Estimates of depressive disorders in the elderly vary depending on how cases are defined. We estimated the prevalence of subthreshold depression (SD) and clinically significant depression (D) in a population of 70–74-year-olds. We also looked for associations with sociodemographic factors and perceptions of self. Participants underwent a multidimensional assessment (social, medical, and neuropsychological). The estimated prevalence of SD was 15.71% (95% CI: 13.70–17.72), while that of D was 5.58% (95% CI: 4.31–6.85). Multinomial logistic regression analysis revealed that female gender and dissatisfaction with family relationships were related to SD and D. A self-perception of physical age as older than actual age (but not comorbidity) and greater self-perceived stress caused by negative life events both increased the probability of SD. The likelihood of D was decreased in those who perceived their own health as good, whereas a self-perception of mental age as older than actual age and dissatisfaction with relationships with friends were both significantly associated with D. Both SD and D emerged as key problems in our population. Female gender and self-perceptions of various characteristics, which can be explored through simple questions, are associated with late-life depression in elderly people independently of their actual physical condition and other characteristics.
Depressive disorders in aging are associated with increased physical morbidity, lower functional status, and a higher risk of dementia, and they are a major economic and social burden on families and on society [
Many social and clinical conditions, such as older age, female gender, low education, being single or uncoupled, living alone, physical illness, and low cognitive functions, in particular executive ones, are known to be associated with both subthreshold and clinically significant depression [
One aspect of depression in the elderly not yet studied in depth is the role of certain perceptions of self (across a range of variables) that may conceivably be associated with mood. Indeed, even though self-perceived health has been found to reflect subjective health-related quality of life in the elderly and appears to show a strong relationship with several health and social characteristics in this population, although, not with depressive symptoms [
The primary aim of the present study was to estimate the prevalence rates of subthreshold depression and clinically significant depression in a section of the elderly Italian population (those aged 70–74 years) living in Abbiategrasso, a northern Italian town on the outskirts of Milan. The secondary aim was to evaluate social, medical, cognitive, and self-perception variables in this population and the influence of these factors on the presence of subthreshold depression and clinically significant depression.
The participants were drawn from the 1321 responders in the cross-sectional phase of InveCe.Ab (Invecchiamento Cerebrale in Abbiategrasso,
All the study procedures were in accordance with the Declaration of Helsinki and the study protocol was approved by the Ethics Committee of the University of Pavia on October 6, 2009 (Committee report 3/2009). In summary, InveCe.Ab is a single-step, multidimensional study, in which all the participants underwent the same assessment (social, medical, and neuropsychological), performed by trained geriatricians and neuropsychologists. All the participants gave their written informed consent to the use of their personal data.
The primary endpoint was depressive status classified as subthreshold depression (SD), clinically significant depression (D), or without depression (noD). The presence of depressed mood was evaluated as part of the medical and neuropsychological assessment. Following the guidelines of the National Institute of Clinical Excellence (NCCMH 2010), SD was defined as the presence of at least one key symptom of depression but with insufficient other symptoms to meet the criteria for a diagnosis of depression. In accordance with the DSM-IV-TR [
All individuals not meeting the above-mentioned criteria for SD or D were considered to be without depression (noD).
The sociodemographic characteristics considered in the present survey were age (70–75), gender (female and male), years of education (≤5 or >5), marital status (defined as coupled [currently married/cohabiting] or single [never married]/uncoupled [separated/divorced/widowed]), living situation (living with spouse and/or others versus living alone), and primary lifetime occupation (considering a series of categories. housewife, blue collar worker, and white collar worker, adapted from the nine classes established by Italian National Institute of Statistics) [
The participants’ general health was evaluated using the Cumulative Illness Rating Scale. This is a standardized tool for detecting multimorbidity, in which a five-level Likert scale is used to rate the severity of medical illness for each of 14 items referring to different domains [
As part of the neuropsychological assessment global cognition was evaluated using the Mini-Mental State Examination (MMSE), whose items investigate orientation, registration and recall, attention, language, following commands, and figure copying [
Perceptions of self were investigated using various self-perception indices. During the social assessment, the subjects were asked to rate their overall health as poor, fair, or good. Self-perception of age was assessed considering two subindices: self-perceived mental and self-perceived physical age. Self-perceived mental age was determined by asking the patient “Do you feel mentally younger, the same as, or older than your actual age?” while the question concerning self-perceived physical age was “Do you feel physically younger, the same as, or older than your actual age?” The answers to these questions were classified as younger than actual age, equal to actual age, or older than actual age [
Mean values with standard deviation (sd), or medians with interquartile range (iqr) if data lacked normality, were used to summarize the quantitative variables, while percentages were used to describe the categorical variables. To evaluate associations between depressive status (SD, D or noD) and categorical variables, the chi-square test or Fisher’s exact test was performed. Similarly, to evaluate relationships with quantitative factors, a parametric analysis of variance (ANOVA) or corresponding nonparametric test (Kruskal-Wallis rank test) was applied. If the associations with different depressive status were significant, then multiple comparison tests with Bonferroni’s correction were applied taking noD as the reference status. Normality distribution was assessed using the Shapiro-Wilk test. A multinomial logistic regression model was performed to investigate the association of depressive status with several social, medical, cognitive, and self-perception variables using noD as the reference category in the model. The results were reported as the relative risk ratios (RRRs) with 95% confidence intervals (95% CI). Multinomial logistic regression allowed simultaneous comparison of the three categories of depressive status while adjusting for all other variables in the model. The goodness-of-fit of the multinomial model was assessed by means of the Hosmer-Lemeshow test. A
The prevalence of SD among the 1254 elderly subjects was 15.71% (95% CI: 13.70–17.72), while that of D was 5.58% (95% CI: 4.31–6.85).
Depressive status was differently associated with all the demographic characteristics except for years of education (
Sociodemographic characteristics in elderly subjects with subthreshold depression (SD), with clinically significant depression (D), and without depression (noD).
SD |
D |
noD |
|
|
---|---|---|---|---|
|
|
|
|
0.013 |
|
<0.0001 | |||
Male | 52 (26%) | 11 (19%) | 519 (52%) | |
Female | 145 (74%) | 59 (81%) | 468 (48%) | |
|
0.266 | |||
≤5 years | 122 (62%) | 42 (60%) | 545 (55%) | |
>5 years | 75 (38%) | 28 (40%) | 442 (45%) | |
|
<0.0001 | |||
Coupled | 106 (54%) | 36 (51%) | 706 (71%) | |
Single/uncoupled | 91 (46%) | 34 (49%) | 281 (29%) | |
|
<0.0001 | |||
Living with spouse and/or others | 124 (63%) | 41 (59%) | 761 (77%) | |
Living alone | 73 (37%) | 29 (41%) | 226 (23%) | |
|
<0.0001 | |||
Housewife | 34 (17%) | 25 (36%) | 124 (13%) | |
Blue collar worker | 95 (48%) | 28 (40%) | 504 (51%) | |
White collar worker | 68 (35%) | 17 (24%) | 359 (36%) |
Comorbidity, neuropsychological, and self-perception variables in elderly subjects with subthreshold depression (SD), with clinically significant depression (D), and without depression (noD).
SD |
D |
noD |
|
|
---|---|---|---|---|
|
2 (3–1) | 3 (4–2) | 2 (3–1) | 0.0001 |
|
28 (29–27) | 27 (29–26) | 29 (29–27) | 0.002 |
|
|
|
|
<0.0001 |
|
<0.0001 | |||
Poor | 3 (2%) | 5 (7%) | 6 (1%) | |
Fair | 101 (51%) | 46 (67%) | 265 (27%) | |
Good | 93 (47%) | 18 (26%) | 714 (72%) | |
|
<0.0001 | |||
Equal to actual age | 88 (45%) | 19 (28%) | 547 (56%) | |
Younger than actual age | 80 (41%) | 36 (53%) | 395 (40%) | |
Older than actual age | 29 (15%) | 13 (19%) | 37 (4%) | |
|
<0.0001 | |||
Equal to actual age | 60 (31%) | 12 (18%) | 482 (49%) | |
Younger than actual age | 90 (46%) | 33 (49%) | 444 (45%) | |
Older than actual age | 45 (23%) | 22 (33%) | 53 (6%) | |
|
<0.0001 | |||
Satisfied | 180 (92%) | 59 (83%) | 996 (98%) | |
Not satisfied | 15 (8%) | 12 (17%) | 18 (2%) | |
|
<0.0001 | |||
Satisfied | 164 (87%) | 42 (65%) | 885 (91%) | |
Not satisfied | 25 (13%) | 23 (35%) | 84 (9%) | |
|
|
|
|
<0.0001 |
|
5 (7–3) | 5 (8–3) | 3 (5–2) | 0.0001 |
|
8 (12–5) | 9 (13–5) | 6 (9–3) | 0.0001 |
According to the results of the multinomial logistic regression analysis (Table
Multinomial logistic regression for identifying factors associated with depressive status (reference group: noD).
SD versus noD | D versus noD | |
---|---|---|
RRR (95% CI) | RRR (95% CI) | |
|
1.08 (0.95–1.24) | 1.08 (0.87–1.35) |
|
||
Male | 1 | 1 |
Female |
|
|
|
||
≤5 years | 1 | 1 |
>5 years | 0.94 (0.59–1.49) | 2.12 (1.00–4.49) |
|
||
Coupled | 1 | 1 |
Single/uncoupled | 1.11 (0.83–1.50) | 0.97 (0.59–1.57) |
|
||
Living with spouse and/or others | 1 | 1 |
Living alone | 1.13 (0.92–1.39) | 1.17 (0.84–1.63) |
|
||
Blue collar worker | 1 | 1 |
Housewife | 0.82 (0.48–1.40) | 1.18 (0.54–2.55) |
White collar worker | 1.29 (0.80–2.08) | 0.71 (0.30–1.70) |
|
1.02 (0.90–1.16) | 1.02 (0.83–1.25) |
|
0.96 (0.88–1.05) | 0.92 (0.81–1.06) |
|
0.97 (0.87–1.08) | 0.89 (0.77–1.04) |
|
||
Poor | 1 | 1 |
Fair | 1.73 (0.28–10.4) | 0.40 (0.77–2.08) |
Good | 0.99 (0.16–6.08) |
|
|
||
Equal to actual age | 1 | 1 |
Younger than actual age | 0.81 (0.52–1.25) | 1.56 (0.72–3.38) |
Older than actual age | 1.93 (0.90–4.15) |
|
|
||
Equal to actual age | 1 | 1 |
Younger than actual age | 1.38 (0.87–2.17) | 1.44 (0.61–3.38) |
Older than actual age |
|
2.14 (0.70–6.52) |
|
||
Satisfied | 1 | 1 |
Not satisfied |
|
|
|
||
Satisfied | 1 | 1 |
Not satisfied | 1.04 (0.78–1.39) |
|
|
|
1.12 (0.88–1.43) |
|
0.97 (0.90–1.05) | 1.00 (0.87–1.15) |
Specifically, in females with respect to males, the relative risk of having SD versus noD was increased almost threefold, while the relative risk of having D as opposed to noD was increased fivefold.
In other words, females were more likely to belong to SD or D group versus noD than males. Similarly, subjects dissatisfied with their relationships with their family compared with those who declared themselves satisfied were more likely to have SD (one and a half times) or D (around three times) as opposed to noD status.
Conversely, variables significantly associated only with SD, independently of other factors, were self-perception of physical age (
Instead, the variables significantly associated with D, independently of other factors, were self-perception of health (
No other variable was associated with SD or D. The Hosmer-Lemeshow goodness-of-fit test indicated that the model described the data well (chi-square (8) = 11.36,
We conducted a comprehensive investigation of late-life depression in which additional features were considered together with clinician-rated DSM diagnoses of depression. The study showed prevalence rates of 15.71% for SD and 5.58% for D in our elderly population. This result confirms that SD is a problem not only in young people and adults but also in the elderly, in line with data reported in other, similar studies [
Furthermore, our data indicate that clinicians should carefully monitor elderly subjects not only for D but also for SD, given the high risk of this latter condition evolving into more severe depression [
In accordance with previously reported findings [
The multinomial regression analysis identified three groups of variables showing associations with depressive status: those associated with both SD and D, those associated only with SD, and those associated only with D, independently of comorbidity and other characteristics.
With regard to the first group, females were more likely to be classified as SD or D versus noD than males, confirming that the female gender is more prone to depressive symptoms [
With regard to the second group, having an older perceived than actual physical age, as opposed to a perception consistent with actual age, increased the probability of being classified as SD versus noD. This, together with the absence of relationships between comorbidity and SD or D in this sample, suggests that subjective but not objective physical difficulties are implicated in SD. It is well known that physical comorbidity is related to a higher level of depression [
As regards the third group of variables showing associations with depressive status, subjects who considered their health as good as opposed to poor were less likely to be classified as D versus noD, a result in line with the findings of others [
The results of our study show that SD and D are key problems in the elderly and that they are associated with mood independently of actual physical conditions and other characteristics; it also showed that they are subjective dimensions that can be easily explored through simple questions. Perceptions of self may offer the clinician an individualised profile of the patient with SD or D.
The strengths of the present study include the fact that it is a population study with a good response rate [
Through a comprehensive investigation of depressive conditions, we obtained estimates of the rates of SD and D in a sizeable sample of elderly subjects (70–74-year-olds). The results of our study confirmed that both SD and D are key problems in the elderly. They also confirmed that female gender and self-perceptions of various variables (health status, mental and physical age, satisfaction with relationships, and stress linked to negative life events) are associated with late-life depression, independently of physical comorbidity and other factors.
The authors declare that there are no conflicts of interest regarding the publication of this paper.
The authors are grateful to Federazione Alzheimer Italia, Milan, for supporting the “Golgi Cenci” Foundation’s research activities.