Mechanisms of Ageing and Development 127 (2006) 123–128

www.elsevier.com/locate/mechagedev

Early detection of dementia in clinical practice

S. Giaquinto a,*, L. Parnetti b
a
Department of Neuromotor Rehabilitation, IRCCS San Raffaele Pisana, via della Pisana 235, 00163 Rome, Italy
b
Department of Neurology, University of Perugia, Italy

Available online 10 November 2005

Early identification of dementia and possible risk factors is a critical issue. The Basic Italian Cognitive Questionnaire (BICQ) is designed as a
routine instrument for screening patients with initial cognitive impairment in daily practice. It is devoted to the general physician who needs a tool
for deciding whether a subject deserves further diagnostic investigation in specialised centres. The administration of BICQ is easy, fast and does not
require any training, since it is composed of 12 simple and ecologic questions referring to daily life. The discriminant analysis, a predictive model
that generates a function based on linear combinations of the predictor variables, provided the best discrimination between controls with respect to
the group of deteriorated patients. According to the classification function coefficients, the questionnaire score predicted correctly 100% of normal
subjects and 85% of patients with cognitive impairment. Moreover, the receiving operating characteristics (ROC) analysis showed that the highest
sensitivity and specificity were obtained at the cut-off value of 10. Thus, a 10 score was indicative of a possible cognitive impairment requiring
further diagnostic work-up. Among 963 randomly selected subjects from the general population over 50 years of age, 130 subjects (13.5%) had a
score 10. A complete diagnostic assessment was carried out in 103 of them. Forty-one (40%) resulted cognitively normal; 34 (33%) fulfilled the
diagnostic criteria for mild cognitive impairment (MCI); 28 (27%) were diagnosed as demented.
# 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Mild cognitive impairment; Dementia; Cognitive scales

1. Introduction In recent years, a new concept has emerged in the field of

mental decline, namely mild cognitive impairment (MCI), as a
Degenerative processes in the brain leading to cognitive transitional condition between normal ageing and dementia
impairment do not have a precise day of onset, differently from (Petersen et al., 1999; Bozoki et al., 2001). MCI criteria are
stroke or from traumatic brain injury. As a consequence, (Winblad et al., 2004): (i) the person is neither normal nor
subjects often begin their examination, when the dementia is demented, (ii) evidence of cognitive deterioration shown by
overt. Clinical variability at the onset with possible atypical either objectively measured decline over time and/or subjective
patterns, depression and confounding comorbidity make the report of decline by self and/or informant in conjunction with
identification difficult. Yet, early stages of cognitive decline are objective cognitive deficits and (iii) activities of daily living are
of the utmost importance, because both pharmacological preserved and complex instrumental functions are either intact
treatment and cognitive remediation have the highest prob- or minimally impaired. Formal studies indicate that prevalence
ability to be beneficial compared to later stages, when overt of the memory complaints within the community varies greatly,
dementia has begun. A beneficial intervention in the early from 22% up to 56% (DeCarli, 2003). There is growing
stages can save months of cognitive competence, improve the evidence that individuals with MCI are at an increased risk of
quality of life and permit patients to set their financial problems developing AD (Bowen et al., 1997; Morris et al., 2001)
before the tragedy of a mental wasting. The identification of especially those having older age, poorer test performance on
subjects at risk for cognitive impairment is of primary memory tasks and orientation deficits (Wolf et al., 1998). The
importance for studying the cascade of events from the first rate of conversion from MCI to AD is estimated at 10–15% a
stages to the last ones. The progression is likely to be qualitative year, but community-based studies tend to show lower rates
rather than quantitative. (DeCarli, 2003). Early diagnosis represents the mandatory
premise in the field of cognitive decline, for the optimal
planning of interventions. Early diagnostic markers of dementia
* Corresponding author. Fax: +39 06 66058246. of AD, such as low linguistic abilities and poor episodic
E-mail address: [email protected] (S. Giaquinto). memory, can precede of many years the fulfilment of diagnostic

0047-6374/$ – see front matter # 2005 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.mad.2005.09.023
124 S. Giaquinto, L. Parnetti / Mechanisms of Ageing and Development 127 (2006) 123–128

criteria (Caltagirone et al., 2001). The diagnostic procedure 2.2. Observational study
starts at the general practitioner (GP), who most probably can
In order to assess accuracy and reliability of BICQ in detecting early phases
suspect a cognitive disorder in its early phase, thus referring
of dementia, and to confirm the correlation with MMSE, a large observational
for the whole diagnostic work up to dedicated centres. The study was started. Thirty-two GPs spontaneously agreed to participate in the
endeavour of finding effective screening tools for early research, motivated by new diagnostic guidelines for early diagnosis of
diagnosis of dementia is documented by the number of studies dementia in clinical practice. GPs involved in the project were from the Perugia
on this issue. However, the majority of such studies were area (no. 26) and from a district in Rome (no. 6). Prior to the study, all of them
carried out in memory clinics (for a review, see Petersen et al., had actively participated in a two-session training focused on the general
problem of dementia diagnosis with special emphasis on the need of early
2001). Therefore, it is necessary for the GPs to have a simple, diagnosis. Furthermore, the characteristics of BICQ were presented, including
fast, though reliable tool able to help them in confirming the the results of the preliminary validation study.
presence of initial cognitive impairment. To our knowledge, After approval was obtained from the Ethical Committee of Perugia and the
no scales specifically addressed to GPs are actually available Internal Review Board in Rome, 50 subjects of both sexes, age  50 years, were
for the Italian population, although new interesting rating randomly selected from the list of patients of each GP. The GP called at least 15
subjects among them, to briefly explain the type of initiative and asking him/her
scales have been recently presented by Italian researchers, to come to the ambulatory accompanied by a relative. In case of refusal to come,
namely the general knowledge of the world and IQCODE, a GPs proceeded to the next of the randomized list. In order to detect those with
retrospective questionnaire for caregivers about changes in cognitive impairment with a 99% probability, the calculated sample size was
the previous 10 years of the patient (Mariani et al., 2002; 850 subjects. Actually, 963 subjects entered the study. The BICQ was admi-
nistered exclusively by GPs, who completed a simple case report form with
Isella et al., 2002).
main clinical-anamnestical details of the examined subject. Relatives were also
Detection of cognitive deficits in general practice through a asked about the subject’s activities of daily living, such as dressing, meal
direct involvement of GPs is feasible. In a recent study preparations, toilette, phoning ability, changes in mood, eating behavior, leisure
assessing this issue (Fabrigoule et al., 2003), the GPs showed and sleep. Answers were dichotomised, YES or NO. These data were not
to be interested and participated well, with good patient statistically analysed.
adherence and satisfactory concordance correlation coeffi- All subjects reporting a score 10 were referred to the memory clinic for
completing the diagnostic work-up of suspected dementia, including a thorough
cients between the GPs and psychologists scores. The present neuropsychological evaluation and neuroimaging. Three diagnostic categories
article describes a very simple, ecologic, training-free and were defined: not impaired subjects, subjects with mild cognitive impairment
time-sparing tool specifically addressed to GPs, as the first and subjects with dementia. The clinical assessment was the same as that used in
diagnostic aid in case of a suspect cognitive impairment. Our the validation study.
research is aimed at providing a structured interview, namely
the Basic Italian Cognitive Questionnaire (BICQ), for a 2.3. Statistical analysis
further evaluation in memory clinics of those patients
2.3.1. Preliminary validation study
suspected of cognitive decline. The design of BICQ was Non-parametric (Spearman’s) test was used for evaluating the linear
the following. The most applied scales in Italy for the first correlation between the MMSE and the questionnaire scores. Partial correlation
assessment of cognitive impairment are the Short Portable coefficients were also calculated in order to have the actual relationship between
Mental Status Questionnaire (SPMSQ; Pfeiffer, 1975) and the scores of the two instruments after weighting for the covariates group, age
and schooling. Finally, the discriminant power of the questionnaire was
even more the Mini Mental State Examination (MMSE;
measured by means of receiving operating characteristics (ROC) curve analysis,
Folstein et al., 1975). Shopping situations were included, in order to evaluate the performance of classification schemes in which there is
since these fields at our experience seem to be more sensitive one variable with two categories by which subjects are classified.
to reveal a cognitive decline.
2.3.2. Observational study
2. Patients and methods Descriptive analysis of all demographic details (absolute and relative fre-
quency of categorical variables, number of subjects, minimum and maximum
value, mean, standard deviation, median for the continuous variables) was done.
2.1. Validation study
In the group of subjects (no. 130) with pathological BICQ score (10) correlation
analysis between the BICQ score and the MMSE score as well as between the
The preliminary validation of the questionnaire was carried out in two
BICQ score and diagnosis (normal/mild cognitive impairment/dementia) was
centres (Department of Neuroscience, Memory Clinic, Perugia University; carried out. In the latter case, a descriptive analysis of the score stratified by
Rehabilitation Centre, IRCCS San Raffaele Pisana, Rome). The study approval diagnosis and calculation of the Pearson’s coefficient between the two scores
was obtained from the local Ethical Committee and Internal Review Board,
within each diagnosis was also performed. The correlation analysis between the
respectively. One hundred cognitively impaired subjects who were already BICQ score and age, and MMSE score and age, was also been carried out. Finally,
diagnosed (30 males, 70 females, mean age 70  7, years of schooling 6  3) the association between gender, age, schooling and the BICQ score classified as
and 100 healthy subjects (40 males, 60 females, mean age 66  7, years of 10 or >10 was assessed by means of odds ratio (95% CI).
schooling 8  4.5) without cognitive deficit were included in the study. The
reliability of the variables was measured by Cronbach’s alpha coefficient and
0.62 was found. Item analysis was also used to see that the individual variables 3. Results
were suited as sum variables. Control subjects had to meet the following
requirements: (1) no previous history of neuropsychiatric disorders; (2) no use 3.1. Validation study
of psycho-active drugs; (3) still active life. Both patients and controls underwent
a clinical assessment including blood chemistry and ECG; patients with
cognitive impairment also underwent brain CT scan and a thorough neurop- Both the control and patient group had a very similar gender
sychological assessment exploring higher cortical functions. All patients ful- composition in the two centres (Perugia: M/F, controls = 23/27;
filled DSM IV criteria for dementia. patients = 13/37; Rome: controls = 26/24; patients = 12/38).
 S. Giaquinto, L. Parnetti / Mechanisms of Ageing and Development 127 (2006) 123–128 125

No statistical difference between the centres was also found on Table 2

Classification results according to the BICQ score (pathological when 10)
age, schooling, MMSE and questionnaire scores. Therefore,
data were pooled together. They are reported in Table 1. Predicted group membership
The correlation between MMSE and questionnaire scores Controls Patients Total
was highly statistically significant using Spearman correlation
Controls 100 0 100
coefficient (r = 0.902, p = 0.0001). After controlling for the Patients 15 85 100
effect of group, age and schooling, by means of partial
Controls 100 0 100
correlation coefficients analysis, the same correlation coeffi-
Patients 15 85 100
cients remained statistically significant (Spearman’s r = 0.809,
p = 0.0001). The discriminant analysis is known to be a 92.5% of cases grouped originally by means of MMSE scores and clinical
assessment were correctly classified.
predictive model that generates a function based on linear
combinations of the predictor variables and provides the best
discrimination between the groups. That analysis confirmed the
lack of overlap between the scores obtained in controls with
respect to the group of deteriorated patients. According to the
classification function coefficients, the questionnaire score
predicted correctly 100% of normal subjects and 85% of
patients with cognitive impairment (Table 2). Fig. 1 shows the
plot of the scores of BICQ and MMSE for both cognitively
impaired patients and controls. In the control group, test–retest
reliability was 0.92, inter-rater reliability was 0.91.
Finally, the ROC analysis showed that the highest sensitivity
and specificity were obtained at the cut-off value of 10
(Table 3). Thus, a 10 score is indicative of a possible cognitive
impairment requiring further diagnostic work-up.

3.2. Observational study

Fig. 1. Scattergram of BICQ (y-axis) and MMSE (x-axis) scores in controls
(red squares) and deteriorated patients (green squares).
After the validation analysis, 963 subjects (429 M, 534 F;
mean age: 66.26  10.29) were enrolled in the study. A low
education (5 years of schooling) was found in 50.6% of them. Table 3
One hundred and thirty subjects (13.5%) had a score 10 ROC analysis: highest sensitivity and specificity is obtained by a score of 10 as
(Table 4). A complete diagnostic assessment was done in 103 of cut-off
them, since 27 refused further investigations. Forty-one (40%) Sensitivity Specificity
were cognitively normal; 34 (33%) fulfilled the diagnostic
1.00 1.000 1.000
criteria for mild cognitive impairment; 28 (27%) were
.50 1.000 .980
diagnosed as demented. The mean and median values of the 1.50 1.000 .970
BICQ and MMSE scores, obtained in the three diagnostic 2.50 1.000 .880
categories are reported in Table 5. A strong correlation between 3.50 1.000 .830
BICQ and MMSE scores was observed (r = 0.860, 4.50 1.000 .760
5.50 1.000 .610
p < 0.0001). Interestingly, coefficients calculated in the three
6.50 1.000 .450
subgroups were not significant in non-deteriorated (r = 0.239, 7.50 1.000 .370
p = 0.1316) and MCI subjects (r = 0.120, p = 0.4977), while a 8.50 1.000 .250
strong association was observed in the dementia subgroup 9.50 1.000 .150
(r = 0.755, p < .0001). A significant correlation with age was 10.50 .990 .060
11.50 .770 .020
observed for both MMSE (r = 0.448) and BICQ
13 .000 .000
(r = 0.386) ( p < 0.0001).

Table 1
Clinical details of controls and cognitively deteriorated patients
Centre Group Age (years) School (years) MMSE BICQ
I Control N = 50, median 66 (55–85) 9 (3–17) 28 (25–30) 11 (10–12)
Deteriorated N = 50, median 76 (54–90) 7 (3–17) 15 (5–22) 5 (2–10)
II Control N = 50, median 68 (55–84) 8 (2–17) 28 (25–30) 11 (10–12)
Deteriorated N = 50, median 76 (52–93) 5 (3–17) 17 (5–22) 6 (2–10)
126 S. Giaquinto, L. Parnetti / Mechanisms of Ageing and Development 127 (2006) 123–128

Table 4 Table 8
Main demographic details of subjects according to the BICQ score Family
BICQ  10 BICQ > 10 95% confidence intervals 8. What are your parents’ names? (28.4%)
9. What are your grandchildren names? (60.0%)
N (%) N (%) Lower OR Upper
(8) Come si chiamano (o si chiamavano) i suoi genitori?; (9) Come si chiamano i
Male 63 (48.5) 366 (43.9) 1
suoi nipoti?
Female 67 (51.5) 467 (56.1) 0.83 1.2 1.74
Question no. 8 taps onto long-term memory. Indeed, elderly people probably
Age (years) lost their parents years before. Maiden names are more likely to be forgotten.
65 29 (22.3) 442 (53.7) 1 Question no. 9 is very delicate. Initial mental deterioration leads to impairment
66–80 62 (47.7) 333 (40.1) 0.21 0.34 0.55 in the recall of grandchildren names. As many as 16 grandchildren can be
>80 39 (30.0) 58 (6.1) 0.05 0.08 0.15 recalled in Italian families. A cohort effect can be present and impairs the recall,
since first names, at least in Italy, follow fashions on the basis of movie-stars,
Education (years)
characters, politicians and so on.
5 88 (67.6) 408 (48.9) 1
6–10 23 (17.7) 206 (24.7) 1.18 1.94 3.19
>10 19 (14.6) 219 (26.2) 1.49 2.54 4.33
Table 9
Shopping
Table 5 10. Can you indicate the price of bread per kilo? (50.7%)
BICQ and MMSE median values in the group of subjects with pathological 11. Can you tell me two brands of pasta? (52.3%)
(10) BICQ score who underwent a complete diagnostic work-up 12. What change do you expect to have in return from a
banknote of s10.00 if you have to pay s6.50? (11.5%)
No impairment MCI (N = 34) Dementia
(N = 41) (N = 28) (10) Può dirmi quanto costa un chilo di pane?; (11) Mi dice due marche di
BICQ MMSE BICQ MMSE BICQ MMSE pasta?; (12) Deve pagare 6.5 s e consegna una banconota da 10 s. Quanto si
aspetta di resto?
Median 10 28 9 23 6 14.5 Elderly people can be divided into two categories, those who go shopping and
Min 8 21 8 18 0 4 those who do not. It is important to evaluate shopping abilities when a mental
Max 10 30 10 28 10 24 deterioration is suspected. Question no. 10 is not easy even for normal persons,
although bread is present on all of the tables. Different answers are admitted on
the basis of the bread quality, but the range is clearly defined.
Question no. 11 taps onto incidental memory. Pasta is a very popular Italian
food. TV commercials are many and they are designed for sending messages to
Table 6 the memory ‘‘without effort’’. Our screening actually showed that two brands of
Personal orientation pasta have the highest probability to be recalled, on the basis of the intensity and
1. What is your age? (13.8%) the frequency of advertisements. In other Countries, where pasta is not so
2. What is your date of birth? (13.0%) popular, different items can be used, e.g., cheese.
3. Where were you born? (13.0%) Question no. 12 requires calculation abilities. But, differently from MMSE, a
4. Until what age did you go to school? (13.0%) possible shopping situation is simulated.
5. Where do you live? (14.6%) When the GP is unable to verify the accuracy of some answers, as in the cases of
proper names, the questions are put again at the end of the questionnaire and
(1) Quanti anni ha?; (2) Qual è la sua data di nascita?; (3) Dove abita?; (4) A che coherence can be ascertained in this way.
età ha lasciato la scuola?; (5) Dove abita?
The question nos. 1 and 2 are not alike. Elderly people suffering from mental
deterioration can say the birth date and yet do not know the age in years. The Tables 6–9 tap on different domains of cognitive function. In
birthplace may be forgotten, if the person moved to another place early in their all the tables, the percentage of wrong responses for each item
life. The question no. 4 gives information on schooling and its temporal given by subjects having a 10 score is in brackets. Italian
location. Address is necessary information for any citizen; persons unable to
translation is in italics.
indicate the address are certainly at risk of dementia. By contrast, the telephone
number is unknown to many healthy elderly people who are not used to dial
their own number. 4. Discussion

An international working group on MCI states that the first

Table 7 step is the identification of persons who are neither normal nor
Reality orientation demented, but there is insufficience evidence so far to
6. What day is it today? (53.0%) recommend specific tests or cut-off scores (Winblad et al.,
7. Who is the person/who are the persons with you? (5.3%) 2004). However, there has been a growing interest in both fast
(6) Oggi quanti ne abbiamo?; (7) Chi c’è con lei? cognitive scales (e.g., Froehlich et al., 1998; Solomon et al.,
Reality can be defined as the integration of self, time and place. Questions on the 1998; Buschke et al., 1999) and simple cognitive assessment
self were indicated in the previous heading. Question no. 6 refers to the date tools for community population (e.g., Monsch et al., 1995;
(year, month and day). The day of the week is often unknown to healthy elderly Cahn et al., 1996; Heun et al., 1998). None of these instruments
subjects who neither read newspapers nor follow working schedules. Question was validated for application by general physicians, differently
no. 7 taps onto the ability to recognize a wrong question. Indeed, if the subject is
alone, the answer will be ‘‘nobody’’ or ‘‘just you, doctor’’. By contrast, in the from the BICQ. This simple questionnaire confirmed the lack
case of accompanying persons, their name and parental relationship, if any, are of overlap between the scores obtained in controls with respect
expected. to the group of deteriorated patients. According to the
 S. Giaquinto, L. Parnetti / Mechanisms of Ageing and Development 127 (2006) 123–128 127

classification function coefficients, the questionnaire score Acknowledgements

correctly predicted 100% of normal subjects (group 1) and
85% of patients with cognitive impairment (group 2). ROC The Authors are indebted to Drs. Aldo Poli and Giusi
analysis was useful to evaluate the performance of Sgroi, OPIS Data, Milan, for checking the quality of data
classification schemes in which there is one variable with collected and carrying out the statistical analysis. Thanks
two categories by which subjects are classified. The highest are also due to Mrs. Astrid Van Rijn, who revised
sensitivity and specificity were obtained at the cut-off English.
value of 10, which means that a score of 10 is indicative of
some cognitive impairment requiring further diagnostic References
work-up. False negative cases are unlikely, since questions
require perfect orientation, memory and calculation that are Bowen, J., Teri, L., Kukull, W., McCormick, W., McCurry, S.M., Larson, E.B.,
at odds with a cognitive decline. According to the ROC 1997. Progression to dementia in patients with isolated memory loss. Lancet
analysis, the score of 10 is a reliable cut-off. If one lowers the 15 (349(9054)), 763–765.
cut-off, the proportion of true positive cases increases, but Bozoki, A., Giordani, B., Heidebrink, J.L., Berent, S., Foster, N.L., 2001. Mild
cognitive impairments predict dementia in nondemented elderly patients
many others are lost. If one pulls up the cut-off to 11 or to 12, with memory loss. Arch. Neurol. 58, 411–416.
then many true negative cases would be sent to a memory Buschke, H., Kuslansky, G., Katz, M., Stewart, W.F., Sliwinski, M.J., Eckholdt,
clinic. H.M., Lipton, R.B., 1999. Screening for dementia with the memory
We propose the questionnaire as a routine instrument for impairment screen. Neurology 52, 231–238.
Cahn, D.A., Salmon, D.P., Monsch, A.U., Butters, N., Wiederholt, W.C., Corey-
screening patients with initial cognitive impairment in daily
Bloom, J., Barrett-Connor, E., 1996. Screening for dementia of the Alz-
practice. It is dedicated to GPs who can decide in a better and heimer type in the community: the utility of the Clock Drawing Test. Arch.
easier way if a subject should undergo further diagnostic Clin. Neuropsychol. 11, 529–539.
investigation in specialised centres. Mini-Mental-State Caltagirone, C., Perri, R., Carlesimo, G.A., Fadda, L., 2001. Early detection and
Examination is the most popular screening tool, but it can diagnosis of dementia. Arch. Gerontol. Geriatr. (suppl. 7), 67–75.
be heavily influenced by several variables depending on the DeCarli, C., 2003. Mild cognitive impairment: prevalence, prognosis, aetiology,
and treatment. Lancet Neurol. 2, 15–21.
characteristics both of subject (e.g., education, compliance, Fabrigoule, C., Lechevallier, N., Crasborn, L., Dartigues, J.F., Orgogozo, J.M.,
emotional reactions, mood) and assessor (e.g., ability to 2003. Inter-rater reliability of scales and tests used to measure mild
induce relaxation and to ascertain that he/she has properly cognitive impairment by general practitioners and psychologists. Curr.
understood the task). Moreover, it has been demonstrated that Med. Res. Opin. 19, 603–608.
the traditional MMSE cut-off values may not be appropriate Folstein, M.F., Folstein, S.E., McHugh, P.R., 1975. Mini-mental state. A
practical method for grading the cognitive state of patients for the clinician.
in detecting early phases of dementia (Pasqualetti et al., J. Psychiatr. Res. 12, 189–198.
2002). The BICQ is easier and faster than MMSE and does Froehlich, T.E., Robison, J.T., Inouye, S.K., 1998. Screening for dementia in the
not require any training, since it is composed of a few simple outpatient setting: the time and change test. J. Am. Geriatr. Soc. 46, 1506–
and ecologic questions referring to daily life. It is better 1511.
Heun, R., Papassotiropoulos, A., Jennssen, F., 1998. The validity of psycho-
accepted and is not influenced by the variables listed above
metric instruments for detection of dementia in the elderly general popula-
for MMSE. The highest number of wrong responses at BICQ tion. Int. J. Geriatr. Psychiatry 13, 368–380.
was found in the items dealing with grandchildren and Isella, V., Villa, M.L., Frattola, L., Appollonio, I., 2002. Screening cognitive
shopping. decline in dementia: preliminary data on the Italian version of the IQCODE.
In conclusion, the statistical analysis has shown a strong Neurol. Sci. 23 (Suppl. 2), S79–S80.
association between the BICQ and MMSE and the high Mariani, C., Sacco, L., Spinnler, H., Venneri, A., 2002. General knowledge of
the world: a standardised assessment. Neurol. Sci. 23, 161–175.
predictivity of the new test, which correctly classifies 100% Monsch, A.U., Bondi, M.W., Salmon, D.P., Butters, N., Thal, L.J., Hansen,
of normal subjects and 85% of cognitively impaired subjects. L.A., Wiederholt, W.C., Cahn, D.A., Klauber, M.R., 1995. Clinical validity
In practical terms, if a subject scores 10 or less, GPs are of the Mattis Dementia Rating Scale in detecting dementia of the Alzheimer
motivated to proceed with further diagnostic work-up in a type. A double cross-validation and application to a community-dwelling
sample. Arch. Neurol. 52, 899–904.
memory clinic or any equivalent specialised centre. Further
Morris, J.C., Storandt, M., Miller, J.P., McKeel, D.W., Price, J.L., Rubin, E.H.,
researches have to verify as to what extent BICQ can be Berg, L., 2001. Mild cognitive impairment represents early-stage Alzheimer
applied in everyday practice, outside the framework of an disease. Arch. Neurol. 58, 397–405.
experimental study. The same conclusion was reached by Pasqualetti, P., Moffa, F., Chiovenda, P., Carlesimo, G.A., Caltagirone, C.,
Fabrigoule et al. (2003) who successfully trained French GPs Rossini, P.M., 2002. Mini-mental state examination and mental deteriora-
to use a simple battery of functional scales and neuropsy- tion battery: analysis of the relationship and clinical implications. J. Am.
Geriatr. Soc. 50, 1577–1581.
chological test to detect individual with incoming dementia. Petersen, R.C., Doody, R., Kurz, A., Mohs, R.C., Morris, J.C., Rabins, P.V.,
We, therefore, plan to involve the Italian Society of GPs and Ritchie, K., Rossor, M., Thal, L., Winblad, B., 2001. Current concepts in
form a task force to verify the real application over a long mild cognitive impairment. Arch. Neurol. 58, 1985–1992.
time period. These simple screening tools offer the advantage Petersen, R.C., Smith, G.E., Waring, S.C., Ivnik, R.J., Tangalos, E.G., Kokmen,
of a closer communication between case managers, namely E., 1999. Mild cognitive impairment: clinical characterization and outcome.
Arch. Neurol. 56, 303–308.
the GPs, and specialised memory clinics, where neurobio- Pfeiffer, E., 1975. A short portable mental status questionnaire for the assess-
logical mechanisms of cognitive decline can be studied under ment of organic brain deficit in elderly patients. J. Am. Geriatr. Soc. 23,
different profiles. 433–441.
128 S. Giaquinto, L. Parnetti / Mechanisms of Ageing and Development 127 (2006) 123–128

Solomon, P.R., Hirschoff, A., Kelly, B., Relin, M., Brush, M., DeVeaux, R.D., Petersen, R.C., 2004. Mild cognitive impairment beyond controversies,
Pendlebury, W.W., 1998. A seven minute neurocognitive screening battery towards a consensus: report of the Intemational Working Group on mild
highly sensitive to Alzheimer’s disease. Arch. Neurol. 55, 349–355. cognitive impairment. J. Intern. Med. 256, 240–246.
Winblad, B., Palmer, K., Kivipelto, M., Jelic, V., Fratiglioni, L., Wahlund, L.- Wolf, H., Grunwald, M., Ecke, G.M., Zedlick, D., Bettin, S., Dannenberg, C.,
O., Nordberg, A., Baeckman, L., Albert, M., Almkvist, O., Arai, H., Basun, Dietrich, J., Eschrich, K., Arendt, T., Gertz, H.J., 1998. The prognosis of
H., Blennow, K., de Leon, M., DeCarli, C., Erkinjuntti, T., Giacobini, E., mild cognitive impairment in the elderly. J. Neural. Transm. (Suppl. 54),
Graff, C., Hardy, J., Jack, C., Jorm, A., Ritchie, K., van Duijn, C., Visser, P., 31–50.