Abstract

Background: cognitive test scores and visual acuity are strongly associated in older people. This may be due to poor vision limiting performance on cognitive tasks specifically requiring vision, or an association between visual and neurodegenerative disorders.

Objective: to explore, using data from the Newcastle 85+ cohort study, the impact of sight impairment (SI) on Mini-Mental State Examination (MMSE) scores and whether reduced scores among SI participants are limited to tasks requiring vision.

Results: of 839 participants aged 85 years, 44 (5.2%) were registered SI. Median (inter-quartile range) sMMSE scores were 25 (22–29) for SI and 28 (25–29) for non-SI participants (P = 0.006). SI participants had lower subscale scores on tasks requiring vision (< 0.001 for each) but also for some subscale scores not obviously requiring vision: orientation (P = 0.018) and repetition (P = 0.030). Excluding visual items, there was no significant difference in MMSE scores between those with/without SI.

Conclusion: SI may be an obstacle to older people completing cognitive assessments including tasks requiring vision. People with SI also scored lower on some tasks not obviously requiring vision. An association between cognitive impairment and SI may exist beyond simply being unable to see the test material in cognitive tests.

Introduction

There is a strong association between impaired visual acuity and poor cognitive test scores in older people [1, 2]. This could be because those with poor vision are less able to see the test material and so score poorly on tasks requiring vision; or, alternatively, that there is an association between vision disorders and neurodegenerative conditions and those with poor vision are more likely to be cognitively impaired.

The Mini-Mental State Examination (MMSE) is a widely used clinical tool for assessing cognition [3]. The MMSE is scored out of 30 points and includes a number of items which explicitly require vision, including: naming two objects, following a written command, writing a sentence and copying a line drawing. Vision is also likely to have a role, albeit less central, in the praxis of the three stage command. The removal of all these tasks requiring vision gives a score out of 22, which has been described previously as the MMblind [4].

The Newcastle 85+ study is a population-based cohort study which aims to examine the health trajectories and outcomes of very old people in the North East of England [5, 6]. Using data from the baseline phase of the Newcastle 85+ study, we aimed to explore the impact of sight impairment on MMSE scores in people aged around 85 years. We aimed to investigate whether reduced scores on the MMSE in those with sight impairment could be either fully accounted for by poor vision affecting an individual's ability to complete the tasks or suggested an association between cognitive impairment and visual impairment.

Methods

The protocol for the Newcastle 85+ study has been published previously [5, 6]. In brief, the sampling frame comprised all people born in 1921, aged around 85 when recruitment commenced in 2006, who were permanently registered with a participating general practice in Newcastle upon Tyne or North Tyneside NHS Primary Care Trusts, in the UK. Participation at baseline entailed a detailed multidimensional health assessment and a review of medical records held by the general practice. Assessment was carried out in the participant's usual residence (home or institution) by a research nurse and included the standardised MMSE (sMMSE) [5–7]. If any sMMSE items were not asked by the nurse, or if the participant refused to answer, then the reasons were recorded but the missing item scored zero points. The MMblind was calculated by removing all tasks that involve vision to give a score out of 22 [4].

Data were collected from the general practice records on whether the participant was registered as blind or partially sighted, or neither. In the UK, registration as severely sight impaired (blind) or sight impaired (partially sighted) requires the completion of a certificate of visual impairment by a consultant ophthalmologist [8]. The current analysis was based on the 851 participants who had both health assessment and review of general practice records (59% of the 1,453 eligible to take part). Comparison of these with non-responders and those who agreed to record review only (n = 188) showed little difference although there were slightly fewer women in the health assessment group [6].

Data analysis was carried out using the Statistical Package for Social Sciences (SPSS version 17). A Mann–Whitney U test was used to assess differences in sMMSE scores between those with and without impaired sight registration. The subscore analysis was carried out using Chi-squared tests for subscores out of 1 point and Mann–Whitney U tests for subscores greater than 1 point. All reported P-values are two-tailed and a P-value of ≤ 0.05 is considered statistically significant.

Results

A total of 839 participants (99% of the 851 who had both health assessment and general practice record review) had complete data on sMMSE and vision. Of these, 2.4% (20/839) were registered severely sight impaired (blind) and 2.9% (24/839) as sight impaired (partially sighted). No significant difference was found between the educational status (total years in education) of those registered severely sight impaired/sight impaired and those not registered (two-sample t-test, P = 0.799). Of those registered sight impaired, 9% (4/44) had a diagnosis of dementia in the general practice records compared with 8% (64/795) of those not registered; this difference was not significant (χ2 test, P = 0.806).

Participants who were registered as sight impaired were more likely than those not registered to have one or more questions of the sMMSE not asked by the research nurse (χ2 test, P < 0.001) and were more likely to refuse items of the sMMSE (P = 0.014). In no cases were the orientation, registration, calculation or recall questions not asked.

A significant difference in total sMMSE scores was seen between those who were registered sight impaired or severely sight impaired when compared with those who were not. The median sMMSE score for the sight impaired and severely sight impaired group was 25 (inter-quartile range 22–29), and for those without sight impairment was 28 (inter-quartile range 25–29), (Mann–Whitney U test, P = 0.006). Table 1 shows the comparison of the sMMSE subscale scores. As expected, those with sight impairment had lower scores on items requiring vision including naming two objects (Mann–Whitney U test, P < 0.001), writing a sentence (χ2 test, P < 0.001), copying a diagram (χ2 test, < 0.001) and reading and following an instruction (χ2 test, P < 0.001). However, we also found that those registered sight impaired had lower scores on the orientation subscale (which included orientation in time and place; Mann–Whitney U test, P = 0.018) and repetition of a phrase (χ2 test, P = 0.030) which do not require vision.

Table 1.

Comparison of standardised MMSE subscale scores according to sight registration

  SI/SSI (n = 44)
 
Non-SI (n = 795)
 
P-value* 
Median (IQR) Whole section correct [n (%)] Median (IQR) Whole section correct [n (%)] 
Tasks not requiring vision 
 Orientation (0–10) 10 (8–10) 25 (57) 10 (9–10) 568 (71) 0.018** 
 Registration (0–3) 3 (3–3) 42 (95) 3 (3–3) 753 (95) 0.848 
 Calculation (0–5) 5 (3–5) 31 (70) 5 (4–5) 596 (80) 0.419 
 Recall (0–3) 2.5 (1–3) 22 (50) 2 (1–3) 320 (40) 0.444 
 Language: repetition (0–1) NA 16 (36) NA 423 (53) 0.030** 
Tasks which require vision 
 Language: following a written command (0–1) NA 32 (73) NA 765 (96) <0.001** 
 Language: 3 stage command (0–3) 3 (2–3) 27 (61) 3 (2–3) 586 (74) 0.063 
 Language: writing a sentence (0–1) NA 28 (64) NA 722 (91) <0.001** 
 Language: naming two objects (0–2) 2 (2–2) 35 (80) 2 (2–2) 772 (97) <0.001** 
 Visuospatial abilities: copying a diagram (0–1) NA 14 (32) NA 629 (79) <0.001** 
  SI/SSI (n = 44)
 
Non-SI (n = 795)
 
P-value* 
Median (IQR) Whole section correct [n (%)] Median (IQR) Whole section correct [n (%)] 
Tasks not requiring vision 
 Orientation (0–10) 10 (8–10) 25 (57) 10 (9–10) 568 (71) 0.018** 
 Registration (0–3) 3 (3–3) 42 (95) 3 (3–3) 753 (95) 0.848 
 Calculation (0–5) 5 (3–5) 31 (70) 5 (4–5) 596 (80) 0.419 
 Recall (0–3) 2.5 (1–3) 22 (50) 2 (1–3) 320 (40) 0.444 
 Language: repetition (0–1) NA 16 (36) NA 423 (53) 0.030** 
Tasks which require vision 
 Language: following a written command (0–1) NA 32 (73) NA 765 (96) <0.001** 
 Language: 3 stage command (0–3) 3 (2–3) 27 (61) 3 (2–3) 586 (74) 0.063 
 Language: writing a sentence (0–1) NA 28 (64) NA 722 (91) <0.001** 
 Language: naming two objects (0–2) 2 (2–2) 35 (80) 2 (2–2) 772 (97) <0.001** 
 Visuospatial abilities: copying a diagram (0–1) NA 14 (32) NA 629 (79) <0.001** 

SSI, severely sight impaired; SI, sight impaired; IQR, inter-quartile range.

*P-values relate to the Pearson Chi-squared test for scores of 0 or 1 points, and Mann–Whitney U comparison of medians for scores of >1 point.

**Indicates significant P-value.

When all items requiring vision were removed to create the MMblind score out of 22 [4], those registered sight impaired or severely sight impaired had a median score of 20 (inter-quartile range: 17–21) and those not registered had a median score of 20 (inter-quartile range: 18–21), this difference was not significant (Mann–Whitney U test, P = 0.344).

Discussion

Poor vision may be an obstacle for older people completing the MMSE. In this large cohort study, those registered as sight impaired had lower scores on the sMMSE compared with those who were not registered. The reasons for this are likely to include the higher rates of items not being asked or refused among sight impaired participants, and a difficulty seeing the test material leading to poor scores on attempted items. However, those with visual impairment also had lower scores on some items not obviously requiring vision including orientation and repetition, and all sight impaired participants were asked these questions. This could support the theory that visual and cognitive decline co-occur [1, 2, 9] and it fits in with the findings of others who have shown that the effects of poor vision on cognition are not modality specific; impacting on both visual and verbal tasks [10]. There may also be subtle visual cues involved in tests of orientation in time and place, for example reading the newspaper (for date) or reading street signs (for place); and there may be some advantage in being able to see the interviewer clearly in repetition tasks. The MMblind scores were similar between groups, so did not support a hypothesis of co-occurrence of cognitive and visual decline. This may, in part be due to a compensatory improved performance on tasks such as recall among sight impaired participants. A key strength of these results is that they are from a large population-based study which had good response rates given the age group being assessed and the intensity of the assessment protocol. The assessments were based in participants' homes or institutions which is important to achieve a representative sample in a group who may be unwilling or unable to travel for the clinic-based assessment. Thus the sample importantly included those in institutions and those with cognitive impairment and was socio-demographically broadly representative of the local population, and of England and Wales, except for slightly less ethnic diversity [6]. The sMMSE assessment included collection of data on any missing items in the sMMSE and the reasons why missing. A further strength was the use of general practice medical records for data on sight registration as a robust measure of visual impairment. However, the study was not without limitations. We did not measure visual acuity. The rates of registered sight impairments were lower than reported national prevalence data [11] where visual acuity has been measured, suggesting that this was under-recognised in this study. We have also not looked at the causes of visual impairment, and some causes of visual impairment might be more likely to be associated with cognitive impairment than others. While we have shown an association between orientation and language scores with sight registration, the analysis involves multiple comparisons and caution should be taken in drawing firm conclusions from these associations.

In summary, it is important to consider the possibility of vision impairment in older people when carrying out the MMSE and to consider using the MMblind [4]. Beyond this however, visual impairment may be a marker for cognitive impairment.

Key points

  • Visual impairment may be an obstacle to older people completing standard cognitive assessments such as the MMSE.

  • Poorer scores are seen in those with sight impairment on tasks requiring vision as well as some tasks not requiring vision.

  • An association between visual impairment and cognitive impairment may exist beyond simply being unable to see the test material in cognitive tasks.

  • Optimising vision for older people with sight impairment before cognitive testing and using adapted versions of the MMSE may be useful.

Conflicts of interest

None declared.

Funding

This work was supported by a combined grant from the Medical Research Council and Biotechnology and Biological Sciences Research Council (grant number G0500997) and the National Institute of Health Research (DRF-2010-03-71 to J.M.J.).

Acknowledgements

Thanks are especially due to the 85 year olds of Newcastle and North Tyneside for the generous donation of their time and personal information. In addition we thank: the research nurses; data manager (Pauline Potts); project secretary (Lucy Farfort) and NHS North of Tyne and local general practices.

References

1
Lindenberger
U
Baltes
P
Sensory functioning and intelligence in old age: a strong connection
Psychol Aging
 , 
1994
, vol. 
9
 (pg. 
339
-
55
)
2
Anstey
KJ
Luszcz
MA
Hofer
SM
A latent growth curve analysis of late-life sensory and cognitive function over 8 years: evidence for specific and common factors underlying change
Psychol Aging
 , 
2003
, vol. 
18
 (pg. 
714
-
26
)
3
Folstein
MF
Folstein
SE
McHugh
PR
‘Mini-mental state’. A practical method for grading the cognitive state of patients for the clinician
J Psychiatr Res
 , 
1975
, vol. 
12
 (pg. 
189
-
98
)
4
Busse
A
Sonntag
A
Bischkopf
J
, et al.  . 
Adaptation of dementia screening for vision-impaired older persons: administration of the Mini-Mental State Examination (MMSE)
J Clin Epidemiol
 , 
2002
, vol. 
55
 (pg. 
909
-
15
)
5
Collerton
J
Barrass
K
Bond
J
, et al.  . 
The Newcastle 85+ study: biological, clinical and psychosocial factors associated with healthy ageing: study protocol
BMC Geriatr
 , 
2007
, vol. 
7
 pg. 
14
 
6
Collerton
J
Davies
K
Jagger
C
, et al.  . 
Health and disease in 85 year olds: baseline findings from the Newcastle 85+ cohort study
BMJ
 , 
2009
, vol. 
339
 pg. 
b4904
 
7
Molloy
DW
Standish
TI
A guide to the standardized Mini-Mental State Examination
Int Psychogeriatr
 , 
1997
, vol. 
9
 
Suppl. 1
(pg. 
87
-
94
discussion 143–50
8
Department of Health
Form CVI: explanatory notes for consultants ophthalmologists and hospital eye clinic staff
2003
9
Lindenberger
U
Scherer
H
Baltes
P
The strong connection between sensory and cognitive performance in old age: not due to sensory acuity reductions operating during cognitive assessment
Psychol Aging
 , 
2001
, vol. 
16
 (pg. 
196
-
205
)
10
Drobny
JV
Anstey
KJ
Andrews
S
Visual memory testing in older adults with age-related visual decline: a measure of memory performance or visual functioning?
J Clin Exp Neuropsychol
 , 
2005
, vol. 
27
 (pg. 
425
-
35
)
11
Evans
JR
Fletcher
AE
Wormald
RP
, et al.  . 
Prevalence of visual impairment in people aged 75 years and older in Britain: results from the MRC trial of assessment and management of older people in the community
Br J Ophthalmol
 , 
2002
, vol. 
86
 (pg. 
795
-
800
)

Comments

0 Comments