15525279, 0, Downloaded from https://alz-journals.onlinelibrary.wiley.com/doi/10.1002/alz.12572 by National Institute Of Mental, Wiley Online Library on [03/01/2023].

See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Received: 25 October 2021 Accepted: 3 December 2021

DOI: 10.1002/alz.12572

F E AT U R E D A R T I C L E

COGNITO (Computerized assessment of adult information

processing): Normative scores for a rural Indian population
from the SANSCOG study
Bratati Kahali1 Aditi Balakrishnan1 Sneha Dhanavanthri Muralidhara1
Graciela Muniz-Terrera2 Karen Ritchie3 SANSCOG Study Team*
Vijayalakshmi Ravindranath1

1
Centre for Brain Research, Indian Institute of
Science, Bangalore, India Abstract
2
Centre for Clinical Brain Sciences, The Introduction: Neuropsychological assessments are inexpensive and efficient methods
University of Edinburgh, Edinburgh, UK
to understand the cognitive abilities of individuals in research studies and clinical set-
3
Centre for Dementia Prevention, Inserm,
U1061, Montpellier, 34093 France and
tings. Normative scores for such measures are crucial in serving as a reference stan-
Université de Montpellier 1, Montpellier, dard for identifying cognitively healthy and impaired individuals belonging to similar
France
sociodemographic characteristics.
Correspondence Methods: Study subjects in rural India recruited into the Srinivaspura Aging, Neuro
Vijayalakshmi Ravindranath, Centre for Brain
Senescence and Cognition (SANSCOG) study were administered the COGNITO bat-
Research, Indian Institute of Science, C.V.
Raman Avenue, Bangalore 560012, India. tery of tests, which traverse cognitive domains of attention, memory, language, and
E-mail: [email protected]
visuospatial abilities. Percentile norms based on age and education stratification were
*
The list of collaborators for SANSCOG study is derived for the above cohort.
provided in the Appendix Results: Percentile norms are commensurate with literacy levels in this population.
The percentile scores for the cognitive tests show a decline for the individuals aged
Funding information
Centre for Brain Research, Indian Institute of 75 years and above indicating lower cognitive functioning in this age group.
Science, India
Discussion: This is the first-ever study reporting norms for diverse cognitive domains
for illiterate, literate, low-literate individuals enrolled in a large-scale community-
based cohort study in rural India.

KEYWORDS
age, COGNITO, literacy levels, neuropsychological assessments, normative scores, rural India

1 INTRODUCTION vital to tackle this enormous public health problem and recent research
has been directed toward this. However, research evidence on demen-
The older population across the world is growing rapidly. It is estimated tia from low- and middle-income countries (LMICs), including India, is
that by 2050, the number of elderly persons (aged 65 years or above) minimal, although the majority of new dementia cases in the coming
will reach 1.5 billion.1 Consequentially, the burden of aging-related dis- decades will be from LMICs.3,4
orders will also escalate significantly. For example, the number of per- It is now well known that the underlying pathological process of
sons with dementia around the world is expected to cross 100 mil- dementia begins decades before presentation of the overt clinical
lion by 2050.2 With no curative or disease-modifying treatments avail- symptoms.5 It is also becoming evident that subtle cognitive changes
able for dementia, primary and secondary prevention strategies are can appear and progress long before the appearance of diagnosable

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any
medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
© 2022 The Authors. Alzheimer’s & Dementia published by Wiley Periodicals LLC on behalf of Alzheimer’s Association.

Alzheimer’s Dement. 2022;1–10. wileyonlinelibrary.com/journal/alz 1

 15525279, 0, Downloaded from https://alz-journals.onlinelibrary.wiley.com/doi/10.1002/alz.12572 by National Institute Of Mental, Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2 KAHALI ET AL .

clinical manifestations.6 Hence, there is a growing effort to detect the

onset and progression of cognitive changes to help early diagnosis of RESEARCH-IN-CONTEXT
the disease.7 Comprehensive neuropsychological test batteries are an
1. Systematic review: The COGNITO neuropsychological
important measure to detect such early cognitive changes. Neuropsy-
battery assesses individuals in attention, memory, lan-
chological batteries assess a wide range of cognitive skills for vari-
guage, and visuospatial cognitive domains. Normative
ous cognitive domains, such as memory, attention, language, process-
scores for locally and culturally adapted COGNITO have
ing speed, reasoning, and visuo-spatial abilities. A reference scoring
not been reported for the Indian population.
of neuropsychological assessments for healthy individuals in a pop-
2. Interpretation: In this article we present the age- and
ulation facilitates correct interpretation of cognitive status of indi-
education level–stratified normative scores for COG-
viduals tested based on their sociodemographic conditions, thereby
NITO in a rural Indian population for healthy individu-
allowing identification of cognitively healthy and demented individuals.
als aged 45 years and above enrolled in the Srinivaspura
Additionally, serial administration of such detailed neuropsychological
Aging, Neuro Senescence and Cognition study.
assessment in longitudinal studies in healthy aging individuals can help
3. Future directions: Our results will aid in assessing the
determine the differential trajectories of normal versus pathological
level of cognitive functioning in individuals based on their
cognitive aging leading to dementia and related disorders.
age and education levels. Because low literacy and illit-
The Srinivaspura Aging, Neuro Senescence and Cognition
eracy can be confounding factors and lead to cognitively
(SANSCOG) is an ongoing, large-scale longitudinal cohort study
normal older adults being misclassified as cognitively
aimed at identifying the risk and protective factors associated with
impaired because of low literacy, our results are a vital
cognitive changes during normal aging, dementia, and other related
development in literacy-specific normative data for sensi-
disorders. This rural Indian study is being conducted in the villages
tive tests adapted to rural India. Addition of more individ-
of Srinivaspura, in the Kolar district in the southern Indian state of
uals in an even more advanced age group (> 80 years) will
Karnataka, and is the twin of the urban Indian study (Tata Longitudinal
further augment the normative scores and facilitate accu-
Study of Aging [TLSA]) that is being conducted in Bangalore city in
rate screening of cognitively healthy and impaired indi-
the same state. The study uses an interdisciplinary approach in which
viduals.
detailed clinical, neurocognitive, biochemical, genetic, and neuroimag-
ing assessments are carried out on individuals aged 45 and above, with
long-term follow-up. Further details of the study are available in the
SANSCOG study protocol paper.8
One of the key strengths of this rural (SANSCOG) study is system- functional or cognitive decline during the clinical history. Most of the
atic and extensive cognitive testing, which is digitized as well as fully neurocognitive assessment tools frequently used have been developed
adapted to the local languages (Indian English, Kannada, and Telugu) and validated in populations with higher levels of literacy.14 Some of
and socio-cultural context.9 Computerized neuropsychological testing these differences in performance can be attributed to lack of familiar-
is known to provide high accuracy and standardized stimulus presenta- ity with testing material or restriction of assessment to those aspects
tion. Because there is significant reduction in the administration time of cognitive screening traditionally taught in schools such as reading
for these tests, it makes comprehensive testing possible within large and writing; however, studies also show that it can also influence per-
cohort studies.10 The computerized cognitive assessment battery, formance on visuo-spatial and language assessments.15,16 It is esti-
namely, the COGNITO (Computerized assessment of adult informa- mated that there are approximately 758 million illiterate adults across
tion processing) covers four principal areas of cognitive functioning— the globe. While prospective cohort studies have shown the relation-
attention, memory, visuospatial processing, and language—and has ship between low literacy levels and higher chances of developing
been shown to detect mild cognitive changes in pre-clinical stages of dementia17,18,19 very few have examined cognitive assessment tools in
dementia. The component tests cover both quantitative and qualita- adults with low literacy levels.20 Hence, it is important to develop cul-
tive aspect of cognitive performance and can be used for clinical as well turally and linguistically adapted norms to the reference population to
as epidemiological research.10 The majority of the tests in this battery accurately detect cognitive impairments.12
record responses through a touchscreen, enabling its use in literate, To date, normative data and psychometric properties of the English
lesser-educated, and illiterate individuals. It offers tests with a wider version of COGNITO were documented in a population of 78 individu-
range of difficulty levels, to avoid ceiling and floor effects.10 als from France and England with university level education.10 To the
Cognitive test performance is established to be influenced by educa- best of our knowledge, there are no studies that have attempted to
tion and culture.11 Various cognitive tests require language and seman- use this battery to assess cognitive functioning among groups of indi-
tic treatment of information, which is influenced by psycholinguistic viduals with varied literacy levels, especially among individuals with
factors (e.g., word frequency or familiarity) that vary between cultures lower literacy levels. In the present study we aim to establish age-
and languages.12 Literacy and low levels of education have been shown and literacy-stratified normative data for the tests in COGNITO for
to be associated with an increased risk for developing dementia.13 a predominantly cognitively normal and lesser literate sample of indi-
Illiterate and low-literacy adults tend to perform poorly on many of viduals participating in this rural India-based SANSCOG longitudinal
the conventional cognitive assessment tools, even without evidence of study.
 15525279, 0, Downloaded from https://alz-journals.onlinelibrary.wiley.com/doi/10.1002/alz.12572 by National Institute Of Mental, Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
KAHALI ET AL . 3

2 METHODS which have been designed to assess phonology, morphology, syntax,

and semantics.
2.1 Sample Memory is assessed through an immediate and delayed recall of
a name list, name–face recall task, visuo-spatial span, and an implicit
This analytical sample consists of a subset of apparently healthy par- memory task.
ticipants (n = 1440, age ≧ 45 years, males and females) from the Visuo-spatial processing is assessed based on a construction task, a
SANSCOG study, who had completed their baseline assessments, matrix reasoning task, and a semantic and functional matching task.10
including the COGNITO neurocognitive battery. These participants The original COGNITO battery has 24 total tests, which fall under
were recruited into the SANSCOG study, following an area sampling one of the four principal cognitive domains (attention, language, mem-
strategy, with mapping of the catchment areas under the coverage of ory, and visuospatial processing). For the present study, tests were
the respective primary health management units in Srinivaspura. Indi- administered to participants based on their literacy categories and
viduals already diagnosed to have mild cognitive impairment or demen- their ability to read and write. Tests that involved reading and writ-
tia were excluded from the study. ing tasks such as Stroop, vocabulary, implicit memory, and visuospatial
The study is approved by the institutional human ethics commit- construction (which require the use of a pencil), were not administered
tees at the Centre for Brain Research (CBR) and the collaborating to illiterates. The narrative recall and descriptive recall, which is a part
institutes—National Institute of Mental Health and Neurosciences of the original COGNITO battery, was not administered to any of the
(NIMHANS) in Bangalore. All participants signed the written informed participants as it could not be cross-culturally adapted to suit the rural
consent form. SANSCOG cohort. Tests were administered in two Indian languages—
Kannada and Telugu—depending on the language preference of the
participant.
2.2 Tools used

A literacy questionnaire21 was used to examine the literacy status of 2.3 Data analysis
participants. This questionnaire assesses an individual’s ability to read
and write in various daily activities. The tool consists of 11 questions The socio-demographic details of the participants were analyzed using
such as “Could you read the tag of an item in a market?,” “Can you use descriptive statistics (mean, standard deviation). Participants were
street signs to find your way?” and “Can you note the name of the caller stratified by age (45–54, 55–64, 65—74, and 75 years and above) and
on the phone?” Based on the responses, the subjects are classified into literacy category: specifically literates, illiterates, and semi- and func-
different literacy statuses, namely “literates” (those who can read and tional literates.
write fluently), “illiterates” (those who cannot read and write at all), The COGNITO battery automatically produces 1005 variables, both
“semi-literates” (those who cannot read or write enough to manage quantitative and qualitative in nature. For ease of normative data cal-
daily tasks) and “functional literates” (those who cannot read and write culation, we have selected or computed a single score that can be used
enough to manage daily tasks). In the present study, as the number of as an outcome measure for each test, consisting of scores such as “total
individuals in the semi- and functional literate category were small, and correct responses,” “total incorrect responses,” and computation of
with the overlap in their abilities to perform on the tasks included in the scores to obtain a single test score10 (e.g., fluency total score was cal-
battery, the individuals in the semi- and functional literate categories culated by summing the semantic fluency score and phonemic fluency
were grouped into a single category. score). More details on the tests and variables used to calculate norma-
COGNITO (a computerized assessment of adult information pro- tive data are provided in Table 1.
cessing) was administered to each participant as part of the neuropsy- Percentile norms were derived for the tests of the COGNITO bat-
chological assessment. This battery, developed by the National Insti- tery for each age group and literacy category. The 5th, 10th, 25th,
tute of Health and Medical Research (Inserm)—University of Mont- 50th , and 75th percentiles were calculated. Statistical analysis was con-
pellier, France, is a computerized, comprehensive neuropsychological ducted by using SPSS version 23.0.
assessment battery. Permission was obtained to cross-culturally adapt
this battery and use it for the SANSCOG study. The battery assesses
four principal neurocognitive domains of attention, language, memory, 3 RESULTS
and visuo-spatial processing.10
Attention is measured through the auditory and visual modalities COGNITO norms were calculated for a sample of 1400 participants
using an auditory discrimination task, a visual attention task wherein from rural India. Table 2 shows the demographic characteristics of the
the participants are required to identify a visual stimulus presented participants, wherein 52.77% of the sample population are females.
among multiple distractor items, and a dual attention task in which the The mean age of the sample is 57.75 years (standard deviation
two tasks are performed together. Language is assessed through tasks [SD] = 9.99) and the mean years of education is 4.53 years (SD = 4.69).
ranging from a reading and comprehension task, phoneme comprehen- Illiterates comprised 46.59% of this sample, of which 76% are females.
sion test, naming and associations test, fluency, and a vocabulary task, Literate individuals comprised 40.83% of the sample (mean years of
 15525279, 0, Downloaded from https://alz-journals.onlinelibrary.wiley.com/doi/10.1002/alz.12572 by National Institute Of Mental, Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4 KAHALI ET AL .

TA B L E 1 Tests comprising attention, language, memory, visuospatial abilities cognitive domains in COGNITO, and the outcome variable used
for derivation of normative scores

Test Variable used Maximum score possible

Attention
Reaction time (RT) Mean RT 500 ms
Auditory attention Total correct answers 10
Visual attention Total correct answers 10
Dual attention Total correct answers 10
Language
Comprehension Total correct answers 5
Phoneme comprehension Total correct answers 10
Naming Total correct answers 10
Associations Total correct answers 10
Fluency Total correct answers (phonemic + No maximum score limit as score is based on the
semantic) number of correct responses given by participant
Vocabulary* Total correct answers 35
Memory
Names immediate recall Total correct answers 9
Names delayed recall Total correct answers 9
Names recognition Total correct answers 18
Name-face association: names Total correct answers 9
Name-face association: faces Total correct answers 18
*,#
Implicit memory Average number of frames taken to –
identify the name by total number of
frames* 10
Visuo-spatial abilities
Stroop test* Total errors 45
Visuo-spatial span Total span 9
Geometric figures Total correct answers 8
Matrices Total correct answers 30
Construction (drawing) *
Total correct elements (House + Abstract) 80 (40 + 40)

*Test not administered to the Illiterate group as it involved reading or drawing with the use of a pencil.
#
Score for this test was calculated based on the number of frames the participant took to identify the name (stimuli) divided by the total number of frames
possible multiplied by the number of trials (10). Based on these calculations, a lower score would indicate better performance on the implicit memory test.

education = 9.30, SD = 2.83), and 12.56% of participants belonged to tests due to reading or writing constraints or difficulties in compre-
either the semi- or functional literate category (mean years of educa- hending certain tasks. For each test, participants with missing data
tion = 4.50, SD = 2.23); 73.13% of participants who belonged to the were excluded, after which the normative data for the sample was com-
literate category were males. The 45 to 54 age range makes up 42.5% puted. Normative data for Stroop, vocabulary, implicit memory, and
of the sample, 29.02% belonged to the 55 to 64 age range, 20.55% visuospatial construction are not provided for illiterates as these tests
belonged to the 65 to 74 range, and 7.92% of the sample was 75 years were not administered to them.
and above. Males constituted 47.22% of the total sample. The 5th percentile corresponds to the threshold under which 5% of
The normative data for the tests in the COGNITO battery under participants have the lowest performances and the 75th percentile is
the domains of attention, language, memory, and visuo-spatial process- the threshold above which 25% of participants have the best perfor-
ing are shown in, respectively, Tables S1A, S1B, S1C, S1D in support- mances. Tables S1A, S1B, S1C, S1D and Figures 1A-D, 2A-F, 3A-F, 4A-E
ing information. Norms are presented for the three literacy categories clearly depict that the normative scores are the highest for the literate
(literates, illiterates, and semi- or functional literates) within the four category across all age groups, the semi-literates have an intermedi-
age groups (45–54, 55–64, 65–74, and 75 years and above). The sam- ate score, and the illiterate individuals have the least scores—a pattern
ple size differs across tests because of participants not completing the observed for most of the cognitive tests in each of the domains. Note
 15525279, 0, Downloaded from https://alz-journals.onlinelibrary.wiley.com/doi/10.1002/alz.12572 by National Institute Of Mental, Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
KAHALI ET AL . 5

TA B L E 2 Sociodemographic characteristics of the study sample in here that for tests’ reaction time, for Stroop C and Implicit Memory, a
the Srinivaspura Aging, Neuro Senescence and Cognition study lower score is indicative of better performance. Therefore, the 5th per-
Standard centile for these two tests corresponds to the lowest performing indi-
Mean deviation viduals but shows a higher normative data, and the 75th percentile nor-
Characteristic N (%) (in years) (in years) mative score is showing a greater value but indicates the higher per-
Age (in years) formers in these tests.
45-54 612 (42.5) 48.33 2.76 These results from the current study dataset across three different
55-64 418 (29.02) 58.93 2.79 literacy groups suggest cognitive abilities are highest in literate, least in

65-74 296 (20.55) 67.60 2.68 illiterate, and intermediate in semi/functionally literate individuals. The
percentile scores for the cognitive tests show a decline for the individ-
75+ 114 (7.92) 78.37 3.90
uals aged 75 years and above, indicating lower cognitive functioning in
Total 1440 57.75 9.99
this age group.
Literacy groupa
Literate 588 (40.83) 9.30 2.83
Illiterate 671 (46.59) 0.36 1.23
4 DISCUSSION
Semi- or functional 181 (12.56) 4.50 2.23
literate The present study aimed at developing normative data for the COG-
NITO battery with a large sample of aging adults from the ongoing
Sex
SANSCOG study—a prospective, community-based, cohort study that
Males 680 (47.22) – -
focuses on understanding the risk and protective factors associated
Females 760 (52.77) - -
with normal as well as pathological aging. This is significant because lit-
a
For literacy group, the mean and standard deviation is represented in num- eracy and level of education have been consistently found to influence
ber of years. cognitive performance on neuropsychological assessments, leading

F I G U R E 1 Box plot representation of normative scores for cognitive domain “attention” for tests: (A) reaction time, (B) auditory attention, (C)
visual attention, (D) dual attention, in different age and literacy categories for 1440 individuals in the Srinivaspura Aging, Neuro Senescence and
Cognition study population
15525279, 0, Downloaded from https://alz-journals.onlinelibrary.wiley.com/doi/10.1002/alz.12572 by National Institute Of Mental, Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
KAHALI ET AL .

comprehension, (C) naming, (D) associations, (E) fluency, (F) vocabulary, in different age and literacy categories for 1440 individuals in the
F I G U R E 2 Box plot representation of normative scores for cognitive domain “language” for tests: (A) comprehension, (B) phoneme

Srinivaspura Aging, Neuro Senescence and Cognition study population

6
 15525279, 0, Downloaded from https://alz-journals.onlinelibrary.wiley.com/doi/10.1002/alz.12572 by National Institute Of Mental, Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
KAHALI ET AL . 7

(A) (B)

(C) (D)

(E) (F)

F I G U R E 3 Box plot representation of normative scores for cognitive domain “memory” for tests: (A) names immediate recall (B) names
delayed recall, (C) names delayed recall recognition, (D) name–face associations names, (E) name–face associations faces, (F) implicit memory, in
different age and literacy categories for 1440 individuals in the Srinivaspura Aging, Neuro Senescence and Cognition study population

cognitively normal older adults to be misclassified as being cognitively In the current article, we have presented normative data for the
impaired.13 As longitudinal epidemiological studies require the use tests in COGNITO battery, which was adapted to suit the language and
of a sensitive battery to assess neurocognitive functioning, it is vital culture of the study population. These norms have been computed for
to develop normative data specific to the cultural and educational a sample of 1440 cognitively normal individuals living in rural India
background of the population under study. in different age and literacy categories. We observed that the scores
 15525279, 0, Downloaded from https://alz-journals.onlinelibrary.wiley.com/doi/10.1002/alz.12572 by National Institute Of Mental, Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
8 KAHALI ET AL .

(A) (B)

(C) (D)

(E)

F I G U R E 4 Box plot representation of normative scores for cognitive domain “visuospatial abilities” for tests: (A) Stroop C, (B) visuospatial
span, (C) geometric figures, (D) matrices, (E) drawing in different age and literacy categories for 1440 individuals in the Srinivaspura Aging, Neuro
Senescence and Cognition study population

for most of the cognitive tests in each of the domains are the high- for the cognitive tests show a decline for the individuals aged 75 years
est (lowest for reaction time, Stroop C and Implicit memory, wherein and above, indicating lower cognitive functioning in this group of indi-
a lower score is indicative of better performance) for the literate cate- viduals (Tables S1A, S1B, S1C, S1D and Figures 1A-D, 2A-F, 3A-F, 4A-E).
gory across all age groups, and lowest for the illiterate individuals, with The strengths of this study lie in the large sample size analyzed to
the semi-literates having an intermediate score. The percentile scores calculate the normative data. Classification of individuals into various
 15525279, 0, Downloaded from https://alz-journals.onlinelibrary.wiley.com/doi/10.1002/alz.12572 by National Institute Of Mental, Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
KAHALI ET AL . 9

literacy categories in this study addresses a significant lacuna as data academic inputs of CBR’s International Advisory Board members. The
from illiterate or less literate population, which is pertinent to India and SANSCOG study is funded through the Centre for Brain Research,
other developing countries, is scarce. The work carried out in this study Indian Institute of Science, India.
based on the individuals’ functional literacy abilities provides us with
norms for individuals across the spectrum of literacy, which is highly CONFLICTS OF INTEREST
valuable in a country such as India, which has such unique diversity The authors have no conflicts, whether personal, financial, or other-
in terms of its language, education, socio-economic status, and rural– wise.
urban ratio.22
This study has certain limitations. The mean age of the participants REFERENCES
was 57.75 years. The number of participants in the older age groups 1. United Nations, Department of Economic and Social Affairs, Popula-
was lower, and hence individuals were grouped into one category for tion Division . New York: UN; 2019. World population ageing 2019:
Highlights. ST/ESA/SER.A/430.
75 years and above for normative data calculation. As the proportion
2. Feigin VL, Vos T, Nichols E, et al. The global burden of neurological dis-
of the population constituting the age group of 80 years and above are orders: translating evidence into policy. Lancet Neurol. 2020;19(3):255-
increasing worldwide,23 and with age being one of the highest risk fac- 265.
tors for the development of dementia,24,25 the norms for this popula- 3. Prince MJ, Wimo A, Guerchet MM, Ali GC, Wu Y-T, Prina M. World
Alzheimer Report 2015 – The Global Impact of Dementia: An analysis
tion will require modification as the population ages or as the number
of prevalence, incidence, cost and trends. Alzheimer’s Disease Interna-
of participants in the > 80 years age “oldest old” category recruited into tional; 2015:84.
the study increases. Additionally, future longitudinal data of the COG- 4. Ferri CP, Prince M, Brayne C, et al. Global prevalence of dementia: a
NITO battery from our study will help to understand the mechanisms Delphi consensus study. Lancet. 2005;366(9503):2112-2117.
5. Braak H, Thal DR, Ghebremedhin E, Del Tredici K. Stages of the patho-
underlying the decline of cognitive activity with old age, and how cog-
logic process in Alzheimer disease: age categories from 1 to 100 years.
nitive functioning is maintained in otherwise healthy individuals in this J Neuropathol Exp Neurol. 2011;70(11):960-969.
cohort. 6. Silverberg NB, Ryan LM, Carrillo MC, et al. Assessment of cognition in
The norms provided in this paper will help facilitate the study of early dementia. Alzheimers Dement. 2011;7(3):e60-e76.
cognitive functioning in a large sample of individuals ranging from lit- 7. Salmon DP, Bondi MW. Neuropsychological assessment of dementia.
Annu Rev Psychol. 2009;60:257-282.
erate and low literacy to illiterate participants with a wide variety of
8. SANSCOG study team et al. Srinivaspura Aging, Neuro Senescence
cognitive assessments under four principal cognitive domains: atten- and COGnition (SANSCOG) study: Study Protocol. Manuscript sub-
tion, memory, language, and visuospatial abilities. Such normative data mitted.
will be of great value as a reference panel to clinicians and researchers 9. Lukose et al., Cross-cultural adaption of the Computerized Assess-
ment of Information Processing battery (COGNITO) for an Indian lon-
alike who are interested in studying and assessing cognitive function-
gitudinal study on rural elderly. Manuscript submitted.
ing for identifying cognitively healthy and impaired individuals; as well 10. Ritchie K, de Roquefeuil G, Ritchie CW, et al. COGNITO: comput-
as studying cognitive changes in populations with varied levels of liter- erized assessment of information processing. J Psychol Psychother.
acy, especially in the low literacy and illiterate individuals living in rural 2014;136(4):1–5.
11. Stewart R, Johnson J, Richards M, Brayne C, Mann A. Medical council
areas constituting a major mass of population in India and developing
cognitive function and ageing study. The distribution of Mini-Mental
nations. State Examination scores in an older UK African-Caribbean popu-
lation compared to MRC CFA study norms. Int J Geriatr Psychiatry.
AUTHOR CONTRIBUTIONS 2002;17(8):745-751.
12. Larouche E, Tremblay MP, Potvin O, et al. Normative Data for the mon-
The SANSCOG team led by Vijayalakshmi Ravindranath conceptu-
treal cognitive assessment in middle-aged and elderly Quebec-French
alized and designed the SANSCOG study. Bratati Kahali and Gra- people. Arch Clin Neuropsychol. 2016;31(7):819-826.
ciela Muniz-Terrera designed the analysis reported in this paper. 13. Meng X, D’Arcy C. Education and dementia in the context of the cog-
Karen Ritchie helped with initial administration of COGNITO in the nitive reserve hypothesis: a systematic review with meta-analyses and
qualitative analyses. PloS One. 2012;7(6):e38268.
SANSCOG study. Acquisition of data was done by Aditi Balakrishnan
14. Youn JH, Siksou M, Mackin RS, Choi JS, Chey J, Lee JY. Differenti-
and Sneha Dhanavanthri Muralidhara. Analysis was done by Aditi Bal- ating illiteracy from Alzheimer’s disease by using neuropsychological
akrishnan, Sneha Dhanavanthri Muralidhara, and Bratati Kahali; and assessments. Int Psychogeriatr. 2011;23(10):1560-1568.
interpretation of data was done by Bratati Kahali and Graciela Muniz- 15. Ostrosky-Solis F, Ardila A, Rosselli M, Lopez-Arango G, Uriel-Mendoza
Terrera. Drafting the manuscript along with figures and tables was by V. Neuropsychological test performance in illiterate subjects. Arch Clin
Neuropsychol. 1998;13(7):645-660.
Aditi Balakrishnan, Sneha Dhanavanthri Muralidhara, Bratati Kahali,
16. Rosselli M, Ardila A, Rosas P. Neuropsychological assessment in
and Vijayalakshmi Ravindranath and Graciela Muniz-Terrera gave crit- illiterates. II. Language and praxic abilities. Brain Cogn. 1990;12(2):
ical suggestions on the draft. All authors approved the final version of 281-296.
the manuscript for publication. 17. Da Silva CG, Petersson KM, Faísca G, Ingvar H, Reis A. The effects
of literacy and education on the quantitative and qualitative aspects
of semantic verbal fluency. J Clin Exp Neuropsychol. 2004;26(2):
ACKNOWLEDGMENTS 266-277.
The authors are grateful to the volunteers who participated in the 18. Wajman JR, Oliveira FF, Schultz RR, Marin SMC, Bertolucci PHF.
SANSCOG study. The authors acknowledge the valuable advice and Educational bias in the assessment of severe dementia: brazilian
 15525279, 0, Downloaded from https://alz-journals.onlinelibrary.wiley.com/doi/10.1002/alz.12572 by National Institute Of Mental, Wiley Online Library on [03/01/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
10 KAHALI ET AL .

cutoffs for severe Mini-Mental State Examination. Arq Neuro-Psiquiatr.

2014;72(4):273-277. How to cite this article: Kahali B, Balakrishnan A,
19. Scazufca M, Almeida OP, Vallada HP, Tasse WA, Menezes PR. Limi- Dhanavanthri Muralidhara S, et al. COGNITO (Computerized
tations of the Mini-Mental State Examination for screening demen-
assessment of adult information processing): Normative
tia in a community with low socioeconomic status: results from the
Sao Paulo Ageing & Health Study. Eur Arch Psychiatry Clin Neurosci. scores for a rural Indian population from the SANSCOG study.
2009;259(1):8-15. Alzheimer’s Dement. 2022;1-10.
20. Tavares-Júnior JW, de Souza AC, Alves GS, de Bonfadini J, Siqueira- https://doi.org/10.1002/alz.12572
Neto JI, Braga-Neto P. Cognitive assessment tools for screening older
adults with low levels of education: a critical review. Frontiers in Psychi-
atry. 2019;878(10):1–12.
21. Moon HS, Chey JY. Literacy and neuropsychological functions in
APPENDIX A
the older Korean adults. J Korean Geriatr Psychiatry. 2004;8(2):
Collaborators
113-120.
22. Dhillon P, Ladusingh L, Agrawal G. Aging and changing patterns in SANSCOG STUDY TEAM
familial structure for older persons in India: a decomposition analysis. List of Collaborators—arranged in alphabetical order.
Qual Aging Older Adults. 2016;17(2):83-96.
23. United Nations, Department of Economic and Social Affairs, Popula-
tion Division. New York: UN; 2013. World population ageing 2013: Expertise/area
Highlights. ST/ESA/SER.A/348 Name and title Affiliation of interest Contact
24. Corrada MM, Brookmeyer R, Paganini-Hill A, Berlau D, Kawas CH. 1
BN Gangadhar, NIMHANS Psychiatry kalyanybg@
Dementia incidence continues to increase with age in the oldest old:
MD yahoo.com
the 90+ study. Ann Neurol. 2010;67(1):114-121.
25. Giulioli C, Meillon C, Gonzalez-Colaço Harmand M, Dartigues JF, Girish N. Rao, NIMHANS1 Public Health girishnrao@
Amieva H. Normative scores for standard neuropsychological tests in MD yahoo.com
the oldest old from the French population-based PAQUID study. Arch Jonas S. Sun- CBR, IISc2 Psychiatry [email protected]
Clin Neuropsychol. 2016;31(1):58-65. darakumar,
MRCPsych
Naren P. Rao, NIMHANS1 Psychiatry docnaren@
MD gmail.com
SUPPORTING INFORMATION Palanimuthu T. NIMHANS1 Psychiatry sivakumar.
Additional supporting information may be found in the online version Sivakumar, nimhans@
of the article at the publisher’s website. MD gmail.com

1
National Institute of Mental Health and NeuroSciences, Bangalore, India
2
Centre for Brain Research, Indian Institute of Science, Bangalore, India