, Alexander, Levine, & Katz, 2001). However, no clear association with lesion side has been demonstrated.

The TMT can be a useful indicator of neurological integrity (Larrabee, Millis, & Meyers, 2008; Reitan, 1958; Reitan & Wolfson, 1993; Strauss et al., 2006). It is highly sensitive to a variety of neurological conditions, including head injury (e.g., Heled, Hoofien, Margalit, Natovich, & Agranov, 2012; Lange, Iverson, Zakrzewski, Ethel-King, & Franzen, 2005), Alzheimer's disease (e.g., Amieva et al., 1998; Ashendorf et al., 2008), dementia with Lewy bodies (Ferman et al., 2006), Huntington's disease (Lemiere, Decruyenaere, Evers-Kiebooms, Vandenbussche, & Dom, 2004; O'Rourke et al., 2011), and minimal or subclinical hepatic encephalopathy (e.g., Ferenci et al., 2002). The TMT is even sensitive to preclinical manifestations of certain neurodegenerative diseases, such as Alzheimer's disease (Chen et al., 2001) and Huntington's disease (O'Rourke et al., 2011). In clinical practice, TMT is considered a valuable tool to detect and document changes in cognition. It may also contribute to distinguish between neurological conditions (Ashendorf et al., 2008; Ferman et al., 2006; Heidler-Gary et al., 2007).

Several studies have demonstrated the ecological validity of the test. The TMT is significantly related with performance on instrumental activities of daily living (e.g., safe driving), in community-dwelling older adults and neurological patients (e.g., Bell-McGinty, Podell, Franzen, Baird, & Williams, 2002; Dawson, Anderson, Uc, Dastrup, & Rizzo, 2009; Emerson et al., 2012; Hanks et al., 2008; Mitchell & Miller, 2008). Longitudinal studies have also shown that TMT performance is a good predictor of clinical and functional outcomes, namely conversion to dementia, functional decline, and death in community-dwelling older adults (Chen et al., 2001; Johnson, Lui, & Yaffe, 2007; Vazzana et al., 2010) as well as in patients with major depression (Potter et al., in press), mild cognitive impairment (e.g., Chapman et al., 2011; Ewers et al., 2012; Gomar et al., 2011), traumatic brain injury (Hanks et al., 2008), or stroke (Wiberg, Kilander, Sundstro, Byberg, & Lind, 2012).

In clinical practice, in addition to the completion times of TMT-A and TMT-B, derived scores are often used for interpretive purposes (Lezak et al., 2004; Mitrushina et al., 1999; Strauss et al., 2006). The most common derived scores are: difference (B − A), ratio (B/A), and proportion (B − A/A). The main reason to use these scores is to reduce the impact of individual variability on part B, by using the subject as his or her own control and also, to remove the components of motor speed and visual scanning speed from part B, by comparing part B to part A. Numerous authors have argued that these derived scores are purer measures of executive functions than the direct raw measures (e.g., Arbuthnott & Frank, 2000; Axelrod, Aharon-Peretz, Tomer, & Fisher, 2000; Corrigan & Hinkeldey, 1987; Drane, Yuspeh, Huthwaite, & Klingler, 2002; Heaton, Nelson, Thompson, Burks, & Franklin, 1985; Hester, Kinsella, Ong, & McGregor, 2005; Horton & Roberts, 2001; Lamberty, Chatel, Bieliauskas, & Linas, 1994; Sanchez-Cubillo et al., 2009; Stuss et al., 2001). These scores are known to be sensitive to frontal lobe damage (Stuss et al., 2001). However, the difference score is more vulnerable to demographic characteristics than the ratio score (Drane et al., 2002; Hester et al., 2005; Lamberty et al., 1994). Arbuthnott and Frank (2000) provided evidence of a stronger correlation between a non-motor executive control task that involved set-switching and the TMT ratio score, than with direct raw or difference scores. Sum (A + B) and multiplication (A × B/100) have also been explored (Horton & Roberts, 2001; Lange et al., 2005). These two derived scores provide indices of overall cognitive functioning and not just executive functions. Another measure of interest is the number of performance errors (Amieva et al., 1998; Ashendorf et al., 2008; Stuss et al., 2001).

This study aims to explore the influence of sex, age, and education on TMT direct (i.e., time to completion and performance errors for each part) and derived scores (i.e., difference, ratio, proportion, sum, and multiplication scores); and to provide normative data for the Portuguese population that reflect these demographic influences.

Methods

Participants

Subjects included 1,038 community-dwelling Portuguese individuals who were participating in a comprehensive neuropsychological normative project. Participants had between 18 and 93 years of age and between 3 and 22 years of education (i.e., formal schooling completed with success). They were all volunteers and did not receive any monetary compensation. All participants provided their written informed consent in accordance with the Helsinki Declaration.

Recruitment

The participants were recruited in the community via word of mouth. The inclusion criteria were: ≥18 years of age; Portuguese as the first language; have lived in Portugal in the last 5 years; have ≥3 years of education; did ≥50% of formal schooling in Portugal or in a territory with Portuguese administration; and the absence of significant motor, auditory or visual deficits after correction. Each participant's cognitive normalcy was validated via an informant (i.e., personal physician, relative, friend). Individuals with a history of developmental disorders (e.g., learning disability), neurological disorder (e.g., traumatic brain injury, dementia), or moderate to severe psychopathology (e.g., major depression, psychosis, alcoholism) were excluded.

Comparison between Study Sample and Population Census

The data were collected throughout Portugal (mainland), although the majority of the participants (53%) were recruited in the northern region. The south region, which includes the capital, was underrepresented (only 18% of the study sample). Also, unlike the general population (48% men and 52% women), the number of men participating in the study (31%) was considerably lower than the number of women (69%). The age and the educational background of the normative samples were representative of the Portuguese population. According to the 2011 census (Instituto Nacional de Estatística, 2011), 19% of the Portuguese population are illiterates or did not complete the first cycle of basic education (i.e., <4 years of education); 25% have four years of education; 29% attained between 6 and 9 years of education; high school was completed by 12%; and ∼15% had more than 12 years of education.

Procedures

The TMT was part of a comprehensive neuropsychological assessment protocol. Prior to testing, the participant's clinical history was explored with an interview. The examiners were trained psychologists.

Trail Making Test part A was administered to all participants, whereas part B was only applied to participants with ≥4 years of education.

Trail Making Test

Part A

The participants were asked to draw lines to connect consecutively 25 encircled numbers. The participants were urged to connect the circles as quickly as possible. Timing was initiated when the participant was asked to start. Participants were allowed to lift the pencil from the page. Whenever an error was made, the examiner pointed it out and explained it. After the explanation, the examiner marked out the wrong part and guided the participant to the last circle completed correctly.

The test condition was not administered if the participant was unable to perform the practice condition (i.e., made more than two errors). The test condition was discontinued after 200 s or after four errors, unless the patient was within three circles of the end. The instructions in Portuguese are available as Supplementary material online.

Part B

The participants were asked to draw lines to connect consecutively encircled numbers and letters by alternating between the two sequences (e.g., 1-A-2-B, etc.) progressively up to number 13. The participants were urged to connect the circles as quickly as possible.

Timing was initiated when the participant was asked to start. Participants were allowed to lift the pencil from the page. Whenever an error was made, the examiner pointed it out and explained it. After the explanation, the examiner marked out the wrong part and guided the participant to the last circle completed correctly.

The test condition was not administered if the participant was unable to perform the practice condition (i.e., made more than two errors). The test condition was discontinued after 400 s or after four errors, unless the participant was within three circles of the end. The forms and the instructions in Portuguese are available as Supplementary material online.

Scoring

The TMT provided four direct scores and five derived scores. The direct measures of performance were: time (s) to complete part A and part B and performance errors during part A and part B. Based on the direct time scores, five derived scores were calculated: difference score (B − A), ratio score (B/A), proportion score (B − A/A), sum score (A + B), and multiplication score (A × B/100). Lower raw scores and higher adjusted scores correspond to better performance.

Statistical Analyses

Categorical variables were summarized with frequencies and percentages, whereas continuous data were described using means and standard deviations (SD). Scatter plots were used to visualize the associations between demographic variables, and between these variables and test results. Pearson's correlations (r) and explained variance (r2) were used to explore the association of demographic characteristics (i.e., age and education) with test performances. The Mann–Whitney test was applied to test the associations between sex and test results. Multiple logistic regression analysis was used to model the odds of discontinuation of part B.

Both direct (time) and derived scores were log transformed due to the skewedness of the distributions. Multiple linear regression analyses were conducted with test scores in the logarithmic scale as dependent variables and demographic variables as covariates. We considered the possibility of a quadratic effect for age and education. The assumptions of homoscedasticity and normal distribution of the residuals were verified. The adjustment of test scores for demographic characteristics was based on regression coefficients. The standardized regression residuals (standardized residuals = residuals divided by the SD) were used to identify the associated percentiles. The residuals represent the difference between the score of an individual and the mean score of individuals with the same age and education obtained by the linear regression. Higher adjusted scores correspond to better performance.

Results

Normative Sample

Trail Making Test part A was administered and completed by 1,038 study participants. Part B was applied to all participants with ≥4 years of education (n = 986 participants), but only 922 participants were able to complete the test. Among those that failed to complete part B, 84% had 4 years of education. At this level of education, the frequency of discontinuation was 22%. This frequency decreased to 3% and 2%, respectively for individuals with 5–9 years of education and 10–12 years of education. All participants with >12 years of education completed part B. Multiple logistic regression analysis, with sex, age, education, and time to complete part A as independent variables, revealed that the odds of completion increased with education (adjusted OR = 1.4, CI95 = [1.2, 1.7]; p < .001) and decreased with slower performance at part A (adjusted OR = 0.98, CI95 = [0.98, 0.99]; p < .001). Sex and age were not significantly associated with the application of the discontinuation rules.

In a posteriori data inspection, TMT A results from 13 participants and TMT B results from 8 participants were excluded from the normative sample, because they were considered outliers (i.e., respectively ≥230 s and ≥430 s). Trail Making Test part A sample was composed by 1,025 participants (708 women and 317 men; mean age = 54.3 years, SD = 17.9, range: 18–93; and mean education = 9.5 years, SD = 4.8, range: 3–22). Trail Making Test part B sample consisted of 914 participants (622 women and 292 men; mean age = 52.2 years, SD = 17.5, range: 18–93; and mean education = 10.2 years, SD = 4.7, range: 4–22).

Significant negative correlations were observed between participants' age and education on both part A (r = −.419) and part B (r = −.343) samples. Age and education were not statistically different between men and women (p > .05).

Raw Scores

The mean time to complete part A was 58 s (SD = 37) and to complete part B was 119 (SD = 73). However, both direct time scores presented a skewed distribution (TMT A = 1.8 and TMT B = 1.6). Table 1 presents the means and SDs of direct time scores and derived scores for each sex, age group, and education group. All participants, except two, took longer to complete part B than part A.

Table 1.

TMT direct and derived scores per sex, age, and education groups

 Direct scores
 
Derived scores
 
 Part A (n = 1025)
 
Part B (n = 914)
 
B − A B/A B − A/A A + B A × B/100 
 n M (SDMIN—MAX n M (SDMIN—MAX M (SDM (SDM (SDM (SDM (SD
Sex 
 Women 708 69 61 (37) 17–225 622 68 122 (73) 33–401 69 (54) 2.4 (0.9) 1.4 (0.8) 176 (98) 82 (106) 
 Men 317 31 52 (37) 13–224 292 32 113 (71) 29–418 66 (51) 2.5 (1.0) 1.5 (1.0) 160 (96) 69 (95) 
Age 
 18–29 136 13 31 (11) 13–83 136 15 68 (26) 29–198 37 (22) 2.3 (0.8) 1.3 (0.8) 99 (32) 22 (15) 
 30–39 123 12 36 (13) 14–80 120 13 84 (37) 32–278 48 (31) 2.5 (0.9) 1.5 (0.9) 119 (45) 32 (24) 
 40–49 109 11 47 (20) 19–130 107 12 112 (57) 34–300 66 (44) 2.5 (0.9) 1.5 (0.9) 158 (72) 59 (60) 
 50–59 212 21 51 (27) 15–206 200 22 113 (61) 47–360 65 (49) 2.4 (0.9) 1.4 (0.9) 162 (77) 64 (67) 
 60–69 217 21 60 (29) 20–182 191 21 131 (67) 50–401 77 (53) 2.5 (0.9) 1.5 (0.9) 186 (85) 83 (78) 
 70–79 144 14 90 (43) 24–225 105 12 172 (89) 49–401 94 (68) 2.3 (0.8) 1.3 (0.8) 250 (119) 158 (147) 
 ≥80 84 109 (49) 44–224 55 219 (89) 76–418 124 (66) 2.4 (0.7) 1.4 (0.7) 314 (125) 236 (193) 
Education 
 3 51 120 (43) 42–218 — — — — — — — — — 
 4 238 23 84 (43) 25–225 187 21 191 (83) 48–401 118 (64) 2.8 (1.0) 1.8 (1.0) 265 (110) 162 (139) 
 5–9 272 27 52 (27) 15–169 266 29 121 (63) 38–398 69 (46) 2.5 (0.9) 1.5 (0.9) 174 (87) 78 (98) 
 10–12 189 18 46 (23) 13–168 186 20 100 (57) 29–418 54 (41) 2.3 (0.8) 1.3 (0.8) 147 (76) 57 (75) 
 >12 275 27 38 (17) 14–183 275 30 81 (40) 32–385 43 (31) 2.2 (0.8) 1.2 (0.8) 119 (54) 36 (44) 
 Direct scores
 
Derived scores
 
 Part A (n = 1025)
 
Part B (n = 914)
 
B − A B/A B − A/A A + B A × B/100 
 n M (SDMIN—MAX n M (SDMIN—MAX M (SDM (SDM (SDM (SDM (SD
Sex 
 Women 708 69 61 (37) 17–225 622 68 122 (73) 33–401 69 (54) 2.4 (0.9) 1.4 (0.8) 176 (98) 82 (106) 
 Men 317 31 52 (37) 13–224 292 32 113 (71) 29–418 66 (51) 2.5 (1.0) 1.5 (1.0) 160 (96) 69 (95) 
Age 
 18–29 136 13 31 (11) 13–83 136 15 68 (26) 29–198 37 (22) 2.3 (0.8) 1.3 (0.8) 99 (32) 22 (15) 
 30–39 123 12 36 (13) 14–80 120 13 84 (37) 32–278 48 (31) 2.5 (0.9) 1.5 (0.9) 119 (45) 32 (24) 
 40–49 109 11 47 (20) 19–130 107 12 112 (57) 34–300 66 (44) 2.5 (0.9) 1.5 (0.9) 158 (72) 59 (60) 
 50–59 212 21 51 (27) 15–206 200 22 113 (61) 47–360 65 (49) 2.4 (0.9) 1.4 (0.9) 162 (77) 64 (67) 
 60–69 217 21 60 (29) 20–182 191 21 131 (67) 50–401 77 (53) 2.5 (0.9) 1.5 (0.9) 186 (85) 83 (78) 
 70–79 144 14 90 (43) 24–225 105 12 172 (89) 49–401 94 (68) 2.3 (0.8) 1.3 (0.8) 250 (119) 158 (147) 
 ≥80 84 109 (49) 44–224 55 219 (89) 76–418 124 (66) 2.4 (0.7) 1.4 (0.7) 314 (125) 236 (193) 
Education 
 3 51 120 (43) 42–218 — — — — — — — — — 
 4 238 23 84 (43) 25–225 187 21 191 (83) 48–401 118 (64) 2.8 (1.0) 1.8 (1.0) 265 (110) 162 (139) 
 5–9 272 27 52 (27) 15–169 266 29 121 (63) 38–398 69 (46) 2.5 (0.9) 1.5 (0.9) 174 (87) 78 (98) 
 10–12 189 18 46 (23) 13–168 186 20 100 (57) 29–418 54 (41) 2.3 (0.8) 1.3 (0.8) 147 (76) 57 (75) 
 >12 275 27 38 (17) 14–183 275 30 81 (40) 32–385 43 (31) 2.2 (0.8) 1.2 (0.8) 119 (54) 36 (44) 

Test scores are presented as means (M), standard deviations (SD), minimum (MIN), and maximum (MAX).

The majority of the participants did not make any performance errors at part A (84%) or part B (63%). More than one error at part A or two errors at part B were made by less than 5% of the normative sample. Given the low frequency of errors, no further analysis was done.

Demographic Effects

Men outperformed women on part A (p = .015), part B (p = .017), sum (p = .001), and multiplication (p < .001). However, women showed better ratio (p = .006) and proportion (p = .006) scores than men. The difference score was not statistically associated with sex (p = .784). The effects of age and education on test scores were also investigated and significant linear associations were found (Table 2). Both age and education were significantly correlated with direct scores. Education was associated with all derived scores, whereas age was significantly related only with three derived scores (i.e., difference score, sum, and multiplication score). Scatter plots revealed that the relation between demographic variables and TMT scores had quadratic shapes.

Table 2.

Correlations and shared variances between TMT direct and derived scores and demographic variables

TMT  Age
 
Education
 
  r r2 r r2 
Direct time scores Part A .583* .340 −.519* .269 
Part B .513* .263 −.521* .271 
Derived scores B − A .393* .154 −.478* .228 
B/A .002 <.001 −.210* .044 
B − A/A .002 <.001 −.210* .044 
A + B .550* .303 −.516* .266 
A × B/100 .485* .235 −.417* .174 
TMT  Age
 
Education
 
  r r2 r r2 
Direct time scores Part A .583* .340 −.519* .269 
Part B .513* .263 −.521* .271 
Derived scores B − A .393* .154 −.478* .228 
B/A .002 <.001 −.210* .044 
B − A/A .002 <.001 −.210* .044 
A + B .550* .303 −.516* .266 
A × B/100 .485* .235 −.417* .174 

Women = 0; Men = 1; *p < .001.

After TMT direct scores and derived scores were converted to a logarithmic scale, multiple regression analyses were conducted with variables sex, age, age squared, education, and education squared as covariates. The regression model partly explained (r2) the variance of part A (57%), part B (50%), B − A (32%), B/A (7%), B − A/A (7%), A + B (54%), and A × B/100 (55%). Each independent variable of the multiple regression models remained statistically associated (p < .05) with both direct time measures. The univariate (Table 2) and the multivariate regression analyses revealed a similar pattern of significant associations (p < .05) between derived scores and demographic variables. Education was the only demographic variable to be significantly related (p < .05) with all TMT direct and derived measures, after adjusting for sex and age.

Regression-based Normative Data

Regression-based algorithms were developed to adjust direct and derived test scores for sex, age, and education (Table 3). In other words, the algorithms convert “raw” scores of an individual into standardized Z scores. The percentiles and the scaled scores associated with the adjusted scores are shown in Table 4. When time to complete part B is equal or less than part A, the adjusted scores for derived scores difference and proportion could not be calculated.

Table 3.

Regression based algorithms to calculate TMT direct and derived adjusted scores

TMT Algorithms 
Direct scores 
 Part A Time −((log scale score − 4.449433 + (sex × 0.147749) + (age × 0.009158) − (age2 × 0.000228) + (education × 0.133867) − (education2 × 0.004592))/0.360188) 
 Part B Time −((log scale score − 5.283036 + (sex × 0.068572) + (age × 0.006532) − (age2 × 0.000185) + (education × 0.130970) − (education2 × 0.003868))/0.378414) 
Derived scores 
 B − A −((log scale score − 4.865212 − (sex × 0.01674) + (age × 0.005008) − (age2 × 0.000152) + (education × 0.169454) − (education2 × 0.004905))/0.617073) 
 B/A −((log scale score − 1.056782 − (sex × 0.078508) − (age × 0.001797) + (age2 × 0.000032) + (education × 0.035416) − (education2 × 0.000826))/0.324419) 
 B − A/A −((log scale score − 0.632885 − (sex × 0.162626) − (age × 0.003410) + (age2 × 0.000065) + (education × 0.073129) − (education2 × 0.001831))/0.656273) 
 A + B −((log scale score − 5.582173 + (sex × 0.092665) + (age × 0.006851) − (age2 × 0.000192) + (education × 0.120118) − (education2 × 0.003610))/0.341221) 
 A × B/100 −((log scale score − 4.904119 + (sex × 0.215653) + (age × 0.014860) − (age2 × 0.000402) + (education × 0.226524) − (education2 × 0.006909))/0.654732) 
TMT Algorithms 
Direct scores 
 Part A Time −((log scale score − 4.449433 + (sex × 0.147749) + (age × 0.009158) − (age2 × 0.000228) + (education × 0.133867) − (education2 × 0.004592))/0.360188) 
 Part B Time −((log scale score − 5.283036 + (sex × 0.068572) + (age × 0.006532) − (age2 × 0.000185) + (education × 0.130970) − (education2 × 0.003868))/0.378414) 
Derived scores 
 B − A −((log scale score − 4.865212 − (sex × 0.01674) + (age × 0.005008) − (age2 × 0.000152) + (education × 0.169454) − (education2 × 0.004905))/0.617073) 
 B/A −((log scale score − 1.056782 − (sex × 0.078508) − (age × 0.001797) + (age2 × 0.000032) + (education × 0.035416) − (education2 × 0.000826))/0.324419) 
 B − A/A −((log scale score − 0.632885 − (sex × 0.162626) − (age × 0.003410) + (age2 × 0.000065) + (education × 0.073129) − (education2 × 0.001831))/0.656273) 
 A + B −((log scale score − 5.582173 + (sex × 0.092665) + (age × 0.006851) − (age2 × 0.000192) + (education × 0.120118) − (education2 × 0.003610))/0.341221) 
 A × B/100 −((log scale score − 4.904119 + (sex × 0.215653) + (age × 0.014860) − (age2 × 0.000402) + (education × 0.226524) − (education2 × 0.006909))/0.654732) 

Women = 0; Men = 1.

Table 4.

Empirical percentile ranks associated to the TMT direct and derived adjusted scores

Percentile ranks Scaled scores Adjusted scores
 
  Part A Part B B − A B/A B − A/A A + B A × B/100 
−2.5 −2.5 −2.0 −2.7 −2.0 −2.3 −2.5 
−2.1 −2.1 −1.9 −2.3 −1.8 −2.2 −2.1 
3–5 −1.9, −1.6 −2.0, −1.7 −1.7, −1.6 −2.1, −1.7 −1.6, −1.5 −2.0, −1.7 −1.9, −1.7 
6–10 −1.5, −1.3 −1.6, −1.4 −1.5, −1.2 −1.6, −1.4 −1.4, −1.2 −1.6, −1.3 −1.6, −1.3 
11–18 −1.2, −0.9 −1.3, −0.9 −1.2, −0.9 −1.3, −0.9 −1.2, −0.9 −1.3, −1.0 −1.2, −1.0 
19–28 −0.9, −0.6 −0.9, −0.6 −0.8, −0.6 −0.9, −0.5 −0.8, −0.6 −0.9, −0.6 −0.9, −0.6 
29–40 −0.6, −0.3 −0.5, −0.2 −0.5, −0.3 −0.5, −0.2 −0.6, −0.3 −0.5, −0.2 −0.6, −0.2 
41–59 10 −0.3, 0.3 −0.2, 0.3 −0.2, 0.2 −0.2, 0.3 −0.3, 0.1 −0.2, 0.3 −0.2, 0.3 
60–71 11 0.3, 0.6  0.3, 0.6 0.2, 0.5 0.3, 0.6 0.2, 0.5 0.3, 0.6 0.3, 0.6 
72–81 12 0.6, 0.9 0.6, 0.9 0.5, 0.8 0.7, 0.9 0.5, 0.8 0.6, 0.9 0.6, 0.9 
82–89 13 1.0, 1.2 0.9, 1.2 0.8, 1.1 1.0, 1.2 0.8, 1.2 0.9, 1.2 0.9, 1.2 
90–94 14 1.3, 1.5 1.3 1.5 1.2, 1.5 1.3, 1.4 1.2, 1.5 1.2, 1.6 1.2, 1.5 
95–97 15 1.6, 1.8 1.6, 1.8 1.6, 2.0 1.5, 1.7 1.6, 1.8 1.6, 1.8 1.6, 1.9 
98 16 2.0 2.0 2.1 1.9 2.2 2.0 2.0 
99 17 2.2 2.2 2.5 2.0 2.6 2.2 2.2 
Percentile ranks Scaled scores Adjusted scores
 
  Part A Part B B − A B/A B − A/A A + B A × B/100 
−2.5 −2.5 −2.0 −2.7 −2.0 −2.3 −2.5 
−2.1 −2.1 −1.9 −2.3 −1.8 −2.2 −2.1 
3–5 −1.9, −1.6 −2.0, −1.7 −1.7, −1.6 −2.1, −1.7 −1.6, −1.5 −2.0, −1.7 −1.9, −1.7 
6–10 −1.5, −1.3 −1.6, −1.4 −1.5, −1.2 −1.6, −1.4 −1.4, −1.2 −1.6, −1.3 −1.6, −1.3 
11–18 −1.2, −0.9 −1.3, −0.9 −1.2, −0.9 −1.3, −0.9 −1.2, −0.9 −1.3, −1.0 −1.2, −1.0 
19–28 −0.9, −0.6 −0.9, −0.6 −0.8, −0.6 −0.9, −0.5 −0.8, −0.6 −0.9, −0.6 −0.9, −0.6 
29–40 −0.6, −0.3 −0.5, −0.2 −0.5, −0.3 −0.5, −0.2 −0.6, −0.3 −0.5, −0.2 −0.6, −0.2 
41–59 10 −0.3, 0.3 −0.2, 0.3 −0.2, 0.2 −0.2, 0.3 −0.3, 0.1 −0.2, 0.3 −0.2, 0.3 
60–71 11 0.3, 0.6  0.3, 0.6 0.2, 0.5 0.3, 0.6 0.2, 0.5 0.3, 0.6 0.3, 0.6 
72–81 12 0.6, 0.9 0.6, 0.9 0.5, 0.8 0.7, 0.9 0.5, 0.8 0.6, 0.9 0.6, 0.9 
82–89 13 1.0, 1.2 0.9, 1.2 0.8, 1.1 1.0, 1.2 0.8, 1.2 0.9, 1.2 0.9, 1.2 
90–94 14 1.3, 1.5 1.3 1.5 1.2, 1.5 1.3, 1.4 1.2, 1.5 1.2, 1.6 1.2, 1.5 
95–97 15 1.6, 1.8 1.6, 1.8 1.6, 2.0 1.5, 1.7 1.6, 1.8 1.6, 1.8 1.6, 1.9 
98 16 2.0 2.0 2.1 1.9 2.2 2.0 2.0 
99 17 2.2 2.2 2.5 2.0 2.6 2.2 2.2 

A user-friendly program is available online (http://neuropsi.up.pt/) to adjust direct and derived scores (Fig. 1). The clinician only needs to introduce the subject's sex, age, education, and time to complete part A and part B. For instance, if a man with 43 years of age and 9 years of education completes TMT part A in 60 s and part B in 160 s, the adjusted scores are −1.7 for part A and −1.8 for part B. These adjusted scores fall within percentile range 3–5. Thus, 3–5% of the “normal” men population with 43 years of age and 9 years of education take ≥60 s to complete part A and ≥160 s to complete part B. Concerning the derived scores, the performance is between the 3rd and 5th percentiles for A + B and A × B/100; between the 6th and the 10th percentiles for B − A; and between the 41st and the 59th percentiles for B/A and B − A/A.

Fig. 1.

Screenshot of the online tool to compute the adjusted TMT scores (http://neuropsi.up.pt).

Fig. 1.

Screenshot of the online tool to compute the adjusted TMT scores (http://neuropsi.up.pt).

Discussion

Trail Making Test normative data are presented as algorithms to adjust test scores for sex, age, and education, with subsequent correspondence between adjusted scores and percentile distributions. The adjusted scores can be interpreted as z scores, because they have normal distribution with mean ≃ 0 and SD ≃ 1. The percentile ranks can be converted into scaled scores (i.e., mean = 10 and SD = 3). These standardization procedures have significant advantages for both clinical and research practices. The adjustment to the individual's demographic characteristics and the use of common metrics allow the comparisons between tests and between individuals.

In agreement with most normative studies, TMT performance declined with age and improved with education (e.g., Ashendorf et al., 2008; Cangoz, Karakoc, & Selekler, 2009; Giovagnoli et al., 1996; Goul & Brown, 1970; Kennedy, 1981; Lucas et al., 2005; Pena-Casanova et al., 2009; Salthouse et al., 2000; Seo et al., 2006; Steinberg, Bieliauskas, Smith, & Ivnik, 2005; Tombaugh, 2004; Zalonis et al., 2008). The influence of these demographic characteristics on test performance appears to be more pronounced in this study than in previous reports. The strong effects of age and educational level probably reflect the heterogeneity of the normative sample, which is representative of the Portuguese population. Noteworthy, the number of years of education was significantly related to all direct and derived TMT scores, even after adjusting for sex and age. Measures of speed of performance (i.e., completion time, difference, sum, and multiplication) were highly influenced by age, but similar to previous studies (Hester et al., 2005; Lamberty et al., 1994) neither ratio nor proportion scores were related with age. Significant associations were found between sex and TMT performance. In general, women were slower than men, even after adjusting for age and education. However, women showed better ratio and proportion scores than men. Nonetheless, in agreement with other reports (Cangoz et al., 2009; Giovagnoli et al., 1996; Hester et al., 2005; Ivnik, Malec, Smith, & Tangalos, 1996; Seo et al., 2006), the effects of gender on TMT performance were less pronounced than those of age and education.

The influence of demographic characteristics on direct and derived scores varied considerably. As expected (Corrigan & Hinkeldey, 1987; Lamberty et al., 1994; Strauss et al., 2006), these effects were larger for raw completion times and for derived scores that resulted from difference, sum and multiplication, than for ratio or proportion scores. This variability raises the issue of interpreting TMT performance, particularly the derived scores, with reference to the subject's demographic characteristics. As demonstrated by the example presented in the results' section, the same test scores may produce diverging results among demographically adjusted derived scores. The ratio and the proportion scores are believed to be TMT's most sensitive indices for detecting impairments in executive functions, because they are less vulnerable to demographic characteristics (Corrigan & Hinkeldey, 1987; Lamberty et al., 1994).

Educational restrictions were applied to the collection of data because TMT requires specific knowledge (e.g., numerical and alphabetical sequences) that is traditionally acquired in formal education. Even with these restrictions, part B was discontinued in ∼6% of participants. The large majority (84%) of those that failed to complete part B only had 4 years of education. At this level of education, the frequency of discontinuation was considerable (22%). These data suggest that poor performance on TMT B may not be a reliable indicator of cognitive dysfunction in individuals with only 4 years of education.

In this normative study, education was operationalized as the number of years of formal regular schooling completed with success. Equivalences from the “New Opportunities program” initiative (a governmental program designed to enhance school certification and qualification levels of the Portuguese adult population) were not credited. This approach is vulnerable to the numerous changes in the educational system that have occurred throughout the last decades in Portugal. For most of the twentieth century, letters “K”, “W”, and “Y” were not part of the alphabet taught in the Portuguese primary schools. Even though the current teaching includes this set of letters and some foreign words written with “K” have since entered the lexicon, TMT part B stimulus did not include letter “K”. Considering these continuing changes, future studies ought to assess the reliability of the forms and normative algorithms and update the norms if necessary.

The study sample was representative of the Portuguese population regarding age and education, but not for sex or region of residence. It is unlikely that the overrepresentation of women in the study sample has produced a significant bias on the study results, due to (a) the large sample size, (b) the absence of significant associations between sex and other demographic variables, and (c) the adjustment of all TMT scores for sex. On the basis of the relatively small size of the country and its considerable migration, we do not expect geography to have a significant impact on the Portuguese norms.

Trail Making Test data were collected and norms were developed as part of a larger project that aims to standardize a series of widely used neuropsychological instruments. The norms for all measures are being developed within the same (or substantially overlapping) study sample. This co-norming approach is believed to facilitate the comparison between test scores, which enhances the precision of cognitive pattern analysis in clinical settings and allows for a greater diagnostic accuracy (Lucas et al., 2005). The collection of data from multiple tests is also beneficial because it provides a closer parallel to the clinical setting.

Supplementary Material

Supplementary material is available at Archives of Clinical Neuropsychology online.

Funding

This research received no specific grant from any funding agency, commercial or not-for-profit sectors. Armando Teixeira-Pinto was supported by the Australian National Health and Medical Research Council Grant 402764 to the Screening and Test Evaluation Program.

Conflict of Interest

None declared.

Acknowledgements

Psychologists Alexandra Pereira, Alexandre Abreu, Ana Cristina Nunes, Andreia Gonçalves Azevedo, Cátia Alves, Cláudia Dias, Cristina Malaquias, Estelle Lima, Filipa Júlio, Maria do Céu Seabra, Mariana Oliveira, Marta Pinto, Patrícia Matos, Rui Quintas, Sandra Freitas, Sara Cabaço, Susana Pinto, Susana Vieira da Silva, and Vânia Santos collaborated in the collection of data. We would like to thank Ricardo Correia and Tiago Silva Costa for developing an online program to adjust test scores.

References

Amieva
H.
Sylviane
L.
Auriacombe
S.
Rainville
C.
Orgogozo
J.
Dartigues
J.
, et al.  . 
Analysis of error types in the Trail Making Test evidences an inhibitory deficit in dementia of the Alzheimer type
Journal of Clinical and Experimental Neuropsychology
 , 
1998
, vol. 
20
 (pg. 
280
-
285
)
Arbuthnott
K.
Frank
J.
Trail Making Test, Part B as a measure of executive control: validation using a set-switching paradigm
Journal of Clinical and Experimental Neuropsychology
 , 
2000
, vol. 
22
 (pg. 
518
-
528
)
Armitage
S. G.
An analysis of certain psychological tests used for the evaluation of brain injury
Psychology Monographs
 , 
1946
, vol. 
60
  
(Whole No. 277)
Army Individual Test Battery
Manual of directions and scoring
 , 
1944
Washington, DC
War Department, Adjutant General's Office
Ashendorf
L.
Jefferson
A. L.
O'Connor
M. K.
Chaisson
C.
Green
R. C.
Stern
R. A.
Trail Making Test errors in normal aging, mild cognitive impairment, and dementia
Archives of Clinical Neuropsychology
 , 
2008
, vol. 
23
 (pg. 
129
-
137
)
Atkinson
T. M.
Ryan
J. P.
Kryza
M.
Charette
L. M.
Using versions of the Trail Making Test as alternate forms
The Clinical Neuropsychologist
 , 
2011
, vol. 
25
 (pg. 
1193
-
1206
)
Axelrod
B. N.
Aharon-Peretz
J.
Tomer
R.
Fisher
T.
Creating interpretation guidelines for the Hebrew Trail Making Test
Applied Neuropsychology
 , 
2000
, vol. 
7
 (pg. 
186
-
188
)
Bell-McGinty
S.
Podell
K.
Franzen
M.
Baird
A. D.
Williams
M. J.
Standard measures of executive function in predicting instrumental activities of daily living in older adults
International Journal of Geriatric Psychiatry
 , 
2002
, vol. 
17
 (pg. 
828
-
834
)
Cangoz
B.
Karakoc
E.
Selekler
K.
Trail Making Test: Normative data for Turkish elderly population by age, sex and education
Journal of Neurological Sciences
 , 
2009
, vol. 
283
 (pg. 
73
-
78
)
Chapman
R. M.
Mapstone
M.
McCrary
J. W.
Gardner
M. N.
Porsteinsson
A.
Sandoval
T. C.
, et al.  . 
Predicting conversion from mild cognitive impairment to Alzheimer's disease using neuropsychological tests and multivariate methods
Journal of Clinical and Experimental Neuropsychology
 , 
2011
, vol. 
33
 (pg. 
187
-
199
)
Chen
P.
Ratcliff
G.
Belle
S. H.
Cauley
J. A.
DeKosky
S. T.
Ganguli
M.
Patterns of cognitive decline in presymptomatic Alzheimer disease: A prospective community study
Archives of General Psychiatry
 , 
2001
, vol. 
58
 (pg. 
853
-
858
)
Corrigan
J. D.
Hinkeldey
N. S.
Relationships between parts A and B of the Trail Making Test
Journal of Clinical Psychology
 , 
1987
, vol. 
43
 (pg. 
402
-
409
)
Dawson
J. D.
Anderson
S. W.
Uc
E. Y.
Dastrup
E.
Rizzo
M.
Predictors of driving safety in early Alzheimer disease
Neurology
 , 
2009
, vol. 
72
 (pg. 
521
-
527
)
Demakis
G. J.
Frontal Lobe Damage and Tests of Executive Processing: A Meta-Analysis of the Category Test, Stroop Test, and Trail-Making Test
Journal of Clinical and Experimental Neuropsychology
 , 
2004
, vol. 
26
 (pg. 
441
-
450
)
Drane
D. L.
Yuspeh
R. L.
Huthwaite
J. S.
Klingler
L. K.
Demographic characteristics and normative observations for derived-Trail Making Test indices
Neuropsychiatry, Neuropsychology, and Behavioral Neurology
 , 
2002
, vol. 
15
 (pg. 
39
-
43
)
Emerson
J. L.
Johnson
A. M.
Dawson
J. D.
Uc
E. Y.
Anderson
S. W.
Rizzo
M.
Predictors of driving outcome in advancing age
Psychology Aging
 , 
2012
, vol. 
27
 (pg. 
550
-
559
)
Ewers
M.
Walsh
C.
Trojanowski
J. Q.
Shaw
L. M.
Petersen
R. C.
Jack
C. R.
Jr
, et al.  . 
American Alzheimer's Disease Neuroimaging Initiative (ADNI)
Prediction of conversion from mild cognitive impairment to Alzheimer's disease dementia based upon biomarkers and neuropsychological test performance
Neurobiology of Aging
 , 
2012
, vol. 
33
 (pg. 
1203
-
1214
)
Ferenci
P.
Lockwood
A.
Mullen
K.
Tarter
R.
Weissenborn
K.
Biel
A. T.
Hepatic encephalopathy—definition, nomenclature, diagnosis, and quantification: final report of the working party at the 11th World Congresses of Gastroenterology, Vienna, 1998
Hepatology
 , 
2002
, vol. 
35
 (pg. 
716
-
721
)
Ferman
T. J.
Smith
G. E.
Boeve
B. F.
Graff-Radford
N. R.
Lucas
J. A.
Knopman
D. S.
, et al.  . 
Neuropsychological differentiation of dementia with Lewy bodies from normal aging and Alzheimer's disease
The Clinical Neuropsychologist
 , 
2006
, vol. 
20
 (pg. 
623
-
636
)
Giovagnoli
A. R.
Del Pesce
M.
Mascheroni
S.
Simoncelli
M.
Laiacona
M.
Capitani
E.
Trail Making Test: normative values from 287 normal adult controls
Italian Journal of Neurological Sciences
 , 
1996
, vol. 
17
 (pg. 
305
-
309
)
Gomar
J. J.
Bobes-Bascaran
M. T.
Conejero-Goldberg
C.
Davies
P.
Goldberg
T. E.
Alzheimer's Disease Neuroimaging Initiative
Utility of combinations of biomarkers, cognitive markers, and risk factors to predict conversion from mild cognitive impairment to Alzheimer disease in patients in the Alzheimer's disease neuroimaging initiative
Archives of General Psychiatry
 , 
2011
, vol. 
68
 (pg. 
961
-
969
)
Goul
W. R.
Brown
M.
Effects of age and intelligence on Trail Making Test performance and validity
Perceptual and Motor Skills
 , 
1970
, vol. 
30
 (pg. 
319
-
326
)
Gouveia
P. A. R.
Brucki
S. M. D.
Malheiros
S. M. F.
Bueno
O. F. A.
Disorders in planning and strategy application in frontal lobe lesion patients
Brain and Cognition
 , 
2007
, vol. 
63
 (pg. 
240
-
246
)
Hanks
R. A.
Millis
S. R.
Ricker
J. H.
Giacino
J. T.
Nakese-Richardson
R.
Frol
A. B.
, et al.  . 
The predictive validity of a brief inpatient neuropsychologic battery for persons with traumatic brain injury
Archives of Physical Medicine and Rehabilitation
 , 
2008
, vol. 
89
 (pg. 
950
-
957
)
Heaton
R. K.
Nelson
L. M.
Thompson
D. S.
Burks
J. S.
Franklin
G. M.
Neuropsychological findings in relapsing-remitting and chronic-progressive multiple sclerosis
Journal of Consulting and Clinical Psychology
 , 
1985
, vol. 
53
 (pg. 
103
-
110
)
Heidler-Gary
J.
Gottesman
R.
Newhart
M.
Chang
S.
Ken
L.
Hillis
A. E.
Utility of behavioral versus cognitive measures in differentiating between subtypes of frontotemporal lobar degeneration and Alzheimer's disease
Dementia and Geriatric Cognitive Disorders
 , 
2007
, vol. 
23
 (pg. 
184
-
193
)
Heled
E.
Hoofien
D.
Margalit
D.
Natovich
R.
Agranov
E.
The Delis-Kaplan Executive Function System Sorting Test as an evaluative tool for executive functions after severe traumatic brain injury: a comparative study
Journal of Clinical and Experimental Neuropsychology
 , 
2012
, vol. 
34
 (pg. 
151
-
159
)
Hester
R. L.
Kinsella
G. J.
Ong
B.
McGregor
J.
Demographic influences on baseline and derived scores from the Trail Making Test in healthy older Australian adults
The Clinical Neuropsychologist
 , 
2005
, vol. 
19
 (pg. 
45
-
54
)
Horton
A. M.
Roberts
C.
Derived trail making test indices in a sample of substance abusers: demographic effects
International Journal of Neuroscience
 , 
2001
, vol. 
111
 (pg. 
123
-
132
)
Instituto Nacional de Estatística
Censos 2011 – Resultados Provisórios
 , 
2001
Lisboa
INE, I.P
Ivnik
R. J.
Malec
J. F.
Smith
G. E.
Tangalos
E. G.
Neuropsychological tests’ norms above age 55: COWAT, BNT, MAE Token, WRAT-R Reading, AMNART, STROOP, TMT, and JLO
The Clinical Neuropsychologist
 , 
1996
, vol. 
10
 (pg. 
262
-
278
)
Johnson
J. K.
Lui
L. Y.
Yaffe
K.
Executive function, more than global cognition, predicts functional decline and mortality in elderly women
Journal of Gerontology: Medical Sciences
 , 
2007
, vol. 
62A
 (pg. 
1134
-
1141
)
Kaleita
T. A.
Wellisch
D. K.
Cloughesy
T. F.
Ford
J. M.
Freeman
D.
Belin
T. R.
Goldman
J.
Prediction of neurocognitive outcome in adult brain tumor patients
Journal of Neuro-Oncology
 , 
2004
, vol. 
67
 (pg. 
245
-
253
)
Kennedy
K. J.
Age effects on Trail Making Test performance
Perceptual and Motor Skills
 , 
1981
, vol. 
52
 (pg. 
671
-
675
)
Lamberty
G. J. P.
Chatel
S. H.
Bieliauskas
D. M.
Linas
A.
Derived Trail Making Test indices: A preliminary report
Neuropsychiatry, Neuropsychology, and Behavioral Neurology
 , 
1994
, vol. 
7
 (pg. 
230
-
234
)
Lange
R. T.
Iverson
G. L.
Zakrzewski
M. J.
Ethel-King
P. E.
Franzen
M. D.
Interpreting the trail making test following traumatic brain injury: comparison of traditional time scores and derived indices
Journal of Clinical and Experimental Neuropsychology
 , 
2005
, vol. 
27
 (pg. 
897
-
906
)
Larrabee
G. J.
Millis
S. R.
Meyers
J. E.
Sensitivity to brain dysfunction of the Halstead-Reitan vs. an ability-focused neuropsychological battery
The Clinical Neuropsychologist
 , 
2008
, vol. 
22
 (pg. 
813
-
825
)
Lemiere
J.
Decruyenaere
M.
Evers-Kiebooms
H.
Vandenbussche
E.
Dom
R.
Cognitive changes in patients with Huntington's disease (HD) and asymptomatic carriers of the HD mutation: A longitudinal follow-up study
Journal of Neurology
 , 
2004
, vol. 
251
 (pg. 
935
-
942
)
Lezak
M. D.
Howieson
D. B.
Loring
D. W.
Neuropsychological assessment
 , 
2004
4th ed.
New York
Oxford University Press
Lucas
J. A.
Ivnik
R. J.
Smith
G. E.
Ferman
T. J.
Willis
F. B.
Petersen
R. C.
, et al.  . 
Mayo's Oder African Americans Normative Studies: norms for Boston Naming Test, Controlled Oral Word Association, Category Fluency, Animal Naming, Token Test, Wrat-3 Reading, Trail Making Test, Stroop Test, and Judgment of Line Orientation
The Clinical Neuropsychologist
 , 
2005
, vol. 
19
 (pg. 
243
-
269
)
McDonald
C. R.
Delis
D. C.
Norman
M. A.
Tecoma
E. S.
Iragui-Madozi
V. I.
Is impairment in set-shifting specific to frontal-lobe dysfunction? Evidence from patients with frontal-lobe or temporal-lobe epilepsy
Journal of the International Neuropsychological Society
 , 
2005
, vol. 
11
 (pg. 
477
-
481
)
Mitchell
M.
Miller
S.
Prediction of functional status in older adults: The ecological validity of four Delis–Kaplan executive function system tests
Journal of Clinical and Experimental Neuropsychology
 , 
2008
, vol. 
30
 (pg. 
683
-
690
)
Mitrushina
M. N.
Boone
K. B.
D'Elia
I. F.
Handbook Of Normative Data For Neuropsychological Assessment.
 , 
1999
New York
Oxford University Press
O'Rourke
J. J.
Beglinger
L. J.
Smith
M. M.
Mills
J.
Moser
D. J.
Rowe
K. C.
, et al.  . 
The Trail Making Test in prodromal Huntington disease: contributions of disease progression to test performance
Journal of Clinical and Experimental Neuropsychology
 , 
2011
, vol. 
33
 (pg. 
567
-
579
)
Pena-Casanova
J.
Quinones-Ubeda
S.
Quintana-Aparicio
M.
Aguilar
M.
Badenes
D.
Molinuevo
J. L.
, et al.  . 
Spanish multicenter normative studies (NEURONORMA Project): Norms for verbal span, visuospatial span, letter and number sequencing, trail making test, and symbol digit modalities test
Archives of Clinical Neuropsychology
 , 
2009
, vol. 
24
 (pg. 
321
-
341
)
Potter
G. G.
Wagner
H. R.
Burke
J. R.
Plassman
B. L.
Welsh-Bohmer
K. A.
Steffens
D. C.
Neuropsychological predictors of dementia in late-life major depressive disorder
American Journal of Geriatric Psychiatry
  
Rabin
L. A.
Barr
W. B.
Burton
L. A.
Assessment practices of clinical neuropsychologists in the United States and Canada: A survey of INS, NAN, and APA Division 40 members
Archives of Clinical Neuropsychology
 , 
2005
, vol. 
20
 (pg. 
33
-
65
)
Reitan
R. M.
The validity of the Trail Making Test as an indicator of organic brain damage
Perceptual and Motor Skills
 , 
1958
, vol. 
8
 (pg. 
271
-
276
)
Reitan
R. M.
Wolfson
D.
The Halstead–Reitan Neuropsychological Test Battery: Theory and clinical interpretation
 , 
1993
2nd ed.
Tucson, AZ
Neuropsychology Press
Salthouse
T. A.
Toth
J.
Daniels
K.
Parks
C.
Pak
R.
Wolbrette
M.
, et al.  . 
Effects of aging on efficiency of task switching in a variant of the Trail Making Test
Neuropsychology
 , 
2000
, vol. 
14
 (pg. 
101
-
111
)
Sanchez-Cubillo
I.
Perianez
J. A.
Adrover-Roig
D.
Rodriguez-Sanchez
J. M.
Rios-Lago
M.
Tirapu
J.
, et al.  . 
Construct validity of the Trail Making Test: Role of task-switching, working memory, inhibition/interference control, and visuomotor abilities
Journal of the International Neuropsychological Society
 , 
2009
, vol. 
15
 (pg. 
438
-
450
)
Seo
E. H.
Lee
D. Y.
Kim
K. W.
Lee
J. H.
Jhoo
J. H.
Youn
J. C.
, et al.  . 
A normative study of the Trail Making Test in Korean elders
International Journal of Geriatric Psychiatry
 , 
2006
, vol. 
21
 (pg. 
844
-
852
)
Steinberg
B. A.
Bieliauskas
L. A.
Smith
G. E.
Ivnik
R. J.
Mayo's older Americans normative studies: age- and IQ-adjusted norms for the Trail Making Test, the stroop test, and MAE controlled oral word association test
The Clinical Neuropsychologist
 , 
2005
, vol. 
19
 (pg. 
329
-
377
)
Strauss
E.
Sherman
E. M. S.
Spreen
O.
A compendium of neuropsychological tests: Administration, norms, and commentary
 , 
2006
3rd ed.
New York
Oxford University Press
Stuss
D.
Bisschop
M.
Alexander
M.
Levine
B.
Katz
D.
The Trail Making Test: A study in focal lesion patients
Psychological Assessment
 , 
2001
, vol. 
13
 (pg. 
230
-
239
)
Tombaugh
T. N.
Trail Making Test A and B: Normative data stratified by age and education
Archives of Clinical Neuropsychology
 , 
2004
, vol. 
19
 (pg. 
203
-
214
)
Vazzana
R.
Bandinelli
S.
Lauretani
F.
Volpato
S.
Lauretani
F.
Di Iorio
A.
, et al.  . 
Trail Making Test predicts physical impairment and mortality in older persons
Journal of the American Geriatrics Society
 , 
2010
, vol. 
58
 (pg. 
719
-
723
)
Wiberg
B.
Kilander
L.
Sundstro
J.
Byberg
L.
Lind
L.
The relationship between executive dysfunction and post-stroke mortality: a population-based cohort study
BMJ Open
 , 
2012
, vol. 
2
  
pii: e000458
Zalonis
I.
Kararizou
E.
Triantafyllou
N. I.
Kapaki
E.
Papageorgiou
S.
Sgouropoulos
P.
, et al.  . 
A Normative Study of the Trail Making Test A and B in Greek Adults
The Clinical Neuropsychologist
 , 
2008
, vol. 
22
 (pg. 
842
-
850
)