Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Developing an Informant Questionnaire for Cognitive Abilities in Down Syndrome: The Cognitive Scale for Down Syndrome (CS-DS)

  • Carla M. Startin ,

    Affiliations UCL Division of Psychiatry, University College London, London, United Kingdom, The LonDownS Consortium

  • Erin Rodger,

    Current address: Faculty of Health and Medical Sciences, University of Surrey, Guilford, United Kingdom

    Affiliations UCL Division of Psychiatry, University College London, London, United Kingdom, The LonDownS Consortium

  • Lucy Fodor-Wynne,

    Affiliations UCL Division of Psychiatry, University College London, London, United Kingdom, The LonDownS Consortium

  • Sarah Hamburg,

    Affiliations UCL Division of Psychiatry, University College London, London, United Kingdom, The LonDownS Consortium

  • André Strydom

    Affiliations UCL Division of Psychiatry, University College London, London, United Kingdom, The LonDownS Consortium


Down syndrome (DS) is the most common genetic cause of intellectual disability (ID). Abilities relating to executive function, memory and language are particularly affected in DS, although there is a large variability across individuals. People with DS also show an increased risk of developing dementia. While assessment batteries have been developed for adults with DS to assess cognitive abilities, these batteries may not be suitable for those with more severe IDs, dementia, or visual / hearing difficulties. Here we report the development of an informant rated questionnaire, the Cognitive Scale for Down Syndrome (CS-DS), which focuses on everyday abilities relating to executive function, memory and language, and is suitable for assessing these abilities in all adults with DS regardless of cognitive ability. Complete questionnaires were collected about 128 individuals with DS. After final question selection we found high internal consistency scores across the total questionnaire and within the executive function, memory and language domains. CS-DS scores showed a wide range, with minimal floor and ceiling effects. We found high interrater (n = 55) and test retest (n = 36) intraclass correlations. CS-DS scores were significantly lower in those aged 41+ with significant cognitive decline compared to those without decline. Across all adults without cognitive decline, CS-DS scores correlated significantly to measures of general abilities. Exploratory factor analysis suggested five factors within the scale, relating to memory, self-regulation / inhibition, self-direction / initiation, communication, and focussing attention. The CS-DS therefore shows good interrater and test retest reliability, and appears to be a valid and suitable informant rating tool for assessing everyday cognitive abilities in a wide range of individuals with DS. Such a questionnaire may be a useful outcome measure for intervention studies to assess improvements to cognition, in addition to detecting dementia-related cognitive decline. The CS-DS may also be a useful tool for other populations with ID.


Down syndrome (DS) is the most common genetic cause of intellectual disability (ID), occurring due to the presence of an extra chromosome 21. The incidence of DS is approximately 1 in 1000 live births [1], with a life expectancy of approximately 60 years old [2]. While almost all individuals with DS have an ID (mean IQ approximately 50), there is great variability in cognitive abilities both across and within individuals [3]. Cognitive abilities that are particularly affected include executive function [4, 5], memory [6, 7], and language [8].

The impairments in executive function and memory in DS have been associated with altered brain development of the prefrontal cortex and hippocampus respectively, with both of these regions having smaller volumes in people with DS in neuroimaging studies [911]. This is possibly related to later developing networks being affected more in DS than other brain structures [12]. A recent functional magnetic resonance imaging (fMRI) study also suggested atypical functional organisation for language processing in DS [13].

Another feature of DS is the ultra-high risk of developing Alzheimer’s disease; a recent study estimated that lifetime risk based on cumulative incidence for dementia by age 68 may be as high as 95.7%, with an age related increase in incidence from 26.1% by age 50 [14, 15]. In comparison, mortality adjusted lifetime risk for Alzheimer’s disease in the general population at age 45 has been estimated at 19.5% for women and 10.3% for men [1618]. There is a large variability in the clinical presentation and age of onset of dementia in DS, with some adults receiving a diagnosis in their late 30s and others not showing any signs of dementia in their 60s [1922]. Abilities associated with cognitive decline in DS show overlap with those abilities affected by the cognitive profile of DS; it has been suggested that frontal function, characterised by executive function impairments [23, 24], may be affected relatively early in the course of dementia [20] alongside memory impairments [25, 26].

A variety of assessment batteries have been published to assess cognitive abilities and cognitive decline in people with DS [3, 2732]. While these test batteries have been developed to be suitable for the majority of people with DS, they are often not suitable for many individuals, who often score at floor levels for many of the tasks [3, 30, 32] or who may have visual or hearing problems. For these individuals informant ratings are invaluable to assess cognitive abilities and any related changes (either possible improvements due to an intervention, or possible decline associated with dementia). Informant ratings of cognitive abilities may also be a useful addition to test batteries in those who are able to engage with formal testing, allowing for comparison of test scores across the ability spectrum. At present there are several informant scales available to assess symptoms related to dementia in ID [3335] but to our knowledge there are no dedicated informant rated scales to assess everyday cognitive abilities in ID that are appropriate for individuals across the age and ability spectrum, in particular younger adults before the onset of possible decline, and that may be used to track change over time. Such changes may include improvements in abilities following interventions, assessing decline in abilities, or predicting future decline.

While several scales assessing executive function and memory for the general population are available (such as the Behavior Rating Inventory of Executive Function—Adult Version (BRIEF-A) [36] and the Observer Memory Quotient (OMQ) [37]), these often contain questions that carers feel are inappropriate and irrelevant for people with an ID. A recent study found scores on the BRIEF and OMQ did not correlate with IQ measures in people with DS [32], suggesting they may not be related to measures of general abilities. Further, the BRIEF asks informants to rate the extent to which particular behaviours have been a problem over the last 4 weeks, and so is not a direct measure of abilities. For some individuals with an ID then often activities are structured in a way so that potential problems can be avoided, and so the lack of a problem with a particular behaviour does not necessarily mean that the individual does not have any difficulties with that behaviour.

We therefore aimed to create a suitable informant scale of everyday cognitive abilities for adults with DS to assess specific aspects of cognition that are affected by DS; in particular executive function, and also memory and language. The approach we adopted was pragmatic to ensure a robust and easy to use scale that can be used to obtain information from carers of individuals with DS regardless of severity of ID, age, or comorbidities, and we therefore aimed to include a broad range of individuals aged 16 and older. Further, informants were both family members and paid carers to reflect the realities of obtaining information about adults with DS who are often living in supported settings.


Questionnaire development

Existing questionnaires were consulted to guide development of relevant questions: these were the Behavior Rating Inventory of Executive Function—Adult Version (BRIEF-A) [36], the Observer Memory Quotient (OMQ) [37], the Vineland Adaptive Behavior Scales [38], and the Dementia Questionnaire for People with Learning Disabilities (DLD) [33]. Concepts for potential questions were identified based on their relevance to one of three main domains (executive function, memory, and language), with the final composite consisting of a total of 66 questions pertaining to executive function (36 questions), memory (18 questions), and language (12 questions). All questions were phrased to be consistent in style, and appropriate for an ID population. Of these, 34 were reverse phrased to reduce response bias, split across the domains. Each question had 3 options to select from: never/rarely true, sometimes true, and often/always true.

The CS-DS was sent to two ID psychiatrists for comments before use. After completion with the first 21 informants, minimal amendments were made to questionnaire wording and options were amended as appropriate based on feedback, to ensure that questions and options were as clear as possible. These changes consisted of changing the wording of question 46 from ‘doesn’t know when their birthday is’ to ‘forgets when their birthday is’, and changing options from rarely true, sometimes true and often true to never/rarely true, sometimes true, and often/always true.

As a part of the CS-DS we collected information about participants’ languages spoken, the presence of any vision or hearing problems, and the presence of any changes over the last year.

Ethical approval

Ethical approval was obtained for the LonDownS study from the North West Wales REC (13/WA/0194). Where individuals had capacity to consent for themselves we obtained written informed consent for the LonDownS study, which included consent to collect information from informants. Where individuals did not have capacity to consent for themselves a consultee was appointed and asked to sign a form to indicate their decision regarding the individuals’ inclusion based on their knowledge of the individual and his/her wishes, in accordance with the UK Mental Capacity Act 2005.

Questionnaire completion

Carers of 156 participants with a clinical diagnosis of DS from the LonDownS study [3] were approached to complete the CS-DS. Participants were located across England. In total, we received completed questionnaires about 130 individuals (5 carers did not want to complete the questionnaire, and the remaining 21 did not respond); two of these were removed from analyses due to more than 5 missing answers leaving a total of 128 questionnaires.

For the first 21 questionnaires we called participants’ main carers to ask them to complete the CS-DS with a researcher. This allowed us to take comments and feedback and assess how easy the wording of the questionnaire was to understand. The remainder of the questionnaires were either completed with a researcher or filled in by carers themselves, depending on carers’ preference. This pragmatic approach allowed for the realities of collecting data from carers about individuals with ID.

Questionnaires were collected about individuals both with and without dementia. Informants were parents (56.3%), another family member (8.6%) or paid carers (35.2%). All raters had known participants for at least 3 months, and see participants at least once a week. Participants were older for questionnaires completed by paid carers compared to family members (t(126) = -6.52, p<0.001, 95% CI (-17.46, -9.32), family member M 31.99 SD 11.15, paid carer M 45.38 SD 11.00), likely reflecting the realities of older participants being more likely to live in a care setting.

Reliability testing

CS-DS test retest reliability was assessed by collecting 36 questionnaires completed for a second time by original raters in the same way (i.e. with a researcher or filled in by themselves; 97.2% parents, 2.8% other relatives). Participants included in the test retest sample consisted of 18 males and 18 females, with an age range from 16 to 49 (M 29.14 SD 8.27). No participants included in the test retest sample had shown any decline related to dementia, and between the two administrations of the questionnaire no participants had shown any changes in abilities; this was checked at the time of the second questionnaire. The second questionnaire was completed up to 12 months following the initial questionnaire using the same administration method (M 6.83 SD 3.00); we included a range of time delays to ensure that raters could not remember their responses to the initial questionnaire. Further, as repeatability over time is essential for outcomes to assess clinical interventions [32] longer delays allowed us to validate this. CS-DS interrater reliability was assessed by collecting 58 questionnaires completed by a second rater on the same day as the first rater. Again, all raters had known participants for at least 3 months, and see participants at least once a week. For the questionnaires collected from a second rater, 3 were removed from analyses due to more than 5 missing answers, leaving a total of 55 questionnaires (52.7% parents, 5.5% other relatives, 41.8% paid carers). Participants included in the interrater sample consisted of 30 males and 25 females, with an age range from 16 to 57 (M 31.80 SD 11.11). Intraclass correlations were determined for both test retest and interrater test scores to assess CS-DS reliability.

Validity testing

Discriminative validity was determined by comparing CS-DS scores for individuals aged 41 and over with (n = 23) and without (n = 28) significant cognitive decline as assessed using the CAMDEX [39] (the group with significant cognitive decline also included those with a clinical diagnosis of dementia). Significant cognitive decline was defined as decline occurring firstly in the memory domain and secondly in either the other cognitive functions or personality and behaviour domains of the CAMDEX, with the decline not co-occurring with other factors such as depression. Concurrent validity was determined by investigating the relationship between CS-DS scores for adults without significant cognitive decline with two measures of general abilities, the Kaufman Brief Intelligence Test 2 (KBIT-2; n = 100) and the short adaptive behavior scale (short ABS; n = 102). The KBIT-2 [40] consists of 3 subtests which assess general cognitive abilities through questions relating to verbal knowledge, pattern completion and riddle completion. All subtests were started at question 1 and stopped after 4 consecutive incorrect answers. In this analysis, we used total raw score as the outcome measure, due to a high floor effect when scores were converted to IQ scores. The short ABS [41] is an informant questionnaire measuring everyday adaptive abilities.


We used SPSS for all analyses. In addition to analyses described above, we investigated a relationship between total score and both age and sex for the subgroup of adults without significant cognitive decline (n = 105). Finally, we performed exploratory factor analysis to assess the underlying factor structure of the CS-DS. We used a maximum likelihood method with oblique direct oblimin rotation, and a fixed number of 5 factors based on Scree plot analysis. Questions with a factor loading of at least 0.5 were considered to load significantly on to the respective factor.


Participant demographics

Analyses were performed for questionnaires on a total of 128 individuals. Of these, 68 were male, and 60 female. The majority of participants were white British (85.2%). For those without significant cognitive decline (n = 105) there was a wide range of ages (16–63) and ID severity as assessed either by parent / carer report (37 mild ID, 54 moderate ID, 14 severe ID) or using DSM-IV criteria to define ID severity based on KBIT-2 scores (15 participants had an IQ between 56 and 70 (i.e. mild ID), 42 participants had an IQ between 41 and 55 (i.e. moderate ID), and 43 participants had an IQ at floor (i.e. 40 and under; severe ID); 2 adults were unable to complete the KBIT-2 due to poor vision and 3 adults refused to complete it). A subgroup of adults aged 41+ without cognitive decline (n = 28) was used as a comparison group in later analyses. The age range for those with significant cognitive decline (n = 23) was 38–66. Full participant demographics across the different participant groups used in analyses can be found in Table 1.

Table 1. Participant demographics and scores across the questionnaire and domains.

Final question selection

We first performed discriminatory analysis to identify any questions where the majority of informants answered the same way. Frequencies for each response for each question were calculated, and any questions showing poor discrimination (more than 80% of respondents answering the same way) were removed (see Table 2). This resulted in removal of 5 questions, leaving a total of 61 questions (16 in the memory domain, 36 in the executive function domain, and 9 in the language domain). There were no questions that were consistently not answered by carers, suggesting the wording of the questions was easy to understand and the questions are suitable. The final questionnaire can be found in the S1 File.

Table 2. Original CS-DS items and percentage of informants selecting each option.

Distribution of responses

CS-DS scores showed a wide range, with a minimum score of 22 and a maximum of 119 and a possible range from 0 to 122. Scores were not normally distributed; there was a slight negative skew (Shapiro-Wilk test W(128) = 0.97, p = 0.005, skewness = -0.39 (SE 0.21), kurtosis = -0.64 (SE 0.43)). For the memory domain the minimum score was 1 and the maximum was 32 (with a possible range from 0 to 32), for the executive function domain the minimum score was 14 and the maximum was 69 (with a possible range from 0 to 72), and for the language domain the minimum score was 0 and the maximum was 18 (with a possible range from 0 to 18). Further details about mean scores and ranges for total and domain scores across the different participant groups can be seen in Table 1.

We found no floor or ceiling effects for the total score on the CS-DS, and limited floor and ceiling effects across the three domains (only 1 participant (0.8%) was at floor for the language domain, 4 participants (3.1%) were at ceiling for the language domain, and 8 participants (6.3%) were at ceiling for the memory domain).


Both test retest reliability and interrater reliability were high; intraclass correlations were 0.95 (95% CI (0.91, 0.98), p<0.001) for test retest reliability and 0.84 (95% CI (0.74, 0.90), p<0.001) for interrater reliability (see Fig 1a and 1b). The length of time between first and second administrations of the scale for test retest reliability was not a significant regressor when comparing the relationship between scores at the two time points (Β = 0.009, p = 0.867). This suggests consistency of CS-DS scores over time and raters.

Fig 1. Assessing CS-DS reliability; (a) CS-DS total scores across two time points to assess test retest reliability, (b) CS-DS total scores across two raters to assess interrater reliability.

For adults without any cognitive decline, there were no differences in CS-DS scores depending on whether the scale was administered by a researcher or completed by the informant themselves (t(103) = 0.69, p = 0.489, 95% CI (-6.63, 13.77), researcher M 83.52 SD 23.97, informant M 79.95 SD 21.96). There were also no differences in CS-DS scores depending on whether the informant was a family member or paid carer (t(103) = 0.47, p = 0.641, 95% CI (-7.22, 11.68), family member M 81.48 SD 23.45, paid carer M 79.25 SD 20.03).

Internal consistency

We next calculated the internal consistency for the three domains and the total score. Scores for Cronbach’s alpha were 0.92 for the memory domain (M 22.05 SD 7.97), 0.93 for the executive function domain (M 44.38 SD 14.11), 0.86 for the language domain (M 9.62 SD 4.89), and 0.96 for the total score (M 76.05 SD 24.73). The removal of any questions did not improve internal consistency scores for any domains. These scores suggest high internal consistency within the domains and the overall questionnaire.

Relationship between scores and cognitive decline status

Scores for adults with significant cognitive decline were significantly lower than scores for adults without significant cognitive decline age 41 and over (t(49) = 4.34, p<0.001, decline M 54.35 SD 23.70, no decline M 82.29 SD 22.20, 95% CI (14.99, 40.88)). This difference was not due to a difference in age between the groups; it remained significant when age was added as a covariate (F(1,48) = 14.93, p<0.001), and there was no significant difference in age between the groups (t(49) = -1.69, p = 0.098, decline M 51.70 SD 7.68, no decline M 48.61 SD 5.36, 95% CI (-6.77, 0.59)). This difference was also not due to a difference in pre-decline ID severity; it remained significant when ID severity was added as a covariate (F(1,48) = 18.66, p<0.001) and there was no significant difference in ID severity between the groups (χ2(2) = 4.30, p = 0.117).

Relationship between scores, general abilities, age and sex

We assessed the relationship between CS-DS scores and KBIT-2 raw scores, short ABS scores, age and sex for adults without significant cognitive decline using Spearman’s rho (see Fig 2a–2c). There were significant positive correlations between CS-DS score and both KBIT-2 total raw score and short ABS total score (KBIT-2 r = 0.56, p<0.001; short ABS r = 0.76, p<0.001). This suggests concurrent validity of CS-DS scores. We found no significant correlation with age (r = 0.09, p = 0.360), and no difference in scores between males and females (t(103) = -1.08, p = 0.283, males M 78.42 SD 22.92, females M 83.13 SD 21.82, 95% CI (-13.37, 3.95)).

Fig 2. Relationship in adults without cognitive decline between CS-DS total score and (a) KBIT-2 raw score, (b) short ABS score, and (c) age.

Exploratory factor analysis

Factor analysis was performed using the final questionnaire (see Tables 3 and 4). Factor 1 (7 questions explaining 32.1% of the variance) consisted of questions relating to individuals’ memory abilities, factor 2 (5 questions explaining 6.2% of the variance) contained questions relating to individuals’ self-regulating and inhibiting abilities, factor 3 (7 questions explaining 4.4% of the variance) contains questions relating to individuals’ self-directing and initiating abilities, factor 4 (3 questions explaining 4.2% of the variance) consists of questions relating to individuals’ communicative abilities, and factor 5 (3 questions explaining 3.4% of the variance) contains questions relating to individuals’ abilities to focus attention. Cronbach’s alpha was 0.88 (M 10.38 SD 3.94) for factor 1, 0.77 (M 6.77 SD 2.67) for factor 2, 0.87 (M 9.95 SD 4.06) for factor 3, 0.77 (M 2.77 SD 2.12) for factor 4, and 0.65 (M 2.54 SD 1.67) for factor 5. This factor model showed a good fit (root mean square error of approximation (RMSEA) = 0.046, Kaiser-Meyer-Olking (KMO) = 0.848, Bartlett’s test of sphericity χ2(1830) = 5088.4, p<0.001).


We have developed a new informant questionnaire, the CS-DS, to assess everyday cognitive abilities in people with DS. This questionnaire was developed to focus on abilities related to executive function, memory and language, which are often affected by DS. The CS-DS shows good reliability, as assessed using two raters and over two time points. The CS-DS also shows good validity, with scores being significantly lower for those with cognitive decline and correlating well with measures of general abilities. We tested the CS-DS over a large, diverse sample of individuals with DS suggesting its suitability for this population; our sample contained a wide range of ages and ID severities, as well as those with significant dementia-related cognitive decline. The domain and total scores for the CS-DS showed a wide range with minimal floor and ceiling effects, and we had few questionnaires with questions not completed due to not being appropriate. Our three domains showed high internal consistency, suggesting that the use of reverse phrasing for half of the questions reduced response set bias. The CS-DS is therefore suitable for informant ratings of cognitive abilities in adults with DS, and also likely suitable for other types of ID.

Validity of the CS-DS

To validate the CS-DS we first compared scores for those with and without significant cognitive decline aged 41+. We found significantly higher scores for those without significant cognitive decline; this was not due to an effect of age or pre-decline ID severity. We secondly determined the relationship between CS-DS total score and scores on the KBIT-2, a measure of IQ, and short ABS, a measure of adaptive abilities, for individuals without significant cognitive decline. Both measures significantly correlated with CS-DS scores, with short ABS scores correlating more significantly. This is likely due to the questions on the CS-DS relating to individuals’ everyday cognitive abilities to plan tasks, remember events and communicate. We did not find any relationship between CS-DS score and age in those who have not shown cognitive decline, suggesting that scores may be stable over adulthood in healthy individuals.

Factor structure

Our exploratory factor analysis revealed five underlying factors within the CS-DS, relating to memory, self-regulation / inhibition, self-direction / initiating, communication, and focussing attention. Based on the relationship between CS-DS scores and general abilities as assessed using the KBIT-2 and short ABS, this suggests that these aspects are important contributors towards everyday abilities. Supporting this, previous work has suggested relationships between everyday adaptive abilities and aspects of memory in individuals with DS [42], adaptive abilities, attention, language abilities and executive functioning in people with DS [29], and adaptive behaviour and inhibitory control in children with a mild ID [43].

Applications and future use of the CS-DS

The CS-DS may be of use for intervention trials. Such trials use a number of outcome measures, with both cognitive task performance and relevant informant ratings being important. The CS-DS may therefore complement test batteries to assess cognitive abilities such as the Arizona Cognitive Test Battery (ACTB) [27] or the TESDAD battery [29]. Informant ratings are an invaluable outcome measure in particular for individuals with more severe IDs who may be unable to complete cognitive tasks, those with vision or hearing problems, or those who score at the floor level of psychometric tests [32]. Informant rated tools may also be useful in identifying early symptoms of dementia [44] and to identify relevant biomarkers associated with cognitive change [45, 46].

It will also be important to determine whether the CS-DS may be useful for detecting and predicting cognitive decline longitudinally. Changes in executive function in particular have been associated with predicting cognitive decline in DS [23, 24], and so it would be of interest to determine whether changes in scores for the executive function domain are able to predict future cognitive decline. As the CS-DS is focussed on assessing cognitive abilities then it may be more sensitive to detecting early changes compared to questionnaires focussed on assessing a variety of changes related to dementia, some of which are associated with the later stages, such as the DLD [33], DSQIID [34] and DSDS [35]. Future longitudinal studies are required to determine the usefulness of this scale for assessing cognitive decline.

As discussed by Liogier d'Ardhuy et al. [32] then measures to be used in intervention studies need to be stable over time in the absence of any intervention to ensure that any change in scores using the measure are due to the intervention. Our test retest results suggest that scores on the CS-DS are stable over time, and so this may be an appropriate measure in intervention studies.

As with many informant questionnaires, responses to questions contain a degree of subjectivity. This is supported by the higher test retest reliability (i.e. same rater) compared to interrater reliability (i.e. different raters). Due to this subjectivity it will be important for longitudinal studies using the CS-DS to use the same rater wherever possible.

Finally, although we developed and validated the CS-DS using an adult population with DS, the questions may also be suitable for a child / adolescent population with DS, and also populations with another cause of ID. Future studies should explore this possibility.

Strengths and limitations

A major strength of our study is our pragmatic approach, and we included a wide range of participants, across a variety of ages and ID severities, including several individuals with visual and hearing difficulties. Using DSM-IV criteria for ID severity our sample without cognitive decline contained 43% of individuals with a severe ID, supporting the use for this scale in adults with a severe ID. This proportion is much higher than the proportion determined to have a severe ID due to carer assessment, suggesting a disconnect between carer and clinical judgement of ID severity. We recruited participants from a variety of settings, including local ID teams and individuals who had voluntarily contacted us about our research, suggesting that our sample should be broadly representative of people with DS.

However, the CS-DS will need further validation in other samples to confirm its applicability across the wider population with DS. In particular, we may have included a slight underrepresentation of individuals with a severe ID as they are less likely to take part in research studies. In addition, the majority of our sample were white British, and it will be important to confirm the validity of the questionnaire in different ethnic groups. The tool will also require further validation to assess its suitability for use in other populations with ID.

Another limitation of our study is that an exploratory factor analysis was based on a relatively small sample size, though the high internal consistency and good model fit may compensate for this. In the future it would be of interest to perform confirmatory factor analysis in a larger sample size.


We report the development of an informant scale to assess cognitive abilities in individuals with DS. Our scale shows high reliability and validity with a range of scores, and is applicable to individuals with DS across a range of ages and ID severities. In the future this scale may be useful to assess changes in cognition due to interventions or the development of cognitive decline.

Supporting Information


The authors would like to thank all the participants in this study for their time. We would also like to thank Fatima Jichi for statistical advice, and Amanda Sinai for comments during questionnaire development.

Author Contributions

Conceived and designed the experiments: CMS SH AS. Performed the experiments: CMS ER LFW. Analyzed the data: CMS ER AS. Wrote the paper: CMS ER LFW SH AS.


  1. 1. Wu J, Morris JK (2013) The population prevalence of Down's syndrome in England and Wales in 2011. Eur J Hum Genet 21: 1016–1019. pmid:23321618
  2. 2. Englund A, Jonsson B, Zander CS, Gustafsson J, Anneren G (2013) Changes in mortality and causes of death in the Swedish Down syndrome population. Am J Med Genet A 161A: 642–649. pmid:23436430
  3. 3. Startin CM, Hamburg S, Hithersay R, al Janabi T, Strydom A (submitted) LonDownS: assessing cognitive decline and cognitive abilities in people with Down syndrome.
  4. 4. Lanfranchi S, Jerman O, Dal Pont E, Alberti A, Vianello R (2010) Executive function in adolescents with Down Syndrome. J Intellect Disabil Res 54: 308–319. pmid:20202074
  5. 5. Rowe J, Lavender A, Turk V (2006) Cognitive executive function in Down's syndrome. Br J Clin Psychol 45: 5–17. pmid:16480563
  6. 6. Pennington BF, Moon J, Edgin J, Stedron J, Nadel L (2003) The neuropsychology of Down syndrome: evidence for hippocampal dysfunction. Child Dev 74: 75–93. pmid:12625437
  7. 7. Visu-Petra L, Benga O, Tincas I, Miclea M (2007) Visual-spatial processing in children and adolescents with Down's syndrome: a computerized assessment of memory skills. J Intellect Disabil Res 51: 942–952. pmid:17991001
  8. 8. Roberts JE, Price J, Malkin C (2007) Language and communication development in Down syndrome. Ment Retard Dev Disabil Res Rev 13: 26–35. pmid:17326116
  9. 9. Carducci F, Onorati P, Condoluci C, Di Gennaro G, Quarato PP, Pierallini A, et al. (2013) Whole-brain voxel-based morphometry study of children and adolescents with Down syndrome. Funct Neurol 28: 19–28. pmid:23731912
  10. 10. Pinter JD, Brown WE, Eliez S, Schmitt JE, Capone GT, Reiss AL (2001) Amygdala and hippocampal volumes in children with Down syndrome: a high-resolution MRI study. Neurology 56: 972–974. pmid:11294940
  11. 11. Teipel SJ, Schapiro MB, Alexander GE, Krasuski JS, Horwitz B, Hoehne C, et al. (2003) Relation of corpus callosum and hippocampal size to age in nondemented adults with Down's syndrome. Am J Psychiatry 160: 1870–1878. pmid:14514503
  12. 12. Edgin J (2013) Cognition in Down syndrome: a developmental cognitive neuroscience perspective. WIREs Cogn Sci 4: 307–317.
  13. 13. Jacola LM, Byars AW, Hickey F, Vannest J, Holland SK, Schapiro MB (2014) Functional magnetic resonance imaging of story listening in adolescents and young adults with Down syndrome: evidence for atypical neurodevelopment. J Intellect Disabil Res 58: 892–902. pmid:23962356
  14. 14. McCarron M, McCallion P, Reilly E, Mulryan N (2014) A prospective 14-year longitudinal follow-up of dementia in persons with Down syndrome. J Intellect Disabil Res 58: 61–70. pmid:23902161
  15. 15. Wiseman FK, Al-Janabi T, Hardy J, Karmiloff-Smith A, Nizetic D, Tybulewicz VL, et al. (2015) A genetic cause of Alzheimer disease: mechanistic insights from Down syndrome. Nat Rev Neurosci 16: 564–574. pmid:26243569
  16. 16. Seshadri S, Beiser A, Kelly-Hayes M, Kase CS, Au R, Kannel WB, et al. (2006) The lifetime risk of stroke: estimates from the Framingham Study. Stroke 37: 345–350. pmid:16397184
  17. 17. Alzheimer's A (2015) 2015 Alzheimer's disease facts and figures. Alzheimers Dement 11: 332–384. pmid:25984581
  18. 18. Chene G, Beiser A, Au R, Preis SR, Wolf PA, Dufouil C, et al. (2015) Gender and incidence of dementia in the Framingham Heart Study from mid-adult life. Alzheimers Dement 11: 310–320. pmid:24418058
  19. 19. Coppus A, Evenhuis H, Verberne GJ, Visser F, van Gool P, Eikelenboom P, et al. (2006) Dementia and mortality in persons with Down's syndrome. J Intellect Disabil Res 50: 768–777. pmid:16961706
  20. 20. Holland AJ, Hon J, Huppert FA, Stevens F, Watson P (1998) Population-based study of the prevalence and presentation of dementia in adults with Down's syndrome. Br J Psychiatry 172: 493–498. pmid:9828989
  21. 21. Margallo-Lana ML, Moore PB, Kay DW, Perry RH, Reid BE, Berney TP, et al. (2007) Fifteen-year follow-up of 92 hospitalized adults with Down's syndrome: incidence of cognitive decline, its relationship to age and neuropathology. J Intellect Disabil Res 51: 463–477. pmid:17493029
  22. 22. Tyrrell J, Cosgrave M, McCarron M, McPherson J, Calvert J, Kelly A, et al. (2001) Dementia in people with Down's syndrome. Int J Geriatr Psychiatry 16: 1168–1174. pmid:11748777
  23. 23. Adams D, Oliver C (2010) The relationship between acquired impairments of executive function and behaviour change in adults with Down syndrome. J Intellect Disabil Res 54: 393–405. pmid:20367747
  24. 24. Ball SL, Holland AJ, Treppner P, Watson PC, Huppert FA (2008) Executive dysfunction and its association with personality and behaviour changes in the development of Alzheimer's disease in adults with Down syndrome and mild to moderate learning disabilities. Br J Clin Psychol 47: 1–29. pmid:17681112
  25. 25. Kittler P, Krinsky-McHale SJ, Devenny DA (2006) Verbal intrusions precede memory decline in adults with Down syndrome. J Intellect Disabil Res 50: 1–10. pmid:16316425
  26. 26. Ball SL, Holland AJ, Hon J, Huppert FA, Treppner P, Watson PC (2006) Personality and behaviour changes mark the early stages of Alzheimer's disease in adults with Down's syndrome: findings from a prospective population-based study. Int J Geriatr Psychiatry 21: 661–673. pmid:16802281
  27. 27. Edgin JO, Mason GM, Allman MJ, Capone GT, Deleon I, Maslen C, et al. (2010) Development and validation of the Arizona Cognitive Test Battery for Down syndrome. J Neurodev Disord 2: 149–164. pmid:21274406
  28. 28. Witts P, Elders S (1998) The 'Severe Impairment Battery': assessing cognitive ability in adults with Down syndrome. Br J Clin Psychol 37 (Pt 2): 213–216. pmid:9631208
  29. 29. de Sola S, de la Torre R, Sanchez-Benavides G, Benejam B, Cuenca-Royo A, Del Hoyo L, et al. (2015) A new cognitive evaluation battery for Down syndrome and its relevance for clinical trials. Front Psychol 6: 708. pmid:26089807
  30. 30. Sinai A, Hassiotis A, Rantell K, Strydom A (submitted) Assessing specific cognitive deficits associated with dementia in older adults with Down syndrome: validity of the Arizona Cognitive Test Battery (ACTB).
  31. 31. Oliver C, Crayton L, Holland A, Hall S, Bradbury J (1998) A four year prospective study of age-related cognitive change in adults with Down's syndrome. Psychol Med 28: 1365–1377. pmid:9854278
  32. 32. Liogier d'Ardhuy X, Edgin JO, Bouis C, de Sola S, Goeldner C, Kishnani P, et al. (2015) Assessment of Cognitive Scales to Examine Memory, Executive Function and Language in Individuals with Down Syndrome: Implications of a 6-month Observational Study. Front Behav Neurosci 9: 300. pmid:26635554
  33. 33. Evenhuis HM (1992) Evaluation of a screening instrument for dementia in ageing mentally retarded persons. J Intellect Disabil Res 36 (Pt 4): 337–347. pmid:1525439
  34. 34. Deb S, Hare M, Prior L, Bhaumik S (2007) Dementia screening questionnaire for individuals with intellectual disabilities. Br J Psychiatry 190: 440–444. pmid:17470960
  35. 35. Gedye A (1995) Dementia Scale for Down Syndrome. Gedye Research & Consulting.
  36. 36. Roth RM, Isquith PK, Gioia GA (2005) Behavioral Rating Inventory of Executive Function—Adult version. Lutz FL, Psychological Assessment Resources Inc.
  37. 37. O'Shea MF (1996) The cognitive and affective correlates of the memory complaint in temporal lobe epilepsy. University of Melbourne, Melbourne.
  38. 38. Sparrow SS, Cicchetti DV, Balla DA (2007) Vineland-II: Vineland Adaptive Behavior Scales (Second edition): Pearson, Oxford, UK.
  39. 39. Ball SL, Holland AJ, Huppert FA, Treppner P, Watson P, Hon J (2004) The modified CAMDEX informant interview is a valid and reliable tool for use in the diagnosis of dementia in adults with Down's syndrome. J Intellect Disabil Res 48: 611–620. pmid:15312062
  40. 40. Kaufman AS, Kaufman NL (2004) Kaufmann brief intelligence test (Second edition): Pearson Assessments.
  41. 41. Hatton C, Emerson E, Robertson J, Gregory N, Kessissoglou S, Perry J, et al. (2001) The adaptive behavior scale-residential and community (part I): towards the development of a short form. Res Dev Disabil 22: 273–288. pmid:11523952
  42. 42. Edgin JO, Pennington BF, Mervis CB (2010) Neuropsychological components of intellectual disability: the contributions of immediate, working, and associative memory. J Intellect Disabil Res 54: 406–417. pmid:20537047
  43. 43. Gligorovic M, Buha Durovic N (2014) Inhibitory control and adaptive behaviour in children with mild intellectual disability. J Intellect Disabil Res 58: 233–242. pmid:23082754
  44. 44. Dekker AD, Strydom A, Coppus AM, Nizetic D, Vermeiren Y, Naude PJ, et al. (2015) Behavioural and psychological symptoms of dementia in Down syndrome: Early indicators of clinical Alzheimer's disease? Cortex 73: 36–61. pmid:26343344
  45. 45. Zis P, Dickinson M, Shende S, Walker Z, Strydom A (2012) Oxidative stress and memory decline in adults with Down syndrome: longitudinal study. J Alzheimers Dis 31: 277–283. pmid:22561328
  46. 46. Zis P, McHugh P, McQuillin A, Pratico D, Dickinson M, Shende S, et al. (2014) Memory decline in Down syndrome and its relationship to iPF2alpha, a urinary marker of oxidative stress. PLoS One 9: e97709. pmid:24901945