**1. Introduction**

One of the most common neurological conditions that affect older adults is dementia1 . In addition, public health reports in the United States show that one of the most frequent physical illnesses found in older adults is diabetes2 . Dementia affects approximately 6–10% of people 65 years or older [1], and this prevalence rate increases with age. Neuropsychological impair‐ ment associated with dementia includes poor judgment, difficulty with calculation, and getting lost while driving [2]. Older patients with both dementia and depression typically show impairment in the domains of attention, memory, and psychomotor speed [3]. On the other hand, diabetes affects about 20% of people older than 65 years [4]. Patients treated for type II diabetes show cognitive deficits on brief cognitive screening [5], and subtle decrements in verbal memory and processing speed (mean difference in z scores -0.37 and -0.25 respectively) on in‐depth cognitive testing. This suggests that diabetes is a significant risk factor for dementia [6]. Depression has been estimated to affect 1–5% of community‐dwelling older adults [7]. Depression is common in dementia, with substantial variability in the reported base‐rate ranging from 20 to 40% [4, 8, 9], and it is a probable risk factor for both vascular dementia (VaD) and Alzheimer's disease (AD) [10]. As for diabetes, the prevalence of depression is three times higher in these patients than in individuals free of diabetes (OR = 2.9, 95% CI 2.3–3.7) [10]. Diabetes and depression exhibit a closely linked bi‐directional relationship: between 15 and 20% of people with diabetes will develop clinical depression, while depression is two to three times more common among people with diabetes compared to those without. Comorbid depression and diabetes have a 2.7‐fold increased risk for dementia [11].

Taking these factors into account, our chief interest was to compare cognitive function in individuals diagnosed with dementia with comorbid diabetes and depressive symptoms, alone or combined, compared to those with dementia alone, and against healthy controls.

### **2. Methods**

#### **2.1. Subject**

All subjects from that study come from the Study on Aging and Dementia in Mexico (SADEM). SADEM was a cross‐sectional study conducted to determine the prevalence of MCI and dementia, between September 2009 and March 2010. Individuals age 60 years and older were invited into the study through a random sampling of the eligible population registered within 24 family medicine units from the Instituto Mexicano del Seguro Social (IMSS) encompassing all of Mexico City. All subjects were beneficiaries (users and non‐users) of the IMSS. The inclusion criteria for SADEM were 1) community‐dwelling individuals aged 60 years and older, living in Mexico City, 2) registered with IMSS; and 3) accepting to take part in the study through informed consent. We excluded individuals 1) resident in other states; 2) living in an institution; 3) altered mental status secondary to delirium; 4) died before study start; 5) currently taking antipsychotic medication (other psychotropic medications including antide‐ pressants were allowed because of the potential negative impact of non‐treatment on cogni‐ tion); and 6) those who refused to participate or who after two attempted visits could not be located. All participants were assessed at doctor's office by a geriatric specialized in geriatric cognitive disorders. The diagnosis of dementia was performed during the study SADEM and was based on the DSM‐IV. All subjects were clinically assessed. Final diagnoses were assigned by a consensus expert panel made up of neuropsychologists, neurologists, and geriatricians [12]. All subjects participating in the study, or their caregivers, gave signed informed consent. The research protocol was reviewed and approved by The National Commission of Scientific Research as well as by the IMSS Ethics Commission (registration number 2010‐785‐005).

In this way for 3D study, we included all patients diagnosed with dementia vascular, Alzheimer disease, or both, diagnosed during the study SADEM, both sexes. Subjects were excluded if they had a) problems with vision, b) poor auditory capacity, c) history of alcohol abuse, d) Parkinsonism or meningioma, and e) incomplete assessment scales or neurological examination. Finally, 330 patients were included and stratified into following groups: a) patients with only dementia: Alzheimer (AD), vascular (VaD) and mixed (MD) (1D†), b) patients with only depression (1D†), c) patients with only diabetes mellitus (1D†), d) patients with AD and diabetes mellitus (2D†), e) patients with AD and depression (2D†), f) patients with VaD and diabetes mellitus (2D†), g) patients with VaD and depression (2D†), h) patients with MD and diabetes mellitus (2D†), i) patients with MD and depression (2D†), j) patients with AD and diabetes mellitus and depression (3D†), k) patients with VaD and diabetes mellitus and depression (3D†), and l) patients with MD and diabetes mellitus and depression (3D†).

#### **2.2. Dementia evaluation**

**1. Introduction**

516 Update on Dementia

**2. Methods**

**2.1. Subject**

illnesses found in older adults is diabetes2

One of the most common neurological conditions that affect older adults is dementia1

addition, public health reports in the United States show that one of the most frequent physical

65 years or older [1], and this prevalence rate increases with age. Neuropsychological impair‐ ment associated with dementia includes poor judgment, difficulty with calculation, and getting lost while driving [2]. Older patients with both dementia and depression typically show impairment in the domains of attention, memory, and psychomotor speed [3]. On the other hand, diabetes affects about 20% of people older than 65 years [4]. Patients treated for type II diabetes show cognitive deficits on brief cognitive screening [5], and subtle decrements in verbal memory and processing speed (mean difference in z scores -0.37 and -0.25 respectively) on in‐depth cognitive testing. This suggests that diabetes is a significant risk factor for dementia [6]. Depression has been estimated to affect 1–5% of community‐dwelling older adults [7]. Depression is common in dementia, with substantial variability in the reported base‐rate ranging from 20 to 40% [4, 8, 9], and it is a probable risk factor for both vascular dementia (VaD) and Alzheimer's disease (AD) [10]. As for diabetes, the prevalence of depression is three times higher in these patients than in individuals free of diabetes (OR = 2.9, 95% CI 2.3–3.7) [10]. Diabetes and depression exhibit a closely linked bi‐directional relationship: between 15 and 20% of people with diabetes will develop clinical depression, while depression is two to three times more common among people with diabetes compared to those without. Comorbid

Taking these factors into account, our chief interest was to compare cognitive function in individuals diagnosed with dementia with comorbid diabetes and depressive symptoms, alone or combined, compared to those with dementia alone, and against healthy controls.

All subjects from that study come from the Study on Aging and Dementia in Mexico (SADEM). SADEM was a cross‐sectional study conducted to determine the prevalence of MCI and dementia, between September 2009 and March 2010. Individuals age 60 years and older were invited into the study through a random sampling of the eligible population registered within 24 family medicine units from the Instituto Mexicano del Seguro Social (IMSS) encompassing all of Mexico City. All subjects were beneficiaries (users and non‐users) of the IMSS. The inclusion criteria for SADEM were 1) community‐dwelling individuals aged 60 years and older, living in Mexico City, 2) registered with IMSS; and 3) accepting to take part in the study through informed consent. We excluded individuals 1) resident in other states; 2) living in an institution; 3) altered mental status secondary to delirium; 4) died before study start; 5) currently taking antipsychotic medication (other psychotropic medications including antide‐ pressants were allowed because of the potential negative impact of non‐treatment on cogni‐

depression and diabetes have a 2.7‐fold increased risk for dementia [11].

. Dementia affects approximately 6–10% of people

. In

Dementia case newly recognized was conducted in two steps. First all participants of SADEM study were screened with the Mini‐Mental State Examination (MMSE) [13]. All participants with a cutoff ≤24 adjustment for educational level in an aging Mexican population were underwent a battery of neuropsychological measures and a standardized neurological examination. We used the Clinical Dementia Rating or CDR to quantify the severity of symptoms of dementia [14]. Complete details of the evaluation and diagnostic procedures have been described earlier [12]. The final diagnosis of dementia was determined by consensus expert panel review including neuropsychologists, neurologists, and geriatric physicians. Each diagnosis was based on based on the Diagnostic and Statistical Manual of Mental Disorders criteria for dementia [15], (DMS‐IV‐R) criteria for dementia. Once dementia was diagnosed, subjects were further grouped according to whether they met the National Institute for Neurological and Communicative Disorders and Stroke and the Alzheimer's Disease and Related Disorders Association (NINCDS‐ADRDA) [16], and/or the National Institute of Neurological Disorders and Stroke Association Internationale Pour la Recherche et l'Enseigne‐ ment en Neurosciences (NINDS‐AIREN) criteria [17]. Diagnoses fell into three categories: a) probable Alzheimer's disease (AD), b) vascular dementia (VaD), and c) mixed type dementia (MD) and were ascertained using a two‐step procedure: (1) diagnosis of dementia and (2) association of cognitive impairment to lesions of vascular of origin. The criteria for diagnosis of MD were that the course was suggestive of AD, and in addition, there were focal neurologic symptoms suggestive of ischemia. The presence of vascular risk factors alone, in a patient with otherwise clinically typical AD, was not enough to support a diagnosis of MD. Hence, patients suspected of MD were subjected to all procedures for diagnosis.

#### **2.3. Control subjects**

The control group consisted of 134 subjects healthy from the SADEM study, which did not meet MCI or dementia criteria, they had a CDR score of zero, and a memory test performance <1.5 standard deviations from the mean for age. There were no significant differences between controls and cases for age or years of education (p > 0.01).

#### **2.4. Cognitive measure**

Cognitive testing was performed on all the patients in whom MMSE [14] and was used to evaluate global cognitive performance. For the purpose of the present study, the 11 MMSE subtests and the global MMSE score were considered independently: *spatial orientation* (state, county, town, place, and floor=5 points), *temporal orientation* (year, season, month, day, and date=5 points), *immediate memory* (immediately repeating three words=3 points), *attention/ concentration* (If the participants had education as serially subtracting 7, beginning with 100, or, alternatively, spelling the word world backward for the participants without education=5 points), *delayed recall* (recalling the previously repeated three words=3 points), *language* (naming two items=2 points), *verbal repetition* (repeating a phrase=1 points), *reading a sentence* (reading aloud and understanding a sentence=1 points), *writing a sentence* (1 points), *verbal comprehension* (following a three‐step command=3 points), and *constructional praxis* (copying a design=1 points). [The MMSE is thoroughly familiar to any readership in geriatrics or neurol‐ ogy. You would only need to specify the above if you were intending to present subscores. That might be a good idea if you do not have access to the neuropsychological scores].

#### **2.5. Depression**

In addition to neuropsychological tests, depression symptoms were evaluated using the Center for Epidemiologic Studies Depression Scale (CES‐D), a validated 20‐item scale consisting of four factors: depressive affect, somatic complaints, positive affect, and interpersonal relations. Scores on the CES‐D ranged from 0 to 60, where 0–15 is indicative of absence of depression, and scores of 16–60 are indicative of depressive symptomatology [18]. The presence of depression was corroborated with the self‐reported treatment with antidepressants, selective serotonin reuptake inhibitors (SSRI—i.e., paroxetine, sertraline, fluoxetine, venlafaxine, citalopram), or tricyclic antidepressants at any time during the four‐month preceding the interview.

#### **2.6. Diabetes assessment**

Diagnosis of diabetes mellitus was based on the patients' self‐report to the question, "Are you taking medication for diabetes?" These diabetes diagnoses were then confirmed by blood glucose measurements, a fasting plasma glucose concentration >7.0 mmol/l (whole blood >6.1 mmol/l) and hemoglobin A1c (HbA1c) (<48 mmol/mol (6.5%), and treatment. All patients in this study were receiving treatment for diabetes, and 90% of patients were taking sulfony‐ lureas for controlling diabetes with the remaining 10% on insulin therapy.

#### **2.7. Statistical analyses**

symptoms suggestive of ischemia. The presence of vascular risk factors alone, in a patient with otherwise clinically typical AD, was not enough to support a diagnosis of MD. Hence, patients

The control group consisted of 134 subjects healthy from the SADEM study, which did not meet MCI or dementia criteria, they had a CDR score of zero, and a memory test performance <1.5 standard deviations from the mean for age. There were no significant differences between

Cognitive testing was performed on all the patients in whom MMSE [14] and was used to evaluate global cognitive performance. For the purpose of the present study, the 11 MMSE subtests and the global MMSE score were considered independently: *spatial orientation* (state, county, town, place, and floor=5 points), *temporal orientation* (year, season, month, day, and date=5 points), *immediate memory* (immediately repeating three words=3 points), *attention/ concentration* (If the participants had education as serially subtracting 7, beginning with 100, or, alternatively, spelling the word world backward for the participants without education=5 points), *delayed recall* (recalling the previously repeated three words=3 points), *language* (naming two items=2 points), *verbal repetition* (repeating a phrase=1 points), *reading a sentence* (reading aloud and understanding a sentence=1 points), *writing a sentence* (1 points), *verbal comprehension* (following a three‐step command=3 points), and *constructional praxis* (copying a design=1 points). [The MMSE is thoroughly familiar to any readership in geriatrics or neurol‐ ogy. You would only need to specify the above if you were intending to present subscores. That might be a good idea if you do not have access to the neuropsychological scores].

In addition to neuropsychological tests, depression symptoms were evaluated using the Center for Epidemiologic Studies Depression Scale (CES‐D), a validated 20‐item scale consisting of four factors: depressive affect, somatic complaints, positive affect, and interpersonal relations. Scores on the CES‐D ranged from 0 to 60, where 0–15 is indicative of absence of depression, and scores of 16–60 are indicative of depressive symptomatology [18]. The presence of depression was corroborated with the self‐reported treatment with antidepressants, selective serotonin reuptake inhibitors (SSRI—i.e., paroxetine, sertraline, fluoxetine, venlafaxine, citalopram), or tricyclic antidepressants at any time during the four‐month preceding the

Diagnosis of diabetes mellitus was based on the patients' self‐report to the question, "Are you taking medication for diabetes?" These diabetes diagnoses were then confirmed by blood glucose measurements, a fasting plasma glucose concentration >7.0 mmol/l (whole blood

suspected of MD were subjected to all procedures for diagnosis.

controls and cases for age or years of education (p > 0.01).

**2.3. Control subjects**

518 Update on Dementia

**2.4. Cognitive measure**

**2.5. Depression**

interview.

**2.6. Diabetes assessment**

We examined effects of dementia by type, depression, and diabetes by themselves. All comparisons were performed with MANOVAs that are being conducted using a Bonferroni correction of m [19], with dementia (AD, VaD, mixed), depression (yes/no), and diabetes mellitus (yes/no) as between‐subject factors. The same MANOVAs models were then used within each diagnostic group (AD, VaD, MD). Main effects and statistically significant difference between groups were assessed by F‐test [20]. Additionally, a complementary analysis to better interpret the results was carried‐out using the standardized mean effect sizes (Hedges' g). Hedges' g is calculated on the basis of the standardized mean difference effect size, which uses the pooled within‐groups SD but corrects for bias from small sample sizes. These effect sizes indicate the mean difference between two variables expressed in standard deviation units. Hedges' g is a conservative estimate of effect size, which typically is interpreted by Cohen's d guidelines (small effect=0.20, medium effect=0.50, large effect=0.80) [21]. A positive effect size indicates that the MMSE score in the control group was superior to the diagnostic groups, whereas a negative effect size indicates that the diagnostic groups outperformed the control.
