The study included a total of 218 female participants with a mean age of 70.11+ 8.42 years. 164 (75.2%) participants had some form of formal education. A majority of the participants, 188 (86.2%) were Hindu, with only 13 (6%) Muslims and 17 (7.8%) Christians. 144 (67%) of participants were widowed, divorced or single. 178(81.7%) of the participants had one or more comorbid condition. 126 (57.8%) of the participants were non consumers of tobacco and 214 (98.2%) were non consumers of alcohol.
A. Becks Depression Inventory
Using Beck’s Depression Inventory, 31.7% (69) participants scored 17 or higher indicating borderline, moderate or severe depression. Of these participants, 57 (82.6%) reported that they feel sad all the time and cannot snap out of it, 43 (62.3%) felt as though they feel they have nothing to look forward to in the future and 53(76.8%) look back on their lives to see a lot of failures. A majority of the participants 61 (88.4%) did not feel particularly guilty nor were they disappointed in themselves, 42 (60.9%), however a majority reported that did feel as though they were being punished 44 (63.8%). Though a majority of the participants determined to have some form of depression had more difficulty working 61(88.3%) and sleeping 49 (71%) as well as they could before, most had no problems with weight loss 61 (88.4%), decision-making 53 (76.8%) or a belief that they look more unattractive than they used to 53 (76.8%).
B. Associations between socio-demographic characteristics and depression
Cross tabulation was performed to check for any associations between socio-demographic parameters and the presence of depression. After an initial univariate analysis was performed, those variable shown to have a p-value of < 0.25 on the Wald Chi Square Test (as per convention) were used to create a binary logistic regression model. The regression model that was made with these variables was statistically significant c2(17) = 127.09, p < .001. The model was able to correctly classify participants as depressed or not depressed 92.5% of the time and accounted for 72.9% of the variance (Nagelkerke R2). When the Hosmer and Lemeshow Test was performed, c2 (8) = 1.73, p = .99 it showed that the predictions made by the model fit well with the observed results of the participants. Out of the 17 variables used to create this model, 7 were seen to be significant. The five significant variables were whether the participant was part of a social group, whether they got a good night’s sleep on most days, whether they had a good relationship with their family, whether they had any concerns about loans that they needed to settle as well as concerns about their health, whether they preferred staying alone or in the company of others, and their educational status.
Table 1: Sociodemographic and General Determinants of Depression amongst Geriatric Women
Parameter
|
Total
n=218
(%)
|
Participants with depression
n=69
(%)
|
Participants without depression
n=149
(%)
|
OR (95% CI)
|
p-value
(univariate)
|
AOR (95% CI)
|
p-value
(multivariate)
|
Literacy
|
Literate
|
163(74.8)
|
35(16.06)
|
128 (58.72)
|
1.0
|
-
|
1.0
|
-
|
Illiterate
|
55(25.2)
|
34(15.60)
|
21(9.63)
|
4.91 (2.56-9.41)
|
<0.001
|
0.46 (.08-2.67)
|
0.39
|
Marital Status
|
Married
|
72(33.0)
|
21(9.6)
|
51(23.4)
|
1.0
|
-
|
|
|
Widowed/
Divorced/
Single
|
146(67.0)
|
48(22.0)
|
98(44.9)
|
1.15 (0.62-2.13)
|
0.65
|
|
|
Education
|
No formal education/primary school
|
138(63.3)
|
57(26.1)
|
81(37.2)
|
3.81(1.89-7.69)
|
<0.001
|
5.95(1.04-34.10)
|
0.045
|
Secondary school/PUC/graduate
|
80(36.7)
|
12(5.5)
|
68(31.2)
|
1.0
|
-
|
1.0
|
-
|
Pension
|
Yes
|
125(57.3)
|
42(19.3)
|
83(38.0)
|
1.0
|
-
|
|
|
No
|
93.00(42.7)
|
27(12.4)
|
66(30.3)
|
0.77(0.43-1.38)
|
0.374
|
|
|
Relationship with Family
|
Good
|
194(89.0)
|
53(24.3)
|
141(64.7)
|
1.0
|
-
|
1.0
|
-
|
Not so good
|
24(11.0)
|
16(7.3)
|
8(3.7)
|
3.63 (1.52-8.67)
|
0.004
|
0.05 (0.03-0.76)
|
0.031
|
Dependent on others for daily activities?
|
Yes
|
56(25.7)
|
21(9.6)
|
35(16.1)
|
1.63(0.86-3.08)
|
0.13
|
0.68(0.01-5.22)
|
0.71
|
No
|
162(74.3)
|
48(22.0)
|
114(52.3)
|
1.0
|
-
|
1.0
|
-
|
Any family member dependent on participant for daily activities?
|
Yes
|
42(19.3)
|
12(5.5)
|
30(13.8)
|
0.99(0.48-2.04)
|
0.97
|
|
|
No
|
176(80.7)
|
57(26.1)
|
119(54.6)
|
1.0
|
-
|
|
|
Is the participant a part of any social group?
|
Yes
|
58(26.6)
|
12(5.5)
|
46(21.2)
|
1.0
|
-
|
1.0
|
-
|
No
|
160(73.4)
|
57(26.2)
|
103(47.3)
|
1.81(0.90-3.64)
|
0.09
|
68.78(2.23-2120.65)
|
0.02
|
Does the participant have any comorbid conditions
|
Yes
|
178(81.7)
|
63(28.9)
|
115(52.8)
|
>20
|
0.99
|
|
|
No
|
40(18.3)
|
6(2.8)
|
34(15.6)
|
1.0
|
-
|
|
|
Has the participant been hospitalized in the last 6 months
|
Yes
|
30(13.8)
|
17(7.8)
|
13(6.0)
|
3.51 (1.59-7.74)
|
0.002
|
2.33 (0.36-14.91)
|
0.37
|
No
|
188(86.2)
|
52(23.9)
|
136(62.4)
|
1.0
|
-
|
1.0
|
-
|
Tobacco consumption
|
Ex-consumer/Non-consumer
|
154(70.6)
|
39(17.9)
|
115(52.8)
|
1.0
|
-
|
1.0
|
-
|
Current consumer
|
64(29.4)
|
29(13.3)
|
35(16.1)
|
2.44 (1.32-4.50)
|
0.004
|
2.44 (0.58-10.16)
|
0.222
|
Alcohol consumption
|
Ex-consumer/Non-consumer
|
214(98.2)
|
65(29.8)
|
149(68.3)
|
1.0
|
-
|
1.0
|
-
|
Current consumer
|
4(1.8)
|
3(1.4)
|
1(0.5)
|
6.88 (0.7-67.36)
|
0.10
|
22.44 (0.70-725.62)
|
0.08
|
Does the participant get a good night’s sleep on most days?
|
Yes
|
117(53.7)
|
14(6.4)
|
103(47.3)
|
1.0
|
-
|
1.0
|
-
|
No
|
101(46.3)
|
55(25.2)
|
46(21.2)
|
7.51 (3.85-14.64)
|
<0.001
|
94.82 (13.97-643.72)
|
<0.001
|
Does the participant feel that they can freely converse with their family?
|
Yes
|
16(7.3)
|
12(5.5)
|
142(65.1)
|
1.0
|
-
|
1.0
|
-
|
No
|
202(92.7)
|
57(26.2)
|
4(1.8)
|
6.21 (2.36-16.33)
|
<0.001
|
9.93 (0.86-114.60)
|
0.07
|
Does the participant prefer staying at home alone or in the company of others?
|
Yes
|
19(8.7)
|
11(5.1)
|
145(66.5)
|
1.0
|
-
|
10.97 (1.43-84.13)
|
0.02
|
No
|
199(91.3)
|
58(26.6)
|
8(3.7)
|
1.40 (0.77-2.56)
|
0.27
|
1.0
|
-
|
Does the participant engage in hobbies/social activities
|
Yes
|
156(87.6)
|
42(23.6)
|
141(64.7)
|
1.0
|
-
|
1.0
|
-
|
No
|
22(12.4)
|
15(8.4)
|
114(64.0)
|
2.26 (1.25-4.07)
|
0.07
|
2.60 (0.56-12.02)
|
0.22
|
Table 1: Describes general characteristics of the population with both univariate and multivariate analysis using the Chi square test.. A relatively small sample size may have been the reason for the wide range in the odds ratio.
Table 2: Personal Stressors and Social Isolation Determining Depression in Geriatric Women
Parameter
|
Total
n=218
(%)
|
Participants with depression
n=69
(%)
|
Participants without depression
n=169
(%)
|
OR (95% CI)
|
p-value
(univariate)
|
AOR (95% CI)
|
p-value
(multivariate)
|
Does the participant have concerns regarding your children’s marriage?
|
Yes
|
19(8.7)
|
9(4.1)
|
10 (4.6)
|
1.23 (0.65-2.32)
|
0.53
|
|
|
No
|
160(73.4)
|
48(22.0)
|
10(4.6)
|
1.0
|
-
|
|
|
Does the participant have concerns regarding your children’s education?
|
Yes
|
5(2.3)
|
3(1.4)
|
112(51.4)
|
3.41 (0.56-20.9)
|
0.18
|
.68 (0.02-20.42)
|
0.82
|
No
|
213(97.7)
|
66(31.6)
|
2(0.9)
|
1.0
|
-
|
1.0
|
-
|
Does the participant have any concerns regarding loans to settle?
|
Yes
|
66(30.3)
|
30(13.8)
|
147(67.4)
|
2.06(1.12-3.79)
|
0.02
|
18.55(2.65-129.64)
|
0.03
|
No
|
152(69.7)
|
39.00(17.9)
|
36(16.5)
|
1.0
|
-
|
1.0
|
-
|
Does the participant have concerns regarding their health?
|
Yes
|
64(29.3)
|
38(17.4)
|
113(51.8)
|
6.04 (3.20-11.44)
|
<0.001
|
5.09 (1.031-25.10)
|
0.046
|
No
|
154(70.6)
|
31(14.2)
|
26(11.9)
|
1.0
|
-
|
1.0
|
-
|
Does the participant have concerns regarding the financial burden of healthcare?
|
Yes
|
56(25.7)
|
36(16.5)
|
123(56.4)
|
5.82 (3.02-11.21)
|
<0.001
|
2.51 (0.46-13.77)
|
0.291
|
No
|
162(74.3)
|
33(15.1)
|
20(9.2)
|
1.0
|
-
|
1.0
|
-
|
Does the participant have concerns regarding the employment of self/family members?
|
Yes
|
29(13.3)
|
22(10.1)
|
129(59.2)
|
6.53 (2.78-15.31)
|
<0.001
|
.96 (.16-5.60)
|
0.96
|
No
|
189(86.7)
|
47(21.6)
|
7(3.2)
|
1.0
|
-
|
1.0
|
-
|
Does the participant feel that they can freely converse with their family?
|
Yes
|
16(7.3)
|
12(5.5)
|
142(65.1)
|
1.0
|
-
|
1.0
|
-
|
No
|
202(92.7)
|
57(26.2)
|
4(1.8)
|
6.21 (2.36-16.33)
|
<0.001
|
9.93 (0.86-114.60)
|
0.07
|
Does the participant prefer staying at home alone or in the company of others?
|
Yes
|
19(8.7)
|
11(5.1)
|
145(66.5)
|
1.0
|
-
|
10.97 (1.43-84.13)
|
0.02
|
No
|
199(91.3)
|
58(26.6)
|
8(3.7)
|
1.40 (0.77-2.56)
|
.27
|
1.0
|
-
|
Does the participant engage in hobbies/social activities
|
Yes
|
156(87.6)
|
42(23.6)
|
141(64.7)
|
1.0
|
-
|
1.0
|
-
|
No
|
22(12.4)
|
15(8.4)
|
114(64.0)
|
2.26 (1.25-4.07)
|
0.07
|
2.60 (0.56-12.02)
|
0.22
|
Table 2: Describes the personal stressors and need for social isolation of the population with univariate and multivariate analysis using the Chi square test. A relatively small sample size may have been the reason for the wide range in the odds ratio.
B.1 Education on depression
The bivariate analysis showed that those participants with no formal education or only a primary school level of education were more likely to be depressed than those who had educational qualifications of secondary school or higher with an AOR of 5.95(1.04-34.10) and was a significant association c2 (1) = 4.01, p = .04.
B.2 Social interactions on depression
Participants who preferred staying at home alone were more likely to be depressed than those who preferred to be in the company of others with an AOR= 10.97 (1.43-84.13) and a statistically significant association c2 (1) = 5.31, p = .02. Participants who felt that they did not have a good relationship with their families had a higher likelihood of being depressed than those who did, with an AOR of .05 (.03-.76) and a statistically significant association c2 (1) = 4.67, p = .03.
B.3 Physical health and daily worries on depression
Presence of a comorbid condition did not seem to have a significant association with depression. Those with daily concerns regarding their health had a higher chance of being depressed with an AOR=5.09 (1.031-25.10) and was statistically significant with c2 (1) = 3.99, p = .05), as well as those participants who had concerns regarding loans to settle who also had a higher chance of being depressed with an AOR of 18.55(2.65-129.64) and was also significant at c2 (1) =8.66, p = .003.
B.4 Sleep quality on depression
Participants who complained of not getting a good night’s sleep had a higher chance of being depressed with an AOR=94.82 (13.97-643.72) and was statistically significant at c2 (1) =8.66, p < .001.
C. Pre and Post intervention
Of the 218 participants in this study, 69 were considered to have borderline, moderate or severe depression and were administered a psycho-education. A questionnaire with 8 questions (each worth 1 point) regarding knowledge and attitudes towards depression was also administered before and after the psycho-education. The median score before administration of the psycho education was 4 while the median score after administration of the psychoeducation was 8. It was seen that 50 of the 69 participants (73.53%) showed improvement in the score after the psycho-education. Of the 19 participants that showed no improvement, the median score in the pre-psychoeducation questionnaire was 8 (16 out of the 19 participants scored a perfect score) indicating that these participants already had good knowledge and attitudes regarding depression and had little room for improvement. Those that showed improvement after the psycho-education had a median score of 3.5 in the pre-psychoeducation questionnaire which increased to 6.96 post psycho-education. Additionally, after administration of the intervention, all 69 participants were willing to meet with a physician to further discuss their mental health issues.
Question No.
|
% with correct answers before Intervention
|
% with correct answers after intervention
|
1
|
52.2
|
82.6
|
2
|
39.1
|
88.4
|
3
|
47.8
|
88.4
|
4
|
55.1
|
84.1
|
5
|
75.4
|
92.8
|
6
|
82.6
|
89.9
|
7
|
49.3
|
82.6
|
8
|
55.1
|
91.2
|
When the one-tailed Wilcoxon signed ranks test was performed, it showed that participants had a tendency to have higher test scores after the administration of the questionnaire rather than before. (p=<0.001).
D.Associations between socio-demographic characteristics and Post-intervention improvement
Significant associations between socio-demographic characteristics and post-intervention improvement included hobbies/ social activities, relationship with the family and concerns about health. Bivariate analysis showed those who engaged in hobbies/social activities to have a higher odds of no improvement than those who did not (OR=3.214(1.053-9.812)) with a significant association of c2 (1) = 4.401, p = 0.036. Those who had good relationships with their family had a higher chance of showing no improvement than those who did not with an OR= 7.286 (.887-59.827) and a statistical significance of c2 (1) = 4.395, p = 0.036. Those who had concerns about their health had a higher chance of no improvement post intervention than those who did not with an OR=3.792 (1.095-13.129) and a statistical significance of x2 (1) = 4.760, p = 0.029.