Introduction
Age structure of population changes over time, and has recently manifested “population ageing” in most countries around the world. Population ageing is defined as “the process whereby older individuals become proportionately the larger share of the total population.” Number of individuals aged over 60 is growing faster than the total number of population and this increase is even faster among those who are over 80. There are now more than 800 million people aged 60 years and over and this number is expected to become more than 2 billion by 2,050, which would account for 22% of the whole population. Population ageing is faster in developing countries versus developed ones. Two third of the people aged 60 and over live in developing countries (UNFPA and HelpAge International, 2012), which also happen to lack sufficient support and sub-structures of healthcare system for the elderly.
Ageing population has many impacts on countries in terms of economics, vocation, and health system. The “dependency ratio” is generally used as a surrogate to estimate the impact of population aging in the society; it can be defined as the number of dependent people divided by the number of working age population (i.e. total aged 15 to 59 years). However, this index is not a precise one because of heterogeneity of vocational status among people aged 60 years and over. Dependency ratio is increasing globally but by variable rates depending on a country's economic milieu. In the Middle East it is expected to only increase slightly, as the working age population is increasing simultaneously. The total number of dependent people is expected to rise to 225 million by 2,050.
Health issues of older population are of the most challenging problems globally. In fact, declines in mortality in old age may be due to reduced “fatality” of disabling chronic diseases rather than reduction in their incidence, hence older people might live longer in state of ill health, and we might instead experience an “expansion of morbidity”. This is even truer among aging women.
Alzheimer's disease is one of the main health issues of aged population. Today, about 4.7% of the population aged over 60 years lives with dementia and this is predicted to triple by 2,050, meaning 115 million people with the disorder (Prince et al., Reference Prince, Prina and Guerchet2013). This challenge is quite different in developed versus developing countries, as their age structures and epidemiologic transitions, population genetics and cognitive function, attitudes toward the illness and available diagnostic and therapeutic procedures are very different (Nitrini et al., Reference Nitrini2004). Moreover, prevalence of dementia is higher among illiterates (stephanie S. Richards, Reference Stephanie, Richards, Sadock, Sadock and Kaplan2009). This might be due to their childhood style development, exposure to dementia risk factors, insufficient mental stimulation and cognitive reserve, poor control of cardiovascular risk factors, poverty and low socioeconomic status. There is also low level of awareness about normal aging and dementia. Furthermore, there are several difficulties in cognitive evaluation for this specific population and adaptation of neuropsychological assessment tools is relatively poor (Ardila et al., Reference Ardila, Ostrosky-Solis, Rosselli and Gomez2000; Nitrini et al., Reference Nitrini2004). It has been estimated that delaying AD onset by 1 or 2 years could decrease the disease burden in 2,050 by 9.5 million or 23 million cases, respectively (Brookmeyer et al., Reference Brookmeyer, Gray and Kawas1998).
Now, in absence of any curative treatment for dementia, early diagnosis and intervention is the only way that can reduce the prevalence (Brookmeyer et al., Reference Brookmeyer, Gray and Kawas1998) and the economic burden of the disorder, and improve patients’ quality of life (Brookmeyer et al., Reference Brookmeyer, Johnson, Ziegler-Graham and Arrighi2007). Currently, the U.S. Preventive Service Task Force does not recommend for or against screening in the primary care setting unless the patient or the family members report memory problems (stephanie S. Richards, Reference Stephanie, Richards, Sadock, Sadock and Kaplan2009). However, regarding the increasing prevalence of dementia, screening in primary care settings, at least among high risk population, needs to be considered. If so, for screening in huge scales, and also for screening people who present memory complaints, a screening tool which is non-expensive, non-invasive, compatible with the level of education and culture, and implementable by trained primary healthcare providers is needed. Among available neuropsychological tools, the level of education is a major confounding factor which makes the results of these measurements unreliable among the illiterate population.
The goal of this review is to address the key issues in assessment tools in developing countries with a high level of illiteracy. To undertake this conceptual review, we searched electronic databases (Medline and Cochrane) using the terms “illiterate,” “dementia,” “developing country,” “elderly,” “cognitive,” and “education.” Then the titles and abstracts were checked for eligibility. More papers were also recruited by reviewing the former papers references.
Literacy: definition, measurement and rates around the world
According to the UNESCO 2013 report, one in five adults in the world today is still illiterate, two third of which are women. Literacy remains a challenge for 773 million adults lacking minimum literacy skills. More than half of the world's adult illiterates live in south and west Asia. Each year, millions of children leave school without acquiring the basic literacy and numeracy skills necessary to empower them for their future life. The rate of illiteracy is specifically higher in developing countries, rural areas and women (UNESCO, 2013); which makes them more prone to develop Alzheimer's disease.
Lack of education is a major risk factor for developing dementia. Prevalence of dementia is about 15.67% among illiterates, while the percentage is 7.16% among literates. Elders with low levels of literacy have a steeper decline in both immediate and delayed recall of a word list over time, as compared to higher literacy elders. Literacy skills are protective against memory decline among non-demented elders (Manly et al., Reference Manly, Touradji, Tang and Stern2003; Scazufca et al., Reference Scazufca, Almeida, Vallada, Tasse and Menezes2009).
As UNESCO defines, a person is literate when he/she can, with understanding, both read and write a short statement on his/her everyday life. In Iran, national literacy agency uses the same definition. There exist also related definitions: Functional literacy is the ability to use reading, writing and numeracy skills for effective functioning and development of the individual in community; Pure illiterates are the real illiterate subjects who were never exposed to the experience of a symbolic written representation and have never learned a grapheme-phoneme matching operation (UNESCO, 2006).
Years of education do not necessarily reflect the level of literacy. As all individuals do not acquire the same amount of learning from a certain grade level, literacy could be a more accurate reflection of innate ability. There are always some subjects who seek learning more than others outside of school, which will be reflected in measurements of literacy, and naturally not in years of education. Both innate ability and educational experience contribute to an individual's literacy level (Manly et al., Reference Manly, Touradji, Tang and Stern2003). Quality of education is an emerging concept which includes not only educational materials but also school environment, trained teachers, more parental involvement and even children's nourishment. Poor education quality, at least indirectly, can increase the rate of illiteracy through increasing drop-out and repetition. As inclusion of effects of innate ability and quality of education on literacy level is quite impracticable, years of education is commonly used as a proxy for level of literacy (Jeanette Colby, Reference Jeanette Colby2000).
Studies on literacy
In human history, writing only dates back to some 5,000–6,000 years ago (Sumer, 3,200 BCE). Just a few centuries ago, reading and writing abilities were uncommon among the general population. The acquisition of reading and writing skills has changed the brain's topographic organization of cognitive function during this period of time. The first neuropsychological studies on illiteracy were performed by Luria and Vygotsky in 1930s during a study in Uzbekistan, Central Asia. As suggested by Vygotsky (1978), reading and writing are cognitive tools that, once acquired, change the way in which stimuli are memorized and conceptualized, thus stimulating abstract thinking. They studied the extent to which reasoning, memory, and categorization are shaped by social and economic practices. They found that the population working in a traditional economy involving agriculture and farming, with many of them being illiterate or with one or two years of formal education, name the stimulus by approximation with real objects (for example, tent for triangle), whereas in those with more education this type of naming is substituted by more abstract names. Some striking differences were identified on tasks of reasoning in illiterates, but two years or more of schooling produced a dramatic improvement in accuracy (Ardila et al., Reference Ardila, Ostrosky-Solis, Rosselli and Gomez2000).
Impact of literacy on the brain
Exposure to cognitive stimulation that occurs in the complex process of schooling has two effects: diffuse and focal. The diffuse effect is related to adaptation to a rich environment like enhancement of abstract thinking, and the focal effect is related to learning specific skills and operations that constitute the ability of reading and writing (Castro Caldas et al., Reference Castro Caldas, Reis and Guerreiro1997). Education is associated with higher brain weight because of enhanced synapto-dendritic enhancement, complex axonal projections and enhanced myelination and neurogenesis.
“Cognitive Reserve” is a hypothesized capacity of mature adult brain to tolerate the effects of disease or injury without clinical manifestations. It accounts for individual differences in the cognitive processes and neural networks which allow one to cope with brain damage better than others. Years of education would be an indicator of the brain's ability to compensate for pathology through more efficient use of existing cognitive networks or recruitment of alternate networks. Education increases synaptic density in neocortical association cortex, leading to a delay of symptoms by 4– 5 years in those with Alzheimer's disease, and probably, other dementing disorders, hence halving the prevalence of dementia. Enriched environment, which includes complex, highly structured enjoyable activities that provide opportunity for self-expression, has been found to promote neurogenesis, synaptogenesis and trans-generational transmission; these cognitively and physically stimulating activities affect the cognitive reserve positively in all stages of life and development. Neural Reserves are networks other than cognitive reserve networks and are less susceptible to damage, more efficient, have greater capacity, and may be involved while coping with increased task demand and help the individual cope with brain pathology. Neural Compensation is a process by which individuals suffering from brain pathology use brain structures or networks not normally used by intact brain to compensate for brain damage. These compensations occur by two methods: in “compensation to improve performance”, the brain recruits additional brain areas which have resilient and healthy networks, i.e. like similar network on contralateral hemisphere; in “compensation to maintain performance”, the brain recruits any network (Snowdon et al., Reference Snowdon, Kemper, Mortimer, Greiner, Wekstein and Markesbery1996; Whalley et al., Reference Whalley, Starr, Athawes, Hunter, Pattie and Deary2000; Manly et al., Reference Manly, Touradji, Tang and Stern2003; Ostrosky-Solís, Reference Ostrosky-Solís2004).
Almost all older people have some pathology in their brain at death but have not necessarily died with dementia. A large body of evidence indicates that lack of education is a robust risk factor to develop dementia (Chen et al., Reference Chen, Lin and Chen2009). Longer years in education have been associated with decreased dementia risk and greater brain weight but have no relationship to neurodegenerative or vascular pathologies; thus it can be said that more education largely reduces risk of dementia, independent of severity of pathology. More education does not protect individuals from developing neurodegenerative and vascular neuropathology by the time they die, but it does appear to reduce the impact of pathology on the clinical expression of dementia before death. Those who remain in education for longer are able to compensate for pathological burden in later life, rather than a hypothesis based on a protective effect of education against the accumulation of pathology arising from life-long environmental factors (Brayne et al., Reference Brayne2010; Mortamais et al., Reference Mortamais2013).
Literacy exerts influence on the cerebral dominance for language. The right hemisphere is more important than the left one for linguistic processing in the illiterate compared to the literate. Cerebral representation of language is more bilateral in illiterates than it is in school-educated subjects. Corpus callosum is thinner in the illiterate group, especially in the segment where the parietal lobe fibers cross. Parietal lobe processing role is different between literates and illiterates. Information processing in occipital lobe is slower in individuals who learned to read as adults compared to those who learned at the usual age (Petersson et al., Reference Petersson, Reis and Ingvar2001; Parente et al., Reference Parente, Fonseca and Scherer2008). Illiterate individuals do not appear to differ from literates in daily conversational and memory abilities or in independent functioning, even though they perform more poorly on a variety of formal tests. Instead of relying on phonological encoding, they may use semantic information and visual imagery to memorize names of objects. Cognitive domains that are most involved in literacy are phonological attention, production of hand motor sequences, declarative memory, working memory, executive functions and visuospatial abilities (Castro Caldas et al., Reference Castro Caldas, Reis and Guerreiro1997; Folia and Kosmidis, Reference Folia and Kosmidis2003; Ardila et al., Reference Ardila2010).
Education and neuropsychological evaluation
There is a strong relationship between psychometric measures of intelligence and the schooling system an individual has been subjected to. Formal education may enhance the innate ability of learning through training individuals in the efficient learning and retrieval strategies (Pirozzolo et al., Reference Pirozzolo, Mahurin, Loring, Appel and Maletta1985; Folia and Kosmidis, Reference Folia and Kosmidis2003). During normal aging, the educational variable is even more influential on neuropsychological performance than the age variable. The most resistant cognitive domains to the impact of aging are orientation to space and person, digit backwards, language repetition, phonological verbal fluency, and changing the position of left and right hands (Ardila et al., Reference Ardila, Ostrosky-Solis, Rosselli and Gomez2000). Some of the variable domains of language function are significantly affected by education, including naming figures, naming body parts, word repetition, naming photographs, naming real objects, repetition of non-words and reaction time (Castro Caldas et al., Reference Castro Caldas, Reis and Guerreiro1997). Other domains of cognitive function that are affected by education include visuospatial ability, all memory tasks -with the exception of the immediate memory of sentences-, and word list recall. Calculation is one of the cognitive domains most related to literacy. There are illiterate subjects who show remarkable expertise in mental calculation due to their job experiences, but this is not the rule. The need to deal with numbers and calculation tasks in daily life requires these subjects to generate strategies of their own (Castro Caldas et al., Reference Castro Caldas, Reis and Guerreiro1997; Nitrini et al., Reference Nitrini2004).
Necessary considerations in the validation of neuropsychological tools
There are even more challenges to assess cognitive function of old minorities and immigrants. There is lack of validated instruments with adjustment of cut-points, translation and replacement of culture-specific items. Their life stressors are more than others. Many of them are socially isolated or are poorly socially engaged. They have poor access to health system resources. It is also difficult for professionals to communicate with people from different cultural backgrounds, especially if the client does not speak the same language. So, there is risk of over-diagnosis of dementia. Functional scales would be the best for them (Parker and Philp, Reference Parker and Philp2004).
Teamwork is needed in order to adapt an existing instrument for use in another culture. First, manuals and instructions must be studied by an experienced team including clinical neuropsychologist, cognitive neurologist, psychiatrist and psychometrist. It is not enough to translate a questionnaire literally. Translation and back-translation by separate groups of bilingual translators is necessary to ensure that the meaning and significance of the items are preserved. The content, semantic, technical, criterion and conceptual equivalence with the parent instrument for every item should be considered. Comparison of the back-translated scale with the original one will reveal problematic items that should be revised until a valid version, which considers the language and sociocultural aspects of the population, is achieved (Sperber, Reference Sperber2004). The next step is to validate the test against a “gold standard” diagnosis of dementia. Establishment of reliability between field worker and clinicians is also mandatory (Chandra et al., Reference Chandra1998).
Current measurement tools and their limitations
The most popular assessment tools which have been used in populations including the ones with low level of education, and their shortcomings are discussed below. These tests are also recommended by Alzheimer's Association for operationalizing the detection of cognitive impairment in a primary care setting. However, these tests are not comprehensively validated among illiterate and low-educated population; also contain pen and paper tasks that directly related to education. Hence none of them are free from educational, also language and culture bias (Cordell et al., Reference Cordell2013).
During 15 years of professional experience with cognitive disorders including dementia in elderly population in Iran, among whom 71.1% are illiterate (UNESCO, 2008), the first author has faced shortcomings of these measurement tools which are defective and cannot be entirely relied upon for diagnosis of cognitive impairment in people who are illiterate or have low levels of education, and even in educated ones who have different culture and life routines. These findings might be also applicable to neighboring, eastern Mediterranean and possibly other developing countries.
Mini-Mental State Examination
Mini-Mental Status Examination (MMSE) is the most commonly used assessment tool in the screening of dementia. It assesses five cognitive domains: orientation, registration, attention, calculation, recall and language. The impact of education on the MMSE is well known (Folstein et al., Reference Folstein, Folstein and McHugh1975; Scazufca et al., Reference Scazufca, Almeida, Vallada, Tasse and Menezes2009) and the actual effect of education, occupation and other cultural factors on MMSE may be dramatic. However, this tool has been validated in various languages and is commonly used, because it is relatively brief and easy to perform. In one study, healthy non-readers received an average score of 21.9 which is below the cut-off score for dementia, whereas readers with the minimum of 8 levels of education received 28 (Weiss et al., Reference Weiss, Reed, Kligman and Abyad1995; Parker and Philp, Reference Parker and Philp2004; Brucki and Nitrini, Reference Brucki and Nitrini2010b). MMSE has both ceiling and floor effect, meaning a high false negative effect in people with mild cognitive impairment or in highly educated people and a high false positive effect in people who are older and illiterate. In fact, the score of 30 in MMSE does not always mean normal cognition and the score of zero does not mean a total lack of cognition either. Nonetheless, in severe Alzheimer's disease, especially in final seven years, the MMSE score is generally zero (Houx et al., Reference Houx2002; Parker and Philp, Reference Parker and Philp2004; Reisberg, Reference Reisberg2007). Also, both age and job can affect MMSE score. MMSE scores decline while age advances and some suggest lowering cut-offs for the elderly. One study found that even after controlling age and education, farmers scored 2.3 points less than white collar workers on MMSE (Scazufca et al., Reference Scazufca, Almeida, Vallada, Tasse and Menezes2009). The Framingham study recommends different cut-off scores depending on the level of education, however it should be measured in different populations specifically (Woodford and George, Reference Woodford and George2007).
A person who is either illiterate or demented may lose scores in MMSE; while there is marked discrepancy among the examinee's scores in various cognitive domains, for example very low scores in drawing interlocked pentagons but higher score in performing complex command, illiteracy bias should be considered; this pattern is not necessarily seen in the patients with dementia.
MMSE is also disadvantageous in diagnosis of non-Alzheimer's dementias such as vascular or frontotemporal dementia, because it does not contain much capacity to test frontal executive functions (Smith et al., Reference Smith, Gildeh and Holmes2007). For example, the pentagon task of the MMSE simply requires the patient to copy the image, and does not assess planning skills like clock drawing test (CDT).
There are also challenges in translating MMSE questions. For example, choosing a five-letter word to replace “world” in the backward-spelling task poses technical difficulties in some languages. The sequence of “L” and “D” as consonant letters needs to be paid attention to. Word by word translation of the repetition phrase “no ifs, ands or buts” would not be equivalent in languages other than English. The repetition of this statement is a little bit difficult because of existence of several small words with prepositions, and therefore, it is suitable to assess aphasia. Actually, these expressions must not be directly translated, as the translation does not represent a good example of the test's aim. Asking about orientation to time and date in a retired old subject -who lives in a rural area- without any weekly schedule seems not to be fair (Parker and Philp, Reference Parker and Philp2004).
Clock drawing test (CDT)
Clock drawing test (CDT) is a simple and brief cognitive screening tool applicable across different cultures and languages. CDT assesses multiple cognitive domains including comprehension, planning ability, executive functioning, visual memory and reconstruction, visuospatial ability and numerical knowledge. There is a large body of evidence that shows it overestimates cognitive impairment due to dementia in illiterates, as they are uncomfortable in using pen and paper to draw lines and numbers; a study shows optimal cut-off score should be lower in people with less than six years of education (Munang et al., Reference Munang, Chan and Lim2010).
Memory impairment screen (MIS)
Memory Impairment Screen (MIS) is a brief screening test, based on free- and cued-recall of a four-item word list. The fact that it takes only few minutes to do, its feasibility of administration by trained non-specialists and high correlation with Alzheimer's pathology, makes MIS a suitable screening tool. But it requires minimum education and that restricts its use among illiterates. A modified version of MIS, called picture-based MIS (PMIS), has been developed to overcome this limitation and has proved to be a reliable screening tool for dementia in adults with little or no education (Verghese et al., Reference Verghese2012).
Category fluency test (CFT)
Another commonly used screening tool is category fluency test (CFT) which is very quick, usually taking one minute, and easy to perform and has proved to be useful for the diagnosis of dementia, with high sensitivity and specificity in detecting the initial stages of AD. CFT reflects left medial temporal lobe function, an area that is known to be affected early in the course of AD. It can be applied to different cultural groups, the illiterate and other low-level educational groups because it does not require reading or writing skills (Caramelli et al., Reference Caramelli, Carthery-Goulart, Porto, Charchat-Fichman and Nitrini2007). In CFT, examinees are required to generate, within one minute, as many words as possible that belong to a certain category. Categories are defined by either phonemic or semantic resemblance, which means to generate words beginning with a given letter or to recite names of a given category respectively (Lezak et al., Reference Lezak, Howieson and Loring2004; Brucki and Nitrini, Reference Brucki and Nitrini2010b). CFT evaluates a variety of cognitive processes including word knowledge, semantic memory, long-term verbal memory, attention, speed of information processing, vocabulary size, working memory, inhibition of irrelevant words, and executive function, especially in terms of flexibility and initiative abilities. When using animals, it requires activation of the semantic memory system, but since many sub-categories within the “animal” field also exist, more flexibility is required. It has a high sensitivity in differentiating early stages of dementia from normal aging. It also detects impairment not only due to Alzheimer's disease, but also in a variety of pathological conditions such as schizophrenia, Parkinson's Disease, HIV associated dementia and traumatic brain injury (Fichman et al., Reference Fichman2009). Scores below 15 in the CFT (animals/min) have a sensitivity of 88%, specificity of 96%, in differentiating AD patients from normal controls.
CFT is commonly used as a brief cognitive assessment tool. Because it is not a pen and pencil test, one may think it is appropriate for illiterate people. Although not completely wrong, literacy does affect vocabulary size.
The variables which affect the score of CFT should be considered precisely. It has been shown that the frequency of words which start with a specific letter and also length of words can affect CFT scores. For word generation in CFT, a proper starting letter needs to be chosen. Spanish speaking people produce fewer words beginning with letters F, A, and S, compared to English speaking ones, as these letters are less common in Spanish than in English. They also generate smaller number of animals than Vietnamese speakers, as animal names are the longest in Spanish and shortest in Vietnamese. Studies have shown that neither gender nor age affects scores of CFT, with the exception of age and category changing. Educational level significantly affects the total number of animals generated in the fluency tasks, and in the various intervals, as well as number of clusters, changes between categories and number of categories.
Other tools
Other tests known as the most appropriate ones to screen cognitive function are The General Practitioner assessment of Cognition (GPCog), Rowland Universal Dementia Assessment Scale (RUDAS), Mini Cog, 7 minutes, Alzheimer's Disease Assessment Scale-Cognitive subscale (ADAS-Cog) and Montreal Cognitive Assessment (MoCA). These tests posit of tasks such as drawing a clock, reading a sentence, reading a word list, and trail making; hence none of them can be considered free from educational bias (Peña-Casanova, Reference Peña-Casanova1997; Cordell et al., Reference Cordell2013).
Evaluating cognitive function in illiterate people
As mentioned above, there are many challenges in screening of dementia among illiterate people, as literacy level, including both formal and informal education such as Quran reading in Muslims, can affect the results (Avila et al., Reference Avila, Moscoso, Ribeiz, Arrais, Jaluul and Bottino2009). Because illiterate people are less familiar with test taking, they may experience more anxiety while testing which can impair test performance. Test instruction is very important and testing without using pen and paper is recommended for them. Some modifications must be done due to age and lifestyle, for example in many developing countries, the elderly population does not know their exact dates of birth and there are no official birth records. Estimation can be done by important events, the date of marriage or starting a job. For naming, photographs and actual objects can be used instead of line diagrams. Serial subtraction (serial seven) which evaluates attention, can be substituted by serial three subtracted from twenty, and test structure can be merged into a life skill of the target population (Chandra et al., Reference Chandra1998; Fichman et al., Reference Fichman2009). Babacan-Yildiz et al. have recently introduced a brief screening battery for Alzheimer's disease in illiterate and literate people in turkey named Cognitive State Test (COST). They assessed the reliability and validity of this test among illiterate people, literate people with one to five years of education and people with more than five levels of education. They defined different cut-off points for each group (Babacan-Yildiz et al., Reference Babacan-Yildiz2013).
Verghese and colleagues wisely replaced written words with figures and developed an illustrative version of MIS named PIMS (Verghese et al., Reference Verghese2012). Such substitution makes the test appropriate to be used among illiterate and low-educated population. While selecting figures, one should bear in mind that the process of decoding a schematic picture is affected by literacy, education, and also exposure to social life (Brucki and Nitrini, Reference Brucki and Nitrini2010a). While selecting illustrated items, they should be among native animals and fruits of each country so that the target population is familiar with. For instance it seems unwise to use the picture of Avocado or Koala for Iranian population.
Proverbs have extensive use as a diagnostic brief test to evaluate cognitive function across a variety of disciplines including behavioral neurology. Appropriate proverbs should be selected for the target population with their culture in mind. Cognitive domains engaged during proverb and metaphor interpretation include abstract thinking, language and working memory. Familiar Proverbs interpretation may involve semantic memory. Unfamiliar proverbs task relies more heavily on working memory and problem solving. The right hemisphere plays a special role in the comprehension of figurative (non-literal) language, and in particular metaphors. Using proverbs is applicable in the illiterate population for assessment of dementia, however, scoring the interpretation is challenging (Van Lancker and Klein, Reference Van Lancker and Klein1990; Brown et al., Reference Brown and Wright-Harp2001; Schmidt and Seger, Reference Schmidt and Seger2009).
Several considerations need to be taken into account in translation of cognitive assessment tests. “You” in English is used both for singular and plural items; in translation to Persian, the researcher should be cautious to use plural form when addressing an old patient for appropriate respect. The translation of the phrase “No ifs ands or buts” in Persian is very simple and, therefore, its repetition cannot be used to assess language fluency and repetition (Seyedian et al., Reference Seyedian, Fallah, Nourouzian, Nejat, Delavar and Ghasemzadeh2008). The term “caregiver” cannot be interpreted in a similar meaning in Persian because of its negative connotation: referring to someone who observes the patient like an investigator. Cultural differences and stigma also should be mentioned. Western culture has a fairly consistent understanding of dementia as a medical condition caused by several diseases, particularly AD, but many countries do not share this Western view. Some communities may see dementia as a normal part of ageing, while others see it as a mental illness or “madness”, wherein may be stigma or shame. Also, the underlying meaning of the translation of “dementia” varies considerably across different languages (Ostrosky-Solís, Reference Ostrosky-Solís2004).
In places where cognition cannot be directly measured, functional level can be measured as its reflection. The tests so designed have some advantages over cognitive assessments for screening of dementia: they are less influenced by factors such as education, occupation, practice effects, personal background, cultural and linguistic factors and are applicable in heterogeneous education and socioeconomic levels; they have greater severity range, have utility in initial and terminal stages of AD, and have higher sensitivity to the longitudinal course of the disease; they are easier to translate, and are less subject to practice effect. Tests that assess functional level include Functional Assessment Staging (FAST), Clinical Dementia Rating (CDR), and Global Deterioration Scale (GDS) (Reisberg, Reference Reisberg2007; Brucki and Nitrini, Reference Brucki and Nitrini2010b).
FAST, introduced in 1988 by professor Reisberg, is one of the most popular scales used for functional staging of patients with cognitive disturbances of Alzheimer's type or other etiologies (Reisberg, Reference Reisberg1988).
Activity Daily Living international scale (ADL-IS) is a useful way to assess capabilities of the old illiterate subjects. Inappropriate items of the Western questionnaires of ADL for less industrialized settings, including using a map for transportation, driving skills especially for women, cooking for men, and telephone usage should be substituted by items which are compatible with socio-economic status of the target subjects by local experts (Reisberg et al., Reference Reisberg2001). Fillenbaum et al. introduced an activities daily living scale appropriate for use in illiterate rural elderly people in India. They have selected items which those elderly people could be expected to perform, including but not limited to necessary seasonal tasks, in which most elderly people might be expected to engage (Fillenbaum et al., Reference Fillenbaum, Chandra, Ganguli, Pandav, Gilby and Seaberg1999).
Functional assessment tools do not measure a specific cognitive domain, as each life skill includes various cognitive functions. As the basic daily activities, for instance dressing, and bathing, are relatively similar around the world, such tests are more universal, and easier to perform, thus usually a trained healthcare worker can evaluate patients; these tests are very useful to monitor a single patient overtime. If a functional test is based on the informant report, it might be biased, also unfortunately, many elderly people live alone and their family does not have enough information about their life details to report. However, in more traditional societies like the Iranian one, we do not face this problem as most elderly people live with their families (Reisberg, Reference Reisberg2007).
Moreover available tools must be validated among illiterate and low-educated people by studies which exclusively include this population. New cut-off points should be defined for each test according to the level of literacy.
Ethics
Finally, ethical issues should be considered for illiterate people. Informed consent should be read to non-demented subjects by the researcher. Signed written voluntary consent must be obtained from all non-demented subjects or the caregiver or legal guardian of the patients with Dementia. The illiterate non-demented subjects should put his/her thumb print on the consent form implying consent to participate in the study, however amongst some societies they may be concerned and refuse to comply.
Conclusion
The rate of illiteracy is strikingly different between developed and developing countries. Illiteracy rate is even higher among elderly population and women in the developing societies and makes the assessment of cognitive dysfunction and dementia a challenge.
Minorities and immigrants pose potential difficulties in screening by conventional screening tools; even though they have a higher rate of developing dementia, cognitive function cannot be accurately assessed because of cultural differences and language barriers.
Considering the important role of early diagnosis of dementia and the impact of appropriate intervention on the reduction of prevalence and the burden of these devastating disorders (NIH, 2011), it seems there is a caveat for more studies on appropriate, non-expensive, non-invasive screening tools, compatible with the level of education and culture in developing countries. Among these, special attention should be focused on the level of education as the major confounding factor on the results of cognitive assessment in the illiterate population.
Considering the above-mentioned facts and obstacles, most of the popular neuropsychological tools are not appropriate enough to be used in developing societies. The shortcoming is not only problematic in the clinical settings, but also poses problem in performing basic epidemiological and public health studies.
In order to develop an appropriate screening tool for illiterate or low-educated population, a team consisting of a neurologists, psychiatrists, neuropsychologists, methodologists and linguists should work together. Such a tool should precisely evaluate appropriateness of the screening questionnaires for measurement of different cognitive domains. Special attention must be paid to assessment of functional scales and activity daily living scales compatible with the culture of each society. Illustrative cards with real, rather than schematic pictures, can be used for delayed recall evaluation. Cognitive test can also benefit from using appropriate proverbs and metaphors compatible with ethnicity, and asking questions about traditions, religious rituals and historical events to assess abstract thinking, and general information and praxis respectively. For planning, especially among women, asking about how to cook a special dish can be useful. Naming months of the year forward and backward can be useful to evaluate attention, concentration and working memory.
Each test has to be precisely validated within literate and illiterate population. If possible, the tests have to be validated in different ethnic groups within the same society.
As various factors can affect cognitive function and dementia manifestation in any affected person, including, but not limited to job exposure, living environment, level of education, and intelligence, it seems fit to select the appropriate cognitive test for each individual person precisely; functional scales have to be more considered in clinical evaluations, especially in cases where a progressive neurodegenerative disorder is suspected because these basic self-care tasks are amongst the last to be impacted in the disease course.
It is recommended that further research be undertaken in developing countries to find either appropriate cultural and lingual items to better assess cognitive function or define which of the currently available tests have higher priority to be used among the illiterate and low-educated people.
Finally, sharing experiences and insights of experts in developing countries can be very valuable in developing appropriate tools compatible with cultural characteristics of each society.
Conflict of interest
None.
Description of authors’ roles
Maryam Noroozian designed the research question(s), carried out the original review; a part of this study was presented in the sixth International Conference on Alzheimer's disease and Related Disorders in the Middle East on October 2013 in Istanbul, Turkey. She also wrote the core of the manuscript and supervised writing the wholepaper. Alia Shakiba wrote the manuscript and added additional materials. Shahrzad Iran-nejad assisted in writing the paper and its edition in English.
