Abstract
Introduction:
In view of the latest evidence, dementia is also regarded as type 3 diabetes mellitus (T3DM) due to the important role of brain insulin resistance in its pathogenesis. Our study objectives were to find out the proportion and predictors of mild cognitive impairment (MCI) and T3DM (dementia) among the elderly people living with and without type 2 diabetes mellitus (T2DM) attending a medical college outpatient department (OPD).
Materials and Methods:
A cross-sectional study was conducted comparing elderly patients from diabetes clinic OPD and general medicine OPD of a medical college in West Bengal. Two groups consisting of 70 participants in each, based on their diabetic status, were made. A predesigned, pretested, and semistructured schedule was prepared incorporating the Alzheimer’s questionnaire (AQ). Chi-square and logistics regression were used to find the predictors of MCI, T3DM in each group, and the median AQ score was compared using Mann–Whitney U-test.
Results:
Among people with T2DM (PwD), 27 (38.6%) had MCI, and 17 (24.3%) had T3DM. Among people without T2DM, 45 (64.3%) had normal cognition. There was a significant difference in the median AQ score of both groups. A significant association was found between cognitive status and T2DM status (P = 0.004). Among diabetics, female gender (adjusted odds ratio [AOR]: 15.38, P = 0.012), primary education (AOR 27.14, P = 0.027), and family history of dementia (AOR 27.65, P = 0.013) were found to be predictors of dementia.
Conclusion:
Significant difference in the proportion of MCI and T3DM among elderly PwD and elderly nondiabetics was noted. Regular screening for T2DM and cognitive status is need of the hour.
Keywords: Alzheimer’s questionnaire, cognitive dysfunction, cognitive status, dementia, diabetes mellitus
Résumé
Introduction:
À la lumière des preuves récentes, la démence est également considérée comme un diabète de type 3 (MDT3) en raison du rôle important de la résistance à l’insuline cérébrale dans sa pathogenèse. Les objectifs de notre étude étaient de déterminer la proportion et les facteurs prédictifs de l’atteinte cognitive légère (MCI) et du MDT3 (démence) chez les personnes âgées vivant avec ou sans diabète sucré de type 2 (MDT2) consultant dans un service ambulatoire (OPD) d’un collège médical.
Matériels et Méthodes:
Une étude transversale a été menée en comparant les patients âgés fréquentant le service de diabétologie et le service de médecine générale d’un collège médical au Bengale occidental. Deux groupes de 70 participants chacun, selon leur statut diabétique, ont été constitués. Un questionnaire prédéfini, prétesté et semi-structuré a été utilisé, incluant le questionnaire d’Alzheimer (AQ). Le test du chi carré et une régression logistique ont été utilisés pour identifier les prédicteurs de MCI et de MDT3 dans chaque groupe, et la médiane du score AQ a été comparée à l’aide du test de Mann–Whitney U.
Résultats:
Parmi les personnes atteintes de MDT2 (PwD), 27 (38,6 %) présentaient une MCI et 17 (24,3 %) souffraient de DT3. Parmi les personnes sans DT2, 45 (64,3 %) avaient une cognition normale. Une différence significative du score médian AQ a été observée entre les deux groupes. Une association significative a été trouvée entre l’état cognitif et le statut diabétique (P = 0,004). Chez les diabétiques, le sexe féminin (odds ratio ajusté [ORA] : 15,38, P = 0,012), le niveau d’éducation primaire (ORA 27,14, P = 0,027) et les antécédents familiaux de démence (ORA 27,65, P = 0,013) ont été identifiés comme prédicteurs de la démence.
Conclusion:
Une différence significative dans la proportion de MCI et de MDT3 a été observée entre les personnes âgées diabétiques et non diabétiques. Un dépistage régulier du MDT2 et de l’état cognitif est essentiel.
Mots-clés: Questionnaire d’Alzheimer, dysfonctionnement cognitif, état cognitif, démence, diabète sucré
INTRODUCTION
The world population is aging at an unprecedented rate and posing some serious public health challenges.[1] Diabetes mellitus (DM) is one of the leading causes of morbidity and mortality globally as well as in India.[2,3] The prevalence of self-reported chronic disease was 21% and DM accounted for 31% of all chronic diseases among elderly Indians.[4] The World Health Organization has declared dementia as a public health priority and has announced the global action plan (2017–2025) on dementia.[5]
The latest research suggests a possible link between insulin resistance and Alzheimer’s disease (AD) based on the fact that both the processing and the clearance of beta-amyloid protein are caused by the insulin signaling pathway.[6,7] The link between insulin resistance and AD is strong enough for many researchers to call it type 3 DM (T3DM).[8] The relationship between type 2 DM (T2DM) and cognitive performance is intricate and remains unanticipated due to the presence of other factors such as age, socioeconomic status, marital status, education level, and other comorbidities.[9]
The worldwide prevalence of dementia will almost double in every 20 years and expected to reach 78 million in 2030 and 139 million in 2050.[10] Despite being a common medical illness, its screening and treatment are yet to gain popularity.
In this context, the present study was carried out to determine the proportion and predictors of mild cognitive impairment (MCI) and T3DM (dementia) among the elderly people living with and without T2DM attending the outpatient department of a medical college in West Bengal.
MATERIALS AND METHODS
Study design, setting, and population
An analytical, hospital-based cross-sectional study was conducted from December 2023 to February 2024 comparing elderly (≥60 years) patients from the diabetes clinic outpatient department (OPD) and general medicine OPD of a medical college in West Bengal. Study participants were selected as per predefined inclusion and exclusion criteria [Figure 1].
Figure 1.
Flowchart showing selection criteria of the study participants and sampling strategy of the study
Sample size and sampling technique
Minimum sample size was calculated using the formula: 
, considering value of Z1−α/2 = 1.96 at 95% confidence interval, absolute error (L) 15%, prevalence of CI in elderly people without T2DM 26% (P1) while the prevalence of CI in elderly people with T2DM (PwD) 30% (P2).[11,12] Calculated sample size was 69. Hence, from each group, 70 participants were taken, and a total of 140 participants were selected.
Systematic random sampling was done considering clinic register as sampling frame. The sampling interval for diabetes and general medicine OPD was considered 5 and 7, respectively [Figure 1].
Data collection
A predesigned, pretested, and semistructured schedule incorporating clinicosocial Characteristics and the Alzheimer’s questionnaire (AQ) was used.
The AQ (developed by Banner Sun Health Research Institute, Cleo Roberts Centre for Clinical Research, 10515 W. Santa Fe Dr. Sun City) is a validated tool for the screening of cognitive dysfunction among elderly.[13] AQ is a 21-item informant-based assessment designed for ease of use in the clinical setting. AQ items are divided into five domains including memory (6 items), orientation (3 items), functional ability (7 items), visuospatial (2 items), and language (3 items). It demonstrated high sensitivity and specificity for detecting MCI (sensitivity 89% and specificity 91%) and AD (sensitivity 99% and specificity 96%). Area under curve values also indicated high diagnostic accuracy for both MCI (0.95) and AD (0.99). Internal consistency of the AQ is also high (Cronbach’s alpha = 0.89).[14] The AQ total score is based on the sum of points for items with a yes response. The total score ranges from 0 to 27 with higher scores corresponding to greater impairment.[14]
Interview of caregivers, review of medical records, anthropometric measurements (height and weight for body mass index [BMI]), and laboratory investigations (fasting blood sugar, postprandial blood sugar, and glycosylated hemoglobin [HbA1c]) were done.
Operational definition
Cognitive impairment
It can be defined as problems with a person’s ability to think, learn, remember, use judgment, and make decisions. Signs of CI include memory loss and trouble concentrating, completing tasks, understanding, remembering, following instructions, and solving problems. Other common signs may include changes in mood or behavior, loss of motivation, and being unaware of surroundings.[15]
It includes MCI and dementia.
MCI: It is a condition in which people have more memory or thinking problems than other people of their age. The symptoms of MCI are not as severe as those of AD or a related dementia. People with MCI can usually take care of themselves and carry out their normal daily activities.[16]
Dementia: Dementia is a collective name for brain syndromes, which affects memory, thinking, behavior, and emotion, and is the leading cause of disability and dependency among the elderly.[17]
It refers to AD (the most common cause of dementia) and other dementia subtypes, including vascular dementia, dementia with Lewy bodies, frontotemporal dementia, mixed forms of dementia, and other rarer types of dementia.[18,19]
In our study, AQ score of 0–4, 5–14, and 15–27 was considered as normal cognition, MCI, and dementia (T3DM), respectively.
Statistical analysis
Data were entered into Microsoft Office Excel. Entered data were checked twice for correctness and consistency. Data analysis was done using IBM SPSS software Version 24.0 (IBM Corp. Released 2016. IBM SPSS Statistics for Windows, Version 24.0. Armonk, NY, USA: IBM Corp.) and Jamovi software (The jamovi project, Version 2.6.44, Sydney, Australia).[20,21] For descriptive statistics, categorical data were presented in frequency and percentages, and continuous data were presented as mean ± standard deviation or median, interquartile range (IQR), and range. Appropriate figures were used for the graphical representation of the data.
Chi-square test, Mann–Whitney U-test, and univariate and multivariate logistic regression were used. Biologically plausible covariates showing significant association in the univariate model were selected in the final multivariate multinomial logistic regression model. P < 0.05 was considered statistically significant.
In the univariate multinomial logistic regression model MCI, dementia (T3DM) was taken as dependent variables with normal cognition as reference category. Age, gender, level of education, below poverty line (BPL) family, BMI, tobacco usage, family history of dementia, and HbA1c were taken as independent variables.
In the multivariate multinomial logistic regression model MCI, dementia (T3DM) was taken as dependent variables with normal cognition as reference category. Age, gender, level of education, BPL family, tobacco usage, and family history of dementia were taken as independent variables.
RESULTS
Clinicosocial attributes
The total number of study participants was 140 (70 in each group).
For people with type 2 diabetes mellitus
More than three-fourth (75.7%) of the study participants belonged to the age group of 60 years to 70 years. Median age (IQR) was 65 (62, 68.50) years. Age was not normally distributed as was evident from significant Kolmogorov–Smirnov test (P < 0.05), positively skewed distribution (mode < median < mean), and characteristic Q-Q plot. Half (50%) of the study participants were male. Two-fifth (40%) of the study participants received only primary education. More than half (51.4%) possessed BPL card. More than two-fifths (47.1%) were known tobacco users. More than half (52.9%) were obese. More than one-third (35.7%) had a family history of dementia [Table 1]. Nearly three-fourths (74.3%) had HbA1c value ≥7%. Median (IQR) for HbA1c was 7.4% (6.9, 8.2) [Table 1].
Table 1.
Clinicosocial characteristics of study participants (n=140)
| Variables | Subcategories | PwD (n=70), n (%) | People without T2DM (n=70), n (%) | Test statistics | P |
|---|---|---|---|---|---|
| Age (years) | Median (IQR) | 65 (62.0–68.7) | 65 (61.7–70.5) | 2549.0 | 0.677‡ |
| Gender | Male | 35 (50.0) | 29 (41.4) | 1.036 | 0.309† |
| Female | 35 (50.0) | 41 (58.6) | |||
| Level of education | Primary (I-V) | 28 (40.0) | 30 (42.9) | 0.149 | 0.928† |
| Secondary (VI-X) | 26 (37.1) | 24 (34.3) | |||
| HS and above | 16 (22.9) | 16 (22.9) | |||
| BPL family | Yes | 36 (51.4) | 37 (52.9) | 0.029 | 0.866† |
| No | 34 (48.6) | 33 (47.1) | |||
| Tobacco usage | Absent | 37 (52.9) | 47 (67.1) | 2.976 | 0.084† |
| Present | 33 (47.1) | 23 (32.9) | |||
| Family history of dementia | Absent | 45 (64.3) | 58 (82.9) | 6.208 | 0.013† |
| Present | 25 (35.7) | 12 (17.1) | |||
| AQ score | Median (IQR) | 7.0 (3.0–14.2) | 4.0 (3.0–7.2) | 1862.5 | 0.014‡ |
| HbA1c level (%) | Median (IQR) | 7.4 (6.9–8.2) | 5.2 (5.0–5.8) | 89.0 | <0.001‡ |
| BMI (kg/m2) | Median (IQR) | 23.7 (20.9–25.7) | 22.2 (19.4–24.3) | 1907.0 | 0.024‡ |
†Chi-square test, ‡Mann–Whitney U-test, Values in bold indicate statistical significance. IQR=Interquartile range, PwD=People with type 2 diabetes mellitus, T2DM=Type 2 diabetes mellitus, HS=Higher secondary, BPL=Below poverty line, AQ=Alzheimer’s Questionnaire, HbA1c=Glycosylated hemoglobin, BMI=Body mass index
For people without type 2 diabetes mellitus
More than three-fifth (68.6%) of the study participants belonged to the age group of 60 years to 70 years. Median age (IQR) was 65 (61.7, 70.5) years. Majority (58.6%) of the study participants were female. More than two-fifth (42.9%) of the study participants received only primary education. More than half (52.9%) possessed BPL card. Almost one-third (32.9%) were known tobacco users. Almost one-third (32.9%) were obese. Less than one-fifth (17.1%) had a family history of dementia [Table 1].
There was no significant difference in the median age of both the study groups (Mann–Whitney U-test was statistically not significant with P = 0.677) [Table 1].
Mild cognitive impairment and type 3 diabetes mellitus among the study participants
Among the study participants with T2DM, majority (38.6%) had MCI and nearly one-fourth (24.3%) had dementia, whereas among the study participants without T2DM, majority (64.3%) had normal cognition [Table 2].
Table 2.
Comparison of cognitive status among people living with type 2 diabetes mellitus and people without type 2 diabetes mellitus (n=140)
| Cognitive status | People with T2DM (n=70), n (%) | People without T2DM (n=70), n (%) | χ2 (df) | P | Cramer’s V |
|---|---|---|---|---|---|
| Normal | 26 (37.1) | 45 (64.3) | 11.051 (2) | 0.004 | 0.281 |
| MCI | 27 (38.6) | 18 (25.7) | |||
| Dementia | 17 (24.3) | 7 (10.0) | |||
| Total | 70 (100.0) | 70 (100.0) |
Values in bold indicate statistical significance. χ2: Chi Square, df=degrees of freedom, MCI=Mild cognitive impairment
In Pearson’s Chi (χ)-square test, statistically significant association was found between cognitive status (MCI and dementia) and T2DM status (χ2 = 11.051, df = 2 and P = 0.004). Cramer’s V value was 0.281 showing positive, medium effect size [Table 2].
For PwD, the median AQ score (IQR) was 7.0 (3.0, 14.2) while for people without T2DM, median AQ score (IQR) was 4.0 (3.0, 7.2) [Figure 2]. There was a significant difference in the median AQ score of both the study groups (Mann–Whitney U-test was statistically significant with P = 0.014) [Table 1].
Figure 2.
Violin plot with box plot showing distribution of Alzheimer’s questionnaire score stratified by type 2 diabetes mellitus status (n = 140)
For PwD, median BMI (IQR) was 23.7 (20.9, 25.7) while for people without T2DM median BMI (IQR) was 22.2 (19.4, 24.3). There was a significant difference in the median BMI of both the study groups (Mann–Whitney U-test was statistically significant with P = 0.024) [Table 1].
For PwD, median HbA1c (IQR) was 7.4% (6.9, 8.2) while for people without T2DM median HbA1c (IQR) was 5.2 (5.5, 5.8). There was a significant difference in the median HbA1c score of both the study groups (Mann–Whitney U-test was statistically significant with P < 0.001) [Table 1].
In Pearson’s Chi (χ)-square test, a significant association was found between diabetic status and family history of dementia (P = 0.013) [Table 1].
Risk factors of mild cognitive impairment and type 3 diabetes mellitus
Univariate multinomial logistic regression was done to find out the risk factors of MCI and dementia (T3DM) among both groups.
For people with type 2 diabetes mellitus
Female gender (OR 6.10, P = 0.004), educational attainment only till primary level (OR 5.85, P = 0.027), BPL family (OR 11.67, P < 0.001), tobacco usage (OR 7.14, P = 0.002), and family history of dementia (OR 26.92, P = 0.003) were found to be the significant risk factors of MCI [Table 3].
Table 3.
Univariate multinomial logistic regression showing risk factors of mild cognitive impairment and dementia among people with type 2 diabetes mellitus and people without type 2 diabetes mellitus (n=140)
| Variables | Subcategories | People with T2DM (n=70) | People without T2DM (n=70) | ||||||
|---|---|---|---|---|---|---|---|---|---|
|
|
|
||||||||
| MCI | Dementia | MCI | Dementia | ||||||
|
|
|
|
|
||||||
| OR | P | OR | P | OR | P | OR | P | ||
| Age (years) | 1.06 | 0.312 | 1.14 | 0.040 | 1.38 | <0.001 | 1.27 | 0.004 | |
| Gender | Female | 6.10 | 0.004 | 19.61 | <0.001 | 10 | 0.004 | 3.13 | 0.200 |
| Male | 1 | 1 | 1 | 1 | |||||
| Education | Primary | 5.85 | 0.027 | 12.38 | 0.010 | 18.91 | 0.008 | 1.77 | 0.567 |
| Secondary | 1.25 | 0.755 | 1.39 | 0.744 | 0.62 | 0.746 | 0.62 | 0.651 | |
| HS and above | 1 | 1 | 1 | 1 | |||||
| BPL family | Yes | 11.67 | <0.001 | 3.75 | 0.049 | 14.50 | 0.001 | 4.53 | 0.091 |
| No | 1 | 1 | 1 | 1 | |||||
| Tobacco usage | Present | 7.14 | 0.002 | 7.69 | 0.004 | 2.27 | <0.001 | 4.88 | 0.072 |
| Absent | 1 | 1 | 1 | 1 | |||||
| Family history of dementia | Present | 26.92 | 0.003 | 35.71 | 0.002 | 28 | 0.003 | 58.67 | 0.001 |
| Absent | 1 | 1 | 1 | 1 | |||||
| HbA1c | 1.16 | 0.373 | 1.23 | 0.244 | 0.79 | 0.709 | 2.99 | 0.200 | |
| BMI | 1.00 | 0.971 | 1.02 | 0.861 | 0.83 | 0.291 | 1.01 | 0.931 | |
Values in bold indicate statistical significance. P: Probability value. OR=Odds ratio, MCI=Mild cognitive impairment, BMI=Body mass index, BPL=Below poverty line, HbA1c=Glycosylated hemoglobin, T2DM=Type 2 diabetes mellitus, HS=Higher secondary
Advanced age (OR 1.14, P = 0.04), female gender (OR 19.61, P < 0.001), educational attainment only till primary level (OR 12.38, P = 0.01), BPL family (OR 3.75, P = 0.049), tobacco usage (OR 7.69, P = 0.004), and family history of dementia (OR 35.71, P = 0.002) were found to be the significant risk factors of dementia [Table 3].
For people without type 2 diabetes mellitus
Advanced age (OR 1.38, P < 0.001), female gender (OR 10, P = 0.004), educational attainment only till primary level (OR 18.91, P = 0.008), BPL family (OR 14.50, P = 0.001), tobacco usage (OR 2.27, P < 0.001), and family history of dementia (OR 28, P = 0.003) were found to be the significant risk factors of MCI [Table 3].
Advanced age (OR= 1.27, P = 0.004) and family history of dementia (OR= 58.67, P = 0.001) were found to be the significant risk factors of dementia [Table 3].
Predictors of mild cognitive impairment and type 3 diabetes mellitus
Multivariate multinomial logistic regression was done to find out predictors of MCI and Dementia (T3DM) among both the groups.
For people with type 2 diabetes mellitus
Tobacco usage (adjusted odds ratio AOR= 6.13, P = 0.045) and family history of dementia (AOR= 19.45, P = 0.016) were found to be significant predictors of MCI while poverty (BPL family) was slightly flattened (AOR= 5.47 P = 0.059) [Table 4].
Table 4.
Multivariate multinomial logistic regression showing predictors of mild cognitive impairment and dementia among people with type 2 diabetes mellitus and people without type 2 diabetes mellitus (n=140)
| Variables | Sub-categories | People with T2DM (n=70) | People without T2DM (n=70) | ||||||
|---|---|---|---|---|---|---|---|---|---|
|
|
|
||||||||
| MCI | Dementia | MCI | Dementia | ||||||
|
|
|
|
|
||||||
| AOR | P | AOR | P | AOR | P | AOR | P | ||
| Age (years) | 0.93 | 0.451 | 1.02 | 0.843 | 1.33 | 0.028 | 1.25 | 0.073 | |
| Gender | Female | 4.61 | 0.102 | 15.38 | 0.012 | 2.44 | 0.631 | 6.94 | 0.281 |
| Male | 1 | 1 | 1 | 1 | |||||
| Education | Primary | 4.25 | 0.187 | 27.14 | 0.027 | 2.91 | 0.537 | 0.15 | 0.257 |
| Secondary | 1.22 | 0.853 | 2.21 | 0.590 | 0.142 | 0.376 | 0.18 | 0.339 | |
| HS and above | 1 | 1 | 1 | 1 | |||||
| BPL family | Yes | 5.47 | 0.059 | 0.72 | 0.759 | 20.34 | 0.057 | 28.28 | 0.075 |
| No | 1 | 1 | 1 | 1 | |||||
| Tobacco usage | Present | 6.13 | 0.045 | 5.32 | 0.112 | 12.50 | 0.038 | 0.443 | 0.670 |
| Absent | 1 | 1 | 1 | 1 | |||||
| Family history of dementia | Present | 19.45 | 0.016 | 27.65 | 0.013 | 35.09 | 0.070 | 379.17 | 0.01 |
| Absent | 1 | 1 | |||||||
Values in bold indicate statistical significance. Goodness of fit of this model, in PwD group: Likelihood ratio test (P<0.001), deviance Chi-square test (P=0.971), Nagelkerke R2=0.666, Cox and Snell R2=0.589, and McFadden R2=0.412. In people without T2DM Group: Likelihood ratio test (P<0.001), deviance Chi-square test (P=1.000), Nagel Kerke R2=0.791, Cox and Snell R2=0.650 and McFadden R2=0.412. P=Probability value, AOR=Adjusted odds ratio, MCI=Mild cognitive impairment, T2DM=Type 2 diabetes mellitus, HS=Higher secondary, BPL=Below poverty line
Female gender (AOR= 15.38, P = 0.012), educational attainment only till primary level (AOR= 27.14, P = 0.027), and family history of dementia (AOR= 27.65, P = 0.013) were found to be significant predictors of dementia [Table 4].
For people without type 2 diabetes mellitus
Advanced age (AOR= 1.33, P = 0.028) and tobacco usage (AOR= 12.50, P = 0.038) were found to be significant predictors of MCI while family history of dementia (AOR= 35.09, P = 0.07) was slightly flattened [Table 4].
Family history of dementia (AOR= 379.17, P = 0.01) was found to be a significant predictor of dementia while advanced age (AOR= 1.25, P = 0.073) and poverty (BPL family) was slightly flattened (AOR= 28.28, P = 0.075) [Table 4].
Model fitting statistics
For people with type 2 diabetes mellitus
Null hypothesis was rejected in the likelihood ratio test (P < 0.001) and retained in the deviance Chi-square test (P = 0.971). The proportion of total variance was explained by the model, i.e., coefficient of determination (Nagelkerke R2 = 0.666, Cox and Snell R2 = 0.589, and McFadden R2 = 0.412), thus determining adequacy of regression fit of the adjusted model.
For people without type 2 diabetes mellitus
Null hypothesis was rejected in likelihood ratio test (P < 0.001) and retained in the deviance Chi-square test (P = 1.000). The proportion of total variance was explained by the model, i.e., coefficient of determination (Nagelkerke R2 = 0.791, Cox and Snell R2 = 0.650, and McFadden R2 = 0.412), thus determining the adequacy of regression fit of the adjusted model.
DISCUSSION
Despite being a common comorbidity, the study comparing CI and T3DM (dementia) among elderly living with and without T2DM is very less in India.
The proportion of MCI and T3DM (dementia) among PwD study participants was 38.6% and 24.3%, respectively, whereas the proportion of MCI and T3DM among people without T2DM was 25.7% and 10%, respectively, in our study.
In the study done by Dasgupta et al. in West Bengal in 2020, the prevalence of MCI was 36.4% using AQ[22] which is quite similar to our study. The proportion of CI among older adults with T2DM was 30.0% in a community-based study in Puducherry.[11] The prevalence of MCI was 26.06% and dementia was 5.63% in the study done by Mohan et al.[12] It is evident here, that we found a higher prevalence of MCI and Dementia among our study participants. Variation in the results between our study and some of the other studies may be due to clinic-based nature of our study, different study area, and time gap between the studies.
Longitudinal Ageing Study in India wave-1 survey (2017–2018) revealed that self-reported prevalence of diagnosed neurological or psychiatric problems (depression, AD, dementia or psychiatric problem or neurological problem [any one or more]) in ≥60 years people in India was 2.6%, among which the prevalence of AD and dementia was 1.0%.[23] The prevalence of diagnosed neurological or psychiatric problems in the geriatric population in West Bengal was substantially high (8.1%), among which the prevalence of AD and dementia was 5.4%.[24] In our study, the proportion of dementia among PwD was 24.3%, while among people without T2DM, it was 10%.
Durazzo et al. in their study found that former or active smoking is related to increased risk of developing AD and dementia.[25] Similarly, in our study, tobacco usage was found to be a predictor of dementia and MCI among diabetic study population and for MCI among people without T2DM.
In a study done by Khanna and Metgud, advancing age, female sex, unmarried or widow/widower, illiterate, not working presently, staying alone, and poverty were found to be significant risk factors of CI in the elderly.[26] In our study, we found advanced age, female gender, poor educational attainment (only till primary level), BPL family, tobacco usage, and family history of dementia as significant risk factors of dementia among elderly PwD.
Strengths
Novelty of the research question along with adequate sample size is a major strength of the study. The use of a validated tool (AQ) with high sensitivity and specificity, high diagnostic accuracy, and high internal consistency makes it stronger.
Limitations
Certain responses by the study participants, especially for addiction and family history of dementia, may have introduced social desirability bias. Since it is a clinic-based study, results may not be generalizable to community or other hospital settings and there may be some Berksonian bias. Due to the cross-sectional study design, there was no scope of follow-up. The cognitive status of the study participants was assessed using AQ which is a screening tool and does not provide confirmatory diagnosis of these disorders.
CONCLUSION
The current study showed that the proportion of MCI and dementia (T3DM) among PwD was higher than people without T2DM. Advanced age, female gender, poor educational attainment (up to primary education), poverty, tobacco use, and family history of dementia were significant risk factors of dementia among PwD.
Family history of dementia turned out to be the most influential risk factor for developing MCI and dementia among our study participants across all groups, i.e., among PwD and people without T2DM. Female gender, poor educational attainment (up to primary education), and family history of dementia were found to be significant predictors of dementia (T3DM) among PwD.
Regular screening for T2DM should be organized at the community level for early diagnosis and its evidence-based management. Among PwD, screening for CI should be organized for early detection and management of MCI and dementia (T3DM). Adequate glycemic control among PwD may decrease the incidence of CI and dementia (T3DM).
Multicentric and community-based longitudinal study may be planned for better understanding of the relationship of T2DM status and dementia (T3DM).
Conflicts of interest
There are no conflicts of interest.
Acknowledgments
We would like to acknowledge all the study subjects for their participation in this study. We are deeply thankful to HOD, Department of Community Medicine; IRC and IEC members of IQ City Medical College and Hospital, Durgapur, West Bengal, for giving necessary approval to conduct the study. We are thankful to Diabetes Awareness and You (DAY), an NGO working in the field of diabetes mellitus, for providing logistic support.
Funding Statement
This study received grant of INR 50,000/- under Indian Council of Medical Research (ICMR)-funded Short-Term Studentship (STS) Program for medical undergraduate students. Final report has already been submitted and accepted by ICMR (Reference ID: 2023-08143).
REFERENCES
- 1.International Institute for Population Sciences. United Nations Population Fund. New Delhi: United Nations Population Fund; 2023. [[Last accessed on 2025 Apr 10]]. India Ageing Report 2023: Caring for Our Elders –Institutional Responses. Available from: https://india.unfpa.org/en/publications/caring-our-elders-institutional-responses-india-ageing-report-2023 . [Google Scholar]
- 2.World Health Organization. Diabetes. [[Last accessed on 2025 Apr 10]]. Available from: https://www.who.int/health-topics/diabetes#tab=tab_1 .
- 3.International Diabetes Federation. 10th ed. Brussels: International Diabetes Federation; 2021. [[Last accessed on 2025 Apr 01]]. IDF Diabetes Atlas. Available from: https://www.idf.org/e-library/epidemiology-research/diabetes-atlas . [Google Scholar]
- 4.Jana A, Chattopadhyay A. Prevalence and potential determinants of chronic disease among elderly in India: Rural-urban perspectives. PLoS One. 2022;17:e0264937. doi: 10.1371/journal.pone.0264937. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.World Health Organization. Geneva: World Health Organization; 2017. [[Last accessed on 2025 Apr 05]]. Global Action Plan on the Public Health Response to Dementia 2017-2025. Available from: https://www.who.int/publications/i/item/global-action-plan-on-the-public-health-response-to-dementia-2017-2025 . [Google Scholar]
- 6.Kroner Z. The relationship between Alzheimer's disease and diabetes: Type 3 diabetes? Altern Med Rev. 2009;14:373–9. [PubMed] [Google Scholar]
- 7.Nguyen TT, Ta QT, Nguyen TK, Nguyen TT, Giau VV. Type 3 diabetes and its role implications in Alzheimer's disease. Int J Mol Sci. 2020;21:3165. doi: 10.3390/ijms21093165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Caberlotto L, Nguyen TP, Lauria M, Priami C, Rimondini R, Maioli S, et al. Cross-disease analysis of Alzheimer's disease and type-2 diabetes highlights the role of autophagy in the pathophysiology of two highly comorbid diseases. Sci Rep. 2019;9:3965. doi: 10.1038/s41598-019-39828-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Khullar S, Kaur G, Dhillon H, Sharma R, Mehta K, Singh M, et al. The prevalence and predictors of cognitive impairment in type 2 diabetic population of Punjab, India. J Soc Health Diabetes. 2017;5:47–53. [Google Scholar]
- 10.Alzheimer's Disease International. Dementia Statistics. [[Last accessed on 2025 Apr 08]]. Available from: https://www.alzint.org/about/dementia-facts-figures/dementia-statistics .
- 11.Subramanian M, Vasudevan K, Rajagopal A. Cognitive impairment among older adults with diabetes mellitus in Puducherry: A community-based cross-sectional study. Cureus. 2021;13:e12488. doi: 10.7759/cureus.12488. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mohan D, Iype T, Varghese S, Usha A, Mohan M. A cross-sectional study to assess prevalence and factors associated with mild cognitive impairment among older adults in an urban area of Kerala, South India. BMJ Open. 2019;9:e025473. doi: 10.1136/bmjopen-2018-025473. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Budolfson K, Malek-Ahmadi M, Belden CM, Powell J, Davis K, Jacobson S, et al. Neuropsychological correlates of the Alzheimer's questionnaire. J Alzheimers Dis. 2015;46:389–97. doi: 10.3233/JAD-142388. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Malek-Ahmadi M, Davis K, Belden C, Laizure B, Jacobson S, Yaari R, et al. Validation and diagnostic accuracy of the Alzheimer's questionnaire. Age Ageing. 2012;41:396–9. doi: 10.1093/ageing/afs008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.National Cancer Institute. NCI Dictionary of Cancer Terms: Cognitive Impairment. [[Last accessed on 2025 Apr 08]]. Available from: https://www.cancer.gov/publications/dictionaries/cancer-terms/def/cognitive-impairment .
- 16.National Institute on Aging. What is Mild Cognitive Impairment? [[Last accessed on 2025 Apr 08]]. Available from: https://www.alzheimers.gov/alzheimers-dementias/mild-cognitive-impairment .
- 17.Alzheimer's Disease International. Dementia Facts and Figures. [[Last accessed on 2025 Apr 08]]. Available from: https://www.alzint.org/about/dementia-facts-figures/
- 18.Alzheimer's Disease International. Types of Dementia. [[Last accessed on 2025 Apr 08]]. Available from: https://www.alzint.org/about/dementia-facts-figures/types-of-dementia/
- 19.Alzheimer's Society. Rarer Types of Dementia. [[Last accessed on 2025 Apr 08]]. Available from: https://www.alzheimers.org.uk/about-dementia/types-dementia/rarer-types-dementia#content-start .
- 20.The jamovi project. jamovi (Version 2.6) [Computer software] 2025. [[Last accessed on 2025 Apr 08]]. Available from: https://www.jamovi.org .
- 21.R Core Team. R: A language and environment for statistical computing (Version 4.4) [Computer software] 2024. [[Last accessed on 2025 Apr 08]]. Available from: https://cran.r-project.org . R packages retrieved from CRAN snapshot 2024-08-07.
- 22.Dasgupta A, Biswas A, Paul B, Roy S, Sarkar D, Ghosh P, et al. Comparative analysis of Alzheimer questionnaire and montreal cognitive assessment tool for cognitive impairment screening among the elderly population. J Clin Diagn Res. 2020;14 LC05-9. [Google Scholar]
- 23.International Institute for Population Sciences. National Programme for Health Care of Elderly. Mumbai: International Institute for Population Sciences; 2021. [[Last accessed on 2025 Apr 08]]. Longitudinal Ageing Study in India (LASI), Wave 1, 2017-18: Factsheet –Elderly in India. Available from: https://www.iipsindia.ac.in/content/lasi-publications . [Google Scholar]
- 24.International Institute for Population Sciences. National Programme for Health Care of Elderly. Mumbai: International Institute for Population Sciences; 2021. [[Last accessed on 2025 Apr 08]]. Longitudinal Ageing Study in India (LASI), Wave 1, 2017-18: Compendium of Factsheets –States/UTs. Available from: https://www.iipsindia.ac.in/content/lasi-publications . [Google Scholar]
- 25.Durazzo TC, Mattsson N, Weiner MW. Alzheimer's Disease Neuroimaging Initiative. Smoking and increased Alzheimer's disease risk: A review of potential mechanisms. Alzheimers Dement. 2014;10:S122–45. doi: 10.1016/j.jalz.2014.04.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Khanna AB, Metgud CS. Prevalence of cognitive impairment in elderly population residing in an urban area of Belagavi. J Family Med Prim Care. 2020;9:2699–703. doi: 10.4103/jfmpc.jfmpc_240_20. [DOI] [PMC free article] [PubMed] [Google Scholar]