Abstract

The aim was to verify the applicability of Reitan and Wolfson's proposed neuropsychological screening battery for adults (2006, 2008) in the Czech population. The sample consisted of 70 participants aged 19–65 years, all of whom were examined using a screening method as well as the full Halstead–Reitan neuropsychological battery (HRNB). The correlation, logistic regression, ROC curve analysis, sensitivity and specificity, and positive and negative predictive values were all calculated. The Pearson correlation between the screening scale of neuropsychological deficit and the General Neuropsychological Deficit Scale (GNDS) from HRNB was 0.78 (p < .001). When optimal cut-off scores of 8 were utilized (in accordance with Horwitz, Lynch, McCaffrey, & Fisher in Screening for neuropsychological impairment using Reitan and Wolfsońs preliminary neuropsychological test battery. Archives of Clinical Neuropsychology, 23, 393–398, 2008, but different from Reitan, & Wolfson in The use of serial testing in evaluating the need for comprehensive neuropsychological testing of adults. Applied Neuropsychology, 15, 21–32, 2008), 78.6% of individuals were correctly classified having neuropsychological impairment or no impairment according to the GNDS. Our results confirm that this neuropsychological screening battery has good psychometric properties in the Czech population.

We have been using the Czech translation of the entire HRNB since 1988 (Preiss & Hynek, 1991), and have completed over 520 assessments using the entire HRNB. We published research related to the HRNB in schizophrenia, including neuropsychological performance in schizophrenic twins, white matter density, and pursuit eye movements (Preiss, Dvořáková, Zvárová, & Hynek, 1992; Preiss, Hynek, Dvořáková, & Zvárová, 1993; Preiss, Bohm, Hynek, Dvořáková, & Zvárová, 1996; Preiss, Hynek, Böhm, & Zvárová, 2004a). With regards to epilepsy, we have also published research on the effects of antiepileptics on individuals with epilepsy (Kolínová & Preiss, 1991), the relationship between nonverbal tests of spatial cognition and standard neuropsychological tests in those with hippocampus lesions (Preiss, Kalová, Štěpánková, Bureš, & Vlček, 2004b), and the impact of surgical resection for refractory temporal lobe epilepsy on HRNB results (Preiss, 1999; Preiss & Vojtěch, 2010b). The entire battery or portions of it was also used in work with patients with chronic or acute intoxication of toxic materials: toluene, mercury, and dioxin (for a review, see Preiss, 2009).

Although we hold the HRNB in high esteem, it demands a great deal of time and financial resources, thus is not optimal for routine use in daily practice. In everyday clinical neuropsychological practice, the clinician needs a screening battery with both well-documented validity and high predictive value, which can guide determination of which individuals are in need of more detailed and time-demanding neuropsychological assessment. We have begun work on the Screening Battery of the Prague Psychiatric Center (Preiss, Rodriguez, Kawaciuková, & Laing, 2007), which includes the Trail Making Test from the Halstead–Reitan Neuropsychological Battery (HRNB).

First in a lecture (2006) and later in three articles (2008a, 2008b, 2008c), Reitan and Wolfson introduced three screening tests of the widely used Halstead–Reitan Neuropsychological Battery (HRNB), for three different age categories (Reitan & Wolfson, 1985, 1993).

The screening battery for adults (SBNDS) uses seven variables from the four subtests of the HRNB: Trail Making Test-Part A (TMT-A); Trail Making Test-Part B (TMT-B); Tactile Form Recognition Test (TFRT); Finger Tapping Test (FTT). Reitan and Wolfson (2008a) found a correlation between GNDS, derived from HRNB, and the screening scale for neuropsychological deficit: 0.78 for controls (n = 50) and 0.81 for a neurologically impaired group (n = 50; individuals in this group demonstrated definitive evidence of neuropsychological impairment independent of the HRNB.

The screening battery for older children is based on the Halstead–Reitan Neuropsychological Test Battery for Older Children (HRNB-OC; Reitan & Wolfson, 2008b). The screening battery for younger children is based on the Reitan–Indiana Neuropsychological Test Battery for Younger Children (RINTB; Reitan & Wolfson, 2008c). The screening version for adults (SBNDS, Screening Battery for Neuropsychological Deficit Scale) was tested by Horwitz, Lynch, McCaffrey and Fisher (2008). These authors recommended some minor changes but were basically supportive of the idea. The correlation between GNDS as derived from HRNB, and the screening scale for neuropsychological deficit, was found to be 0.82, p ≤ .01 (n = 69). The AUC value (area under curve) was 0.882 (SE = 0.041). With a cut-off point of 9, 78.6% of normal individuals and 82.9% of neuropsychologically impaired individuals were correctly classified (Horwitz et al, 2008). The screening battery had excellent specificity and fair sensitivity for identification of individuals with neuropsychological deficit on the HRNB. Using a cut-off score of 8, their results were comparable with the findings of Reitan and Wolfson.

Vanderslice-Barr, Lynch and McCaffrey (2008) verified the suitability of two different screening batteries for older and younger children. They found a correlation between the older children’s GNDS, derived from HRNB-OC, and the screening scale of 0.87, p ≤ .01 (n = 40). The AUC value was 0.833. They also found a correlation between the younger children's GNDS, derived from Reitan–Indiana Neuropsychological Test Battery for Younger Children, and a screening scale of 0.92, p ≤ .01 (n = 28).

Reitan and Wolfson suggested a two-stage procedure: the first uses a single test that requires minimal administration time and is used to identify subjects who should be assessed with the full screening battery; the second identifies individuals who should have an overall neuropsychological examination. Both verification studies (Horwitz, et al., 2008; Vanderslice-Barr et al., 2008) omit the individual test phase. We follow their procedure in this work.

Aim

The aim of this work was to verify the applicability of the SBNDS screening method for adults in predicting neuropsychological impairment in the Czech population.

Method

Procedure

The study was conducted using archival data. All subjects were examined using the full Halstead–Reitan neuropsychological battery. The screening information was extracted from the full HRNB protocol. General Neuropsychological Deficit Scale (GNDS) scores were computed for each participant using 42 variables from the HRNB (Reitan & Wolfson, 1993). Total Neuropsychological Deficit Scale (NDS) scores for the screening battery (SBNDS) were then computed for each participant by converting raw scores from seven variables from the four subtests. These seven variables are: time on the Trail Making Test-Part A (TMT-A); time on the Trail Making Test-Part B (TMT-B); total time and right/left differences on the Tactile Form Recognition Test (TFRT); and the number of dominant hand taps, number of nondominant hand taps, and right/left differences on the Finger Tapping Test (FTT). For each of these variables, raw scores were converted into scores ranging from 0 (normal range) to 3 (severe impairment), see Reitan and Wolfson (1993, 2006, 2008a) for a complete description.

Statistical methods

Procedures for statistical analyses followed those in previous work (Horwitz et al., 2008), including Pearson correlations, logistical regression, ROC curve analysis, sensitivity and specificity, determination of optimal cut-off scores, positive predictive power, and negative predictive power.

Methodology obtained from the University of Chicago (www.metz-ROC.edu) was used to analyze the ROC curve. ROC methodology is a part of a theory of detecting signals from noise, which was developed in the 1940s during the Second World War (Pintea & Moldovan, 2009). It gives the contrast ratio of false-positive and -negative assessments. A higher AUC value (area under curve) is associated with a more accurate classification (e.g., Bewick, Cheek, & Ball, 2004; Gallop, Crits-Christoph, Muenz, & Tu, 2003; Sideridis, Morgan, Botsas, Padeliadu, & Fuchs, 2006). An AUC value between 90 and 99 is considered “excellent,” 80–89 “good,” 70–79 “fair,” 60–69 “poor,” and values <60 represent random classification accuracy (Sideridis et al., 2006). In the context of the current study, the AUC calculation may potentially underestimate the actual AUC due to the use of conservative, non-parametric methods (Horwitz et al., 2008). The AUC is sometimes regarded as the most useful index of the ability of a screening battery to discriminate between positive and negative screening results (Mari & Williams, 1986).

The predictive utility of the screening battery was determined by calculating the sensitivity (the higher the probability that the test results will be positive in impaired patients, the higher the sensitivity and the fewer false negative findings), specificity (the greater the likelihood that the test will be negative in people without impairment, the higher the specificity and the fewer false-positive findings), positive predictive value (PPV, the probability that a person is truly impaired when the test is positive), and negative predictive values (NPV, the probability that the person does not have the disease when there is a negative test result).

Participants

The sample consisted of 70 participants, of whom 69% were men and 91% were right-handed. Their ages ranged from 19 to 65 (mean age 36.3, SD = 11.6). Educational attainment ranged from 7 to 15 years (mean 11.6, SD = 1.6). Overall IQ according to the WAIS-R ranged from 70 to 131 (mean 90.6, SD = 12.6).

The sample consisted of 30 individuals with pharmaco-resistant epilepsy who were being examined prior to a neurosurgical operation, 16 persons with neurotoxic damage (of which 15 had massive exposure to dioxin, heavily intoxicated and with long-term consequences (e.g., Pelclová et al., 2001, 2009; Preiss, Pelclová, Fenclová, & Urban, 2010a), and one to mercury (Pelclová et al., 2002), as well as 24 persons being examined for possible epilepsy (which was confirmed in all cases, but who were not recommended for neurosurgery due to medical reasons).

All participants were hospitalized, in either the Neurological Department Na Homolce Hospital, Prague, Czech Republic (primary site) or the Department of Occupational Medicine, 1st Medical Faculty, Charles University and General Faculty Hospital, Prague, Czech Republic.

Results

The Pearson correlation between the scales of neuropsychological impairment according to the HRNB and the screening battery was 0.777 (p < .001). When using logistic regression to determine how well the screening battery scores (used as a continuous variable) correctly classified persons without neuropsychological impairment (GNDS scores of 25 or less, as suggested by Reitan & Wolfson, 1993) or as neuropsychologically impaired (GNDS ≥ 26), it correctly classified 92.3% as normal (i.e., without neuropsychological impairment) and 73.7% as neuropsychologically impaired persons. Overall, 77.1% of participants were classified correctly.

The performance of individuals was first classified using GNDS values of 26 and higher (of the total battery HRNB) as an indicator of neuropsychological impairment, and SBNDS scores of 9 and above (from the screening battery) as suggestive of suspected neuropsychologic impairment, in accordance with the proposal of Reitan & Wolfson (2008a). With the use of these cut-off values, 17% of individuals were correctly classified as nonimpaired (normal GNDS and SBNDS scores, n = 12), 60% as impaired (according to GNDS and SBNDS, n = 42), 21% as false negatives (impaired according to GNDS and normal SBNDS scores, n = 15), and 1% as false positives (normal GNDS and weakened according to SBNDS, n = 1). See Table 1 for an overview.

Table 1.

Matrix of true, false positives and negatives

 Results of the screening battery
 
Impaired Normal Total true state 
True state 
 Impaired 42 (TP) 15 (FN) 57 
 Normal 1 (FP) 12 (TN) 13 
Overall test results 43 27 70 
 Results of the screening battery
 
Impaired Normal Total true state 
True state 
 Impaired 42 (TP) 15 (FN) 57 
 Normal 1 (FP) 12 (TN) 13 
Overall test results 43 27 70 

Note: Values of the screening battery using a cut-off score of ≥9, and ≥26 on the whole HRNB according to GNDS. TP = true positives; FN = false negatives; FP = false positives; TN = true negatives.

Like Horwitz and colleagues (2008), we were concerned about the high number of false negatives and compared these cases by level of impairment (normal range, mild impairment, moderate impairment, severe impairment), as indicated by the GNDS score (Reitan & Wolfson, 1993). Of the 15 false negatives, 10 fell in the mildly impaired range (GNDS score of 26–40), 5 fell in the moderately impaired range (GNDS score of 41–67), and none fell in the severely impaired range (GNDS score of 68 or greater; see Table 2). Adjusting the cut-off score to 8 resulted in 3 fewer false negatives, including 1 of the moderately impaired cases.

Table 2.

Frequency distribution of cases in each of four impairment categories

GNDS score Level of impairment Number of cases
 
Total 
+ on screening batterya − on screening batterya 
≥68 Severe 
41–67 Moderate 25 5b 30 
26–40 Mild 15 10b 25 
0–25 Normal range 1c 12 13 
Total  43 27 70 
GNDS score Level of impairment Number of cases
 
Total 
+ on screening batterya − on screening batterya 
≥68 Severe 
41–67 Moderate 25 5b 30 
26–40 Mild 15 10b 25 
0–25 Normal range 1c 12 13 
Total  43 27 70 

ªUsing a cut-off score of 9.

bRepresents false negatives.

cRepresents false positives.

When using a cut-off score of 8 for the screening battery, in accordance with Horwitz and colleagues (2008), the results were somewhat improved, so that 14.3% of individuals were correctly classified as negatives (n = 10), 64% correctly as positives (n = 45), 17.5% as false negatives (n = 12), and 4.3% as false positives (n = 3). Overall, 78.6% of participants were classified correctly.

Furthermore, an ROC curve (receiver-operating characteristic) was developed (see Fig. 1). Diagonal lines on the graph indicate a random classification, while the ROC curve shows the correct classification. The greater the area under the curve, the higher the accuracy of the classification. The AUC value was 0.884 (SE = 0.039).

Fig. 1.

ROC curve calculated for the screening battery (n = 70, cut-off = 9, AUC = 0.884). The diagonal line on the graph shows random classification, the ROC curve shows the correct classification (y-axis = sensitivity, x-axis = (1 − specificity)

Fig. 1.

ROC curve calculated for the screening battery (n = 70, cut-off = 9, AUC = 0.884). The diagonal line on the graph shows random classification, the ROC curve shows the correct classification (y-axis = sensitivity, x-axis = (1 − specificity)

For calculations of sensitivity and specificity, a cut-off point of 9, as recommended by Reitan and Wolfson, was first used. Sensitivity was 74%, and specificity was 92%. We also found values for a cut-off point of 8, which was recommended as more useful by Horwitz and colleagues (2008). Sensitivity was 79%, and specificity was 77%. In Table 3 we show the values for cut-off points ranging from 3 to 10 (see Table 3).

Table 3.

Sensitivity, specificity, and predictive values of the screening battery in the sample according to various cut-off points

NDS cut-off Sensitivity Specificity 1 − specificity Positive predictive value Negative predictive value 
≥3 1.000 0.154 0.846 0.838 1.000 
≥4 0.982 0.231 0.769 0.848 0.750 
≥5 0.930 0.385 0.615 0.869 0.556 
≥6 0.895 0.615 0.385 0.911 0.571 
≥7 0.842 0.692 0.308 0.923 0.500 
≥8 0.789 0.769 0.231 0.938 0.455 
≥9 0.737 0.923 0.077 0.977 0.444 
≥10 0.632 1.000 0.000 1.000 0.382 
NDS cut-off Sensitivity Specificity 1 − specificity Positive predictive value Negative predictive value 
≥3 1.000 0.154 0.846 0.838 1.000 
≥4 0.982 0.231 0.769 0.848 0.750 
≥5 0.930 0.385 0.615 0.869 0.556 
≥6 0.895 0.615 0.385 0.911 0.571 
≥7 0.842 0.692 0.308 0.923 0.500 
≥8 0.789 0.769 0.231 0.938 0.455 
≥9 0.737 0.923 0.077 0.977 0.444 
≥10 0.632 1.000 0.000 1.000 0.382 

For the accuracy of classification according to individual cut-off values, see Table 4.

Table 4.

The accuracy of the screening battery classification in the sample according to different cut-off values

NDS cut-off Normal individuals (GNDS ≤ 25), % Individuals with neuropsych. impairment (GNDS ≥ 26), % Total, % 
≥3 15.4 100.0 84.3 
≥4 23.1 98.2 84.3 
≥5 38.5 93.0 82.9 
≥6 61.5 89.5 84.3 
≥7 69.2 84.2 81.4 
≥8 76.9 78.9 78.6 
≥9 92.3 73.7 77.1 
≥10 100.0 63.2 70.0 
NDS cut-off Normal individuals (GNDS ≤ 25), % Individuals with neuropsych. impairment (GNDS ≥ 26), % Total, % 
≥3 15.4 100.0 84.3 
≥4 23.1 98.2 84.3 
≥5 38.5 93.0 82.9 
≥6 61.5 89.5 84.3 
≥7 69.2 84.2 81.4 
≥8 76.9 78.9 78.6 
≥9 92.3 73.7 77.1 
≥10 100.0 63.2 70.0 

Discussion

Above all, the results largely confirm that this screening method demonstrates good psychometric properties in a sample of Czech participants. The Pearson correlation between GNDS and the screening scale for neuropsychological deficit (Reitan & Wolfson 0.78 and 0.81, Horwitz et al. 0.82, in current sample 0.78), the AUC values (Horwitz et al. 0.882, current sample 0.884) and overall correct classification (Horwitz et al. 81.2%, current sample 78.6%) as well as the sensitivity (Reitan & Wolfson 88%, and using a cut-off score of 8, Horwitz et al. 83% and the current sample 79%) and specificity (Reitan & Wolfson 96%, and using a cut-off score of 8, Horwitz et al. 79% and the Czech sample 77%) were all similar to Reitan and Wolfson (2008a) and Horwitz et al. (2008). For the most part, the values were highest in the Reitan and Wolfson battery (2008a) and the lowest in the current sample. However, the differences were not significant.

An exception to this was the negative predictive validity (NVP). In the current sample, NVP was very low (0.444, when using cut-off of 9, and 0.455 with a cut-off of 8; in contrast, prior research has demonstrated NVP of 0.692 with a cut-off of 9 and 0.759 with a cut-off of 8; Horwitz et al., 2008), meaning that there is not a high probability that an individual identified as neuropsychologically “normal” according to the screening battery would demonstrate no neuropsychological impairment on the full HRNB.

The explanation for this may lie in the sample composition. There are not substantial differences in average age (Reitan and Wolfson: 36.7; Horwitz et al.: 39.6; Czech sample: 36.3). There are also not significant differences in gender (Horwitz et al.: 62% men, Czech sample: 69% men) or educational attainment (in the Reitan and Wolfson control sample the average was 12.78 years and in the sample with neuropsychological deficit the average was 12.86; Horwitz et al., 13.3; in the Czech sample it was somewhat lower, 11.6). The percentage of right-handed persons is also similar (Horwitz et al.: 88%; Czech sample: 91%).

However, there are statistically significant differences in the number of individuals with neuropsychological impairment in the different sample populations. While in Reitan & Wolfson those with neuropsychological impairment account for 50% of the sample, in Horwitz et al. it is 59.4%, and in Vanderslice-Barr and colleagues (2008) it is 40% in older children and 11% in younger children. In the current study, 81.4% of the sample demonstrates neuropsychological impairment according to the HRNB. Under these circumstances, the results of the screening battery do not invalidate its predictive validity. In contrast, they are in agreement with the expected results for the different prevalence rates of impairment (e.g., Horwitz et al., 2008). While sensitivity and specificity are characteristics of the actual test, predictive values are strongly dependent upon the prevalence of the disease.

We believe that the neuropsychological screening battery for adults was validated in our sample group of patients with epilepsy and toxic exposure. The observed low NPV is solely due to the high prevalence of the disease (81.4%) in the sample. If we were to recalculate the results based on a sample with 50% neuropsychogical impairment or weakening, the NPVs of 0.778 with a cut-off of 9 and 0.785 with a cut-off of 8 would be satisfactory. A lower prevalence will always result in a raised NPV.

The samples also differed in their diagnostic composition. It is possible that the patient population under study is the source of observed low NPVs, due to differential effects of the type of neurological disorder on neuropsychological performance. For example, patients with epilepsy or toxic exposure have very different HRBN patterns from patients with strokes or tumors. The ability of a screening battery to effectively identify the impairments in these populations will be highly dependent on the types of neuropsychological functions measured. If the screening measures accurately reflect common impairments, the predictive values will be high. If, however, the neurological disorder does not show the same common impairments, the predictive values will be low. This underlies the importance of the full HRNB, as different neurological disorders produce different HRNB patterns of impairment. In light of our findings, different cut-offs may be of more or less utility for use in different clinical populations. This issue may be explored further in future studies comparing samples with different diagnoses.

In our study, the sample consisted of patients with epilepsy and toxic exposure. For both groups there are screening methods for these specific conditions. For instance, for epilepsy there exists the Neuropsychological Battery for Epilepsy (NBE) (Dodrill, 1978) and EpiTrack (Lütz & Helmstaedter, 2005). Information about the numerous methods used for neurotoxicology is available (e.g., Lezak, Howieson, & Loring, 2004). In the future it might be useful to compare these targeted screening methods with Reitan and Wolfson's neuropsychological screening battery. On the other hand, in prior research on the screening battery being discussed (Reitan & Wolfson, 2008a), the brain-damaged group included a diverse representation of types and locations of brain disease or damage. Among the 50 participants, there were 2 patients with major motor epilepsy, 2 patients with complex partial epilepsy, and 1 patient with carbon monoxide poisoning. In a study done by Horwitz and colleagues (2008), the participants were initially evaluated in a private practice setting while pursuing litigation due to head trauma, and some were excluded for failing the Test of Memory Malingering (TOMM).

Similar to Horwitz and colleagues (2008), a cut-off of 8 demonstrated better sensitivity and specificity for the Czech sample (in which the sensitivity was 0.789 and specificity 0.769). However, for the Czech sample it would also be acceptable to use a cut-off value of 7 (sensitivity 0.842, specificity 0.692).

The use of patients with questionable neurological/medical diagnoses could raise the issue of the effect of response validity across multiple measures in a testing battery (Fox, 2011). In other words, one cannot expect a consistent pattern of poor effort across all test measures. This factor could be an important limitation in the present study. The high false-negative rate could in fact reflect a non-neurological condition of some (if not all) of the patients with “suspected neurotoxic damage.” This may require further investigation.

However, when we looked at the patient makeup of the false-negative group, we found that there was not a higher occurrence of participants from the “neurotoxic” group (2 of 16, or 12.5%) when compared with patients prior to neurosurgery (8 of 30, or 27%), nor when compared with participants from a group of possible (and confirmed) epilepsy cases (5 of 24, or 21%).

Conclusions

The results of this Czech sample confirm the good psychometric properties of the neuropsychological screening battery SBNDS for adults (Screening Battery for Neuropsychological Deficit Scale, Reitan & Wolfson, 2006, 2008a). Original literature regarding the procedure of the screening recommends that after identifying individuals with a positive suspected neuropsychological impairment, the next step would be the use of the entire HRNB, which we also consider most appropriate. In further studies it would be useful to regularly utilize effort testing, ideally utilizing tests of symptom validity. Above all, it will be necessary to concentrate research on the areas for which screening methods are most useful. For example, further research should focus on populations within which the prevalence of neuropsychological impairment is significantly lower than in the sample discussed in this paper. A limitation of the study is the high prevalence of individuals with neuropsychological impairment in the presented sample and the combination of participants with different diagnoses and potentially inconsistent effort across all test measures.

Funding

This work has been supported by MH CZ - DRO (PCP, 00023752).

Conflict of Interest

None declared.

References

Bewick
V.
Cheek
L.
Ball
J.
Statistics review 13: Receiver operating characteristic curves
Critical Care
 , 
2004
, vol. 
8
 (pg. 
508
-
512
)
Dodrill
C. B.
A neuropsychological battery for epilepsy
Epilepsia
 , 
1978
, vol. 
19
 (pg. 
611
-
623
)
Fox
D. D.
Symptom validity test failure indicates invalidity of neuropsychological tests
The Clinical Neuropsychologist
 , 
2011
, vol. 
25
 (pg. 
488
-
495
)
Gallop
R. J.
Crits-Christoph
P.
Muenz
L. R.
Tu
X. M.
Determination and interpretation of the optimal operating point for ROC curves derived through generalized linear models
Understanding Statistics
 , 
2003
, vol. 
2
 (pg. 
219
-
242
)
Horwitz
J. E.
Lynch
J. K.
McCaffrey
R. J.
Fisher
J. M.
Screening for neuropsychological impairment using Reitan and Wolfson’s preliminary neuropsychological test battery
Archives of Clinical Neuropsychology
 , 
2008
, vol. 
23
 (pg. 
393
-
398
)
Kolínová
M.
Preiss
J.
Therapy and antiepileptic drug monitoring
 , 
1991
 
5th Epileptologic days, Abstracts, Prague, September 8–11 Prague: Electric Medical Systems
Lezak
M. D.
Howieson
D. B.
Loring
D. W.
Neuropsychological assessment
 , 
2004
4th ed.
New York
Oxford University Press
Lütz
M.
Helmstaedter
C.
EpiTrack: Tracking cognitive side effects of medication on attention and executive functions in patients with epilepsy
Epilepsy & Behavior
 , 
2005
, vol. 
12
 (pg. 
708
-
714
)
Mari
J. J.
Williams
P.
Misclassification by psychiatric screening questionnaires
Journal of Chronic Diseases
 , 
1986
, vol. 
39
 (pg. 
371
-
378
)
Pelclová
D.
Fenclová
Z.
Dlasková
Z.
Urban
P.
Lukáš
E.
Procházka
B.
, et al.  . 
Biochemical, neuropsychological, and neurological abnormalities after 2,3,7,8-TCDD exposure
Archives of Environmental Health
 , 
2001
, vol. 
56
 (pg. 
493
-
500
)
Pelclová
D.
Fenclová
Z.
Urban
P.
Ridzoň
P.
Preiss
J.
Dubská
Z.
, et al.  . 
Chronic health impairment due to 2,3,7,8- tetrachloro-dibenzo-p-dioxin exposure
Neuroendocrinology Letters
 , 
2009
, vol. 
30
 
Suppl. 1
(pg. 
219
-
224
)
Pelclová
D.
Lukáš
E.
Urban
P.
Preiss
J.
Ryšavá
R.
Lebenhart
P.
, et al.  . 
Mercury intoxication from skin ointment containing mercury ammonium chloride
International Archives of Occupational and Environmental Health
 , 
2002
, vol. 
75
 
Suppl.
(pg. 
S54
-
S59
)
Pintea
S.
Moldovan
R.
The receiver-operating characteristic (ROC) analysis: Fundamentals and applications in clinical psychology
Journal of Cognitive and Behavioral Psychotherapies
 , 
2009
, vol. 
9
 (pg. 
49
-
66
)
Preiss
J.
Preoperative neuropsychological examination and evaluation of surgical results
Epilepsia
 , 
1999
, vol. 
40
 
Suppl. 2
pg. 
84
 
Preiss
J.
Halsteadova-Reitanova neuropsychologická baterie (HRNB)
Československá psychologie
 , 
2009
, vol. 
53
 (pg. 
362
-
375
)
Preiss
J.
Bohm
J.
Hynek
K.
Dvořáková
M.
Zvárová
J.
Korelace neuropsychologických zkoušek s denzitou bílé hmoty mozku u schizofrenie
Česká a slovenská psychiatrie
 , 
1996
, vol. 
92
 (pg. 
3
-
17
)
Preiss
J.
Dvořáková
M.
Zvárová
J.
Hynek
K.
Neuropsychologická výkonnost schizofrenních dvojčat
Československá psychologie
 , 
1992
, vol. 
36
 (pg. 
257
-
266
)
Preiss
J.
Hynek
K.
Neuropsychologická baterie Halstead-Reitan. První zkušenosti s českým převodem
Československá psychiatrie
 , 
1991
, vol. 
87
 (pg. 
249
-
254
)
Preiss
J.
Hynek
K.
Böhm
J.
Zvárová
J.
Neuropsychological study of schizophrenic twins
International Journal of Psychology
 , 
2004a
, vol. 
39
 (pg. 
127
-
128
)
Preiss
J.
Hynek
K.
Dvořáková
M.
Zvárová
J.
Neuropsychologické testy a plynulé sledovací oční pohyby u schizofrenních dvojčat
Česko-slovenská psychiatrie
 , 
1993
, vol. 
89
 (pg. 
276
-
286
)
Preiss
J.
Kalová
E.
Štěpánková
K.
Bureš
J.
Vlček
K.
Metody měření prostorové paměti a jejich vztah k standardním neuropsychologickým zkouškám u osob s lézemi hippokampu
Česká a slovenská neurologie a neurochirurgie
 , 
2004b
, vol. 
67/100
 (pg. 
335
-
346
)
Preiss
J.
Pelclová
D.
Fenclová
Z.
Urban
P.
Neuropsychologie a neurotoxicita při dlouhodobém sledování účinků dioxinu
Československá psychologie
 , 
2010a
, vol. 
54
 (pg. 
525
-
545
)
Preiss
M.
Rodriguez
M.
Kawaciuková
R.
Laing
H.
Screeningová baterie PCP
 , 
2007
3rd ed.
Praha
Psychiatrické centrum Praha
Preiss
J.
Vojtěch
Z.
Resekční operace pro farmakorezistentní temporální epilepsii a Halsteadova-Reitanova neuropsychologická baterie (HRNB)
Česká a slovenská psychiatrie
 , 
2010b
, vol. 
106
 (pg. 
150
-
156
)
Reitan
R. M.
Wolfson
D.
The Halstead-Reitan Neuropsychological Test Battery. Theory and clinical interpretation
 , 
1985, 1993
2nd ed.
Tucson, AZ
Neuropsychology Press
Reitan
R. M.
Wolfson
D.
Issues in neuropsychology: Where have we succeeded? Where have we failed? What are the clinical and forensic implications?
2006
Presented at the 26th annual conference of the National Academy of Neuropsychology
San Antonio, TX
Reitan
R. M.
Wolfson
D.
The use of serial testing in evaluating the need for comprehensive neuropsychological testing of adults
Applied Neuropsychology
 , 
2008a
, vol. 
15
 (pg. 
21
-
32
)
Reitan
R. M.
Wolfson
D.
Serial testing of older children as a basis for recommending comprehensive ceuropsychological evaluation
Applied Neuropsychology
 , 
2008b
, vol. 
15
 (pg. 
11
-
20
)
Reitan
R. M.
Wolfson
D.
The use of serial testing to identify young children in need of comprehensive neuropsychological evaluation
Applied Neuropsychology
 , 
2008c
, vol. 
15
 (pg. 
1
-
10
)
Sideridis
G. D.
Morgan
P. L.
Botsas
G.
Padeliadu
S.
Fuchs
D.
Predicting LD on the basis of motivation, metacognition, and psychopathology: An ROC analysis
Journal of Learning Disabilities
 , 
2006
, vol. 
39
 (pg. 
215
-
229
)
Vanderslice-Barr
J. L.
Lynch
J. K.
McCaffrey
R. J.
Screening for neuropsychological impairment in children using Reitan and Wolfson’s preliminary neuropsychological battery
Archives of Clinical Neuropsychology
 , 
2008
, vol. 
23
 (pg. 
243
-
249
)