PSYCHOGERIATRICS 2016; 16: 93101

doi:10.1111/psyg.12126

ORIGINAL ARTICLE

The effect of benzodiazepine and nonbenzodiazepine

prescriptions for diabetes mellitus type 2 in elderly Taiwanese
with depressive symptoms
Min-Ling TSAI,1,2 Chien-Ning HUANG,1,3* Yung-Rung LAI,2 Horng-Rong CHANG1,4 and
Jeng-Yuan CHIOU5**

1
Institute of Medicine, 5School of Health Policy and
Management, Chung Shan Medical University,
Departments of 2Pharmacy and 3Internal Medicine,
4
Division of Nephrology, Department of Internal
Medicine, Chung Shan Medical University
Hospital, Taichung, Taiwan

Correspondence: Jeng-Yuan Chiou, PhD, School of

Health Policy and Management, Chung Shan
Medical University, 110, Section 1, Jianguo North
Road, South District, Taichung, Taiwan. Email:
tom@csmu.edu.tw
Received 28 October 2014; revision received 5 January
2015; accepted 22 February 2015.
*The author contributed equally to the rst author.
**Author to whom correspondence should be
addressed.

The author contributed equally to corresponding

author.

[Correction added on 22 September 2015, after rst

online publication: The following footnotes have
been edited. These authors contributed equally has
been corrected to The author contributed equally to
the rst author. and The author contributed equally
to correspondence has been corrected to The
author contributed equally to corresponding
author.]

Key words: aged, benzodiazepines, depression,

diabetes mellitus, hypnotics, sedatives.

Abstract
Background: This study examined the relationship between depression,
benzodiazepine (BZD)/nonbenzodiazepine hypnotics (non-BZD), and other
risk factors in a national sample of Taiwans elderly diabetic patients.
Methods: Data were drawn from the 2005 Taiwan National Health Interview
Survey and adults aged 65 years and older. A total of 1331 subjects
were included in this study. The Chinese version of Center for Epidemiologic Studies Depression Scale was used to evaluate patients depression
symptoms.
Results: The rates of depression in the diabetes mellitus (DM) and non-DM
groups were 13.5% (39/288) and 9.8% (102/1043) and the average ages
were 73.7 and 73.4 years, respectively. In multivariate regression, the odds
ratio of depression was 1.66-fold higher among BZD/non-BZD users (95%
condence interval: 1.102.51, model 2) than among those without BZD/
non-BZD use. In addition, hyperlipidaemia, poor physical function, and antidepressant use were associated with a higher risk of depressive symptoms.
Meanwhile, a monthly household income of NT$30 000NT$49 999, exercise, and betel chewing were associated with a lower risk of depression. We
performed an additional logistic analysis for which the odds ratio of depression signicantly increased to 1.52 in non-DM elderly patients (95% condence interval: 1.062.19) who were prescribed BZD/non-BZD. In contrast,
there was no signicant difference in the odds ratio of depression in the DM
elderly regardless of BZD/non-BZD use, although there was a slight tendency for depression among those who used BZD/non-BZD.
Conclusion: Depression in non-DM Taiwanese elderly patients was found
to be associated with BZD/non-BZD use, whereas depression in DM Taiwanese elderly patients was not found to be associated with BZD/non-BZD
use.

INTRODUCTION
Taiwans elderly population has experienced rapid
growth in recent decades, reecting a trend observed
in other industrialized countries, and the percentage
of older people aged 65 and older will reach 20% by
the year 2025.1 Older people frequently face major life
events like bereavement, changed social status after
retirement, loss of income, transition from indepen 2015 The Authors
Psychogeriatrics 2015 Japanese Psychogeriatric Society

dent living to assisted care, and loss of physical,

social, or cognitive function from illness.2
Depressive illness is one of the most serious health
problems in older populations, and it may lead to
unnecessary suffering, impaired functional status,
increased mortality, and excessive use of health-care
resources. The risk factors for late-life depression
include female sex, social isolation, marital status/

93

M-L. Tsai et al.

being widowed, divorced, or separated, lower socioeconomic status, comorbid medical conditions,
uncontrolled pain, insomnia, functional impairment,
and cognitive impairment.3 Insomnia is a risk factor for
developing dysthymia and depression in older adults,
and persistence of insomnia has been associated with
persistence of depression.4 Benzodiazepines (BZD)
and non-benzodiazepines (non-BZD) are frequently
used to treat insomnia in clinical practice, and they are
also capable of decreasing anxiety and controlling
convulsions. However, the most common adverse
effects associated with BZD/non-BZD are impaired
memory, residual daytime sedation, drowsiness, dizziness, light-headedness, cognitive impairment,
motor discoordination, and dependence.57 BZD/nonBZD are included in the list of potentially inappropriate
medications (PIM) in the updated 2012 version of the
Beers Criteria and are not recommended for the
elderly.8 It has been reported that improvements in
sleep with BZD/non-BZD use are statistically signicant, but the magnitude of the effect is small.9 In
addition, the increased risk of adverse events is statistically signicant, with an increased risk of falls, hip
fractures, cognitive impairment, and car accidents.10
Depression is common among people with diabetes mellitus (DM) and it is associated with poor outcomes. Recent studies have reported that the
prevalence of depression among Taiwanese with DM
type 2 was about 20.6%.11,12 Raval et al. reported that
depression was strongly associated with age
>54 years, central obesity, neuropathy, nephropathy,
peripheral vascular disease, diabetic foot disease,
and pill burden (>4), and the likelihood of depression
was not signicantly associated with duration of DM
and insulin use.13 Furthermore, people with DM type 2
have a 24% increased risk of developing depression.14 Depression and DM have thus frequently been
found to be associated with each other.
Therefore, in the present study we investigated the
associations between sociodemographic variables,
health-related factors, and the use of PIM BZD/nonBZD in an older population with DM type 2 with a
depressive tendency in Taiwan.

METHODS
Study population
The Taiwan National Health Interview Survey (NHIS) is
a countrywide cross-sectional health survey that provides data on the general health of residents in Taiwan.

94

Figure 1 The process of sample selected. DM, diabetes mellitus;

NHIS, National Health Interview Survey.

The 2005 NHIS was conducted by the Bureau of Health

Promotion in Taiwan. For this study, data were drawn
from the 2005 NHIS, and a representative random
sample of older adults aged 65 years and older was
obtained. There were 2727 subjects in the NHIS and
1760 of them agreed to have their data linked to the
National Health Insurance Research Database, which
includes medical expenditure data. Subjects were
excluded if they responded dont know to any item on
the survey, refused to provide a response to an item, or
were missing responses to any of the questions. The
nal total sample of patients included in the analysis
was 1331. Of these 1331 subjects, 288 (21.6%) were
DM elderly and 1043 (78.4%) were non-DM elderly. DM
status was self-reported by survey participants or
identied by a diagnosis of DM (International Classication of Diseases, Ninth Revision, Clinical Modication = 250 for Outpatient Department visit 2 or
admission 1 in 20042005). The process of sample
selection is shown in Figure 1.
Variables measured
Outcome variable
Depressive tendency was measured by the Chinese
version of the Center for Epidemiologic Studies
Depression Scale 10, which is a modication of the
original 20-item full-length version.1,15,16 The Center for
Epidemiologic Studies Depression Scale is composed
of 10 items, each assigned a weight scale ranging
from 0 to 3; the total maximum possible score is
30. A higher score on an item represents a greater
level of frequency of experiencing a given symptom
in the past week. A total score 12 is dened as
2015 The Authors
Psychogeriatrics 2015 Japanese Psychogeriatric Society

BZD/non-BZD for depression in DM elderly

depression, and <12 indicates no depressive symptoms. Higher scores correlate with a greater burden of
depression.17,18
Other independent variables
BZD/non-BZD, which are often used by patients
with depression and anxiety, may result in reduced
severity of depressive symptoms. However, because
of their common adverse effects, BZD/non-BZD
hypnotics are classied as PIM in the updated 2012
version of the Beers Criteria and are not recommended in elderly patients.8 In the present study,
the denition of BZD/non-BZD hypnotics was
based on the 2012 Beers Criteria, including denitions of short- and intermediate-acting BZD
(alprazolam, estazolam, lorazepam, oxazepam, and
triazolam), long-acting BZD (clorazepate, chlordiazepoxide, chlordiazepoxide-amitriptyline, clidiniumchlordiazepoxide, clonazepam, diazepam, and
urazepam), and non-BZD (eszopiclone, zolpidem,
and zaleplon).
Demographic variables such as age, sex, urbanization, body mass index (BMI), education level, marital
status, monthly household income, exercise, drinking,
smoking, chewing betel, Charlson comorbidity index
(CCI), hypertension, hyperlipidaemia, activities of daily
living (ADL), instrumental activities of daily living (IADL),
and antidepressant use were collected at baseline in
2005. Marital status was classied as married (living
with spouse) or single (living without spouse). The
educational level groups were classied as illiterate or
elementary school education (the reference group),
junior high school education, senior high school education, and university education or higher. In developing countries, urbanization was used as a proxy
measure of the increased risk of DM associated with
altered diet, obesity, decreased physical activity, and
other factors such as stress. Urbanization in this study
was divided into the following categories: (i) highly
urbanized; (ii) moderate urbanization; (iii) newly developed towns; and (ivvii) others (including (iv) general
towns; (v) ageing towns; (vi) agricultural towns; and (vii)
remote towns). BMI 24 was considered to be overweight. Monthly household income was grouped as
follows: (i) less than NT$30 000; (ii) NT$30 000
NT$49 999; and more than NT$50 000. The CCI is a
weighted summary measure of clinically important
concomitant diseases that has been adapted for use
with International Classication of Diseases, Ninth
2015 The Authors
Psychogeriatrics 2015 Japanese Psychogeriatric Society

Revision, Clinical Modication coded administrative

databases.1921 Associated comorbidity was evaluated
by the CCI according to the diagnoses recorded in the
NHIS. A scale of 0 (0 points), 1 (1 point), and 2
(2 points) with CCI was used. A strong association
between functional disability and depressive symptoms in older people has frequently been reported.2225
Therefore, ADL and IADL indicators were used to
evaluate subjects who were classied as having
restricted or unrestricted ADL.
Ethical approval
Our protocol was approved by the Institutional Ethics
Review Board of Chung Shan Medical University
Hospital.
Statistical analysis
This study employed a cross-sectional design. The
study population was characterized by descriptive
statistics. Logistic regression was used to predict the
factors that contributed to depression. All candidate
independent variables were examined for statistical
signicance (P < 0.05) in univariate analyses. Then,
multivariate analyses were performed using variables
that showed statistically signicant associations with
depression in the univariate models. The results of
tests with P < 0.05 were considered statistically signicant. All statistical analyses were performed with
Stata software, version 11.0 (2009, Stata, College
Station, TX).

RESULTS
Demographic data
Table 1 shows the demographic data of the study
population based on DM status: DM elderly group
(n = 288) and non-DM elderly group (n = 1043). There
were no signicant differences in age, sex, urbanization, education level, marital status, monthly household income, exercise, drinking, smoking, betel
chewing, ADL disability, and prevalence of non-BZD
users between the two groups. However, there were
signicant differences between the DM group and
the non-DM group with regard to BMI 24 (56.12 vs
44.88%), CCI 2 (17.34 vs 10.98%), hypertension
(59.65 vs 37.86%), hyperlipidaemia (35.36 vs
23.51%), IADL disability (67.50 vs 49.93%), prevalence of antidepressant users (20.86 vs 11.94%),

95

M-L. Tsai et al.

Table 1 Demographic data of study population
DM (n = 288)

Non-DM (n = 1043)

Unweighted sample Weighted percentage (%) Unweighted sample Weighted percentage (%)
Age, mean 1 SD
Sex
Female
Male
Urbanization
1 (Highest)
2
3
47
BMI
<24
24
Education
Elementary school
Junior school
High school
University
Marital status
No
Yes
Household monthly income#
<NT$30 000
NT$30 000NT$49 999
NT$50 000
Exercise
No
Irregular
Regular
Drinking
No
Yes
Smoking
No
Yes
Betel chewing
No
Yes
CCI
0
1
2
Hypertension
No
Yes
Hyperlipidaemia
No
Yes
ADL
Unrestricted
Restricted
IADL
Unrestricted
Restricted
Antidepressants
No
Yes
BZD/NonBZD
No
Yes

73.66 1 5.74

73.42 1 6.08

P-value
0.484
0.289

143
145

47.25
52.75

439
604

41.80
58.20

49
86
39
114

26.23
28.22
15.60
29.95

170
256
142
475

24.07
24.37
16.09
35.47

122
166

43.88
56.12

589
454

55.12
44.88

216
22
26
24

69.14
10.23
8.64
11.98

775
102
74
92

71.47
10.64
7.35
10.55

90
198

29.59
70.41

314
729

30.12
69.88

150
65
73

48.54
24.35
27.11

599
213
231

54.53
21.53
23.94

101
59
128

35.48
20.66
43.86

410
199
434

37.33
20.33
42.34

242
46

80.62
19.38

802
241

77.50
22.50

215
73

74.37
25.63

717
326

70.43
29.57

263
25

92.82
7.18

954
89

92.62
7.38

97
141
50

34.17
48.49
17.34

543
389
111

52.03
37.00
10.98

120
168

40.35
59.65

649
394

62.14
37.86

190
98

64.64
35.36

816
227

76.49
23.51

255
33

90.75
9.25

967
76

92.01
7.99

91
197

32.50
67.50

537
506

50.07
49.93

232
56

79.14
20.86

925
118

88.06
11.94

128
160

44.42
55.58

571
472

54.29
45.71

0.479

0.003*

0.800

0.858

0.263

0.845

0.509

0.410

0.911
<0.001**

<0.001**
0.001**

0.491
<0.001**
<0.001**
0.005*

*P < 0.5; **P < 0.01. #New Taiwan dollars; Charlson comorbidity index. ADL, activities of daily living; BMI, body mass index; BZD, benzodiazepines; CCI,
Charlson comorbidity index; DM, diabetes mellitus; IADL, instrumental activities of daily living; non-BZD, nonbenzodiazepines.

96

2015 The Authors

Psychogeriatrics 2015 Japanese Psychogeriatric Society

BZD/non-BZD for depression in DM elderly

and prevalence of BZD/non-BZD users (55.58 vs

45.71%).
Univariate analysis
Univariate correlates of depression are shown in
Table 2. Male sex, a higher level of education (university or higher), being married, monthly household
income equal to or more than NT$50 000, exercise,
and drinking were protectively associated with
depression. However, CCI = 1, chronic diseases
such as dyslipidaemia, ADL disability, IADL disability,
antidepressant use, and BZD/Non-BZD use were
associated with increased depressive tendency.
DM, age, urbanization, BMI, smoking, betel chewing, and hypertension showed no association with
depression.
Multivariate analysis
In multivariate logistic regression analyses, monthly
household income of NT$30 000NT$49 999 (odds
ratio (OR) = 0.52, 95% condence interval (CI) = 0.31
0.87), regular exercise (OR = 0.31, 95%CI = 0.14
0.67), irregular exercise (OR = 0.49, 95%CI =
0.290.84), and betel chewing (OR = 0.29, 95%CI =
0.100.85) were signicantly associated with a
decreased risk of depression. Nevertheless, hyperlipidaemia (OR = 1.52, 95%CI = 1.022.27), restricted
ADL (OR = 3.49, 95%CI = 1.806.75), restricted IADL
(OR = 4.79, 95%CI = 2.459.39), antidepressants use
(OR = 2.77, 95%CI = 1.754.37), and BZD/non-BZD
use (OR = 1.66, 95%CI = 1.102.51) were more likely
to have an increased risk of depressive symptoms.
Age, DM, sex, urbanization level, BMI, education
level, marital status, alcohol use, smoking, CCI, and
hypertension were not signicant factors (P > 0.05).
Logistic regression analyses were performed after
adjustment for age, sex, urbanization, BMI, education,
marital status, monthly household income, exercise,
drinking, smoking, betel chewing, CCI, hypertension,
hyperlipidaemia, ADL, IADL, and antidepressant use
(Table 3). Compared to non-DM elderly who did not use
BZD/non-BZD, non-DM elderly who did use BZD/nonBZD (OR = 1.52, 95%CI = 1.062.19) had a statistically signicant increased risk of depression. In
contrast, depression had no signicant effects on
non-DM elderly who did not use BZD/non-BZD. Nevertheless, DM elderly who were prescribed BZD/nonBZD seemed to have an increasing risk of depression
2015 The Authors
Psychogeriatrics 2015 Japanese Psychogeriatric Society

compared to non-DM elderly who did not use BZD/

non-BZD.

DISCUSSION
In this cross-sectional study, logistic regression analysis found that the OR of depression signicantly
increased to 1.52 in non-DM elderly Taiwanese
patients who were prescribed BZD/non-BZD
(95%CI = 1.062.19). However, there was no signicant difference in risk of depression in DM elderly
patients regardless of whether they were prescribed
BZD/non-BZD. These results seem to demonstrate
that the use of BZD/non-BZD has a greater effect on
depressive symptoms in non-DM elderly.
Several previous studies have reported that DM
patients have a higher prevalence of depression than
non-DM patients,13,26,27 and the depression level of
DM patients may be positively correlated with complications, disease duration, and treatment modality.28
Nevertheless, in the present results, there was no
signicant difference between DM elderly and
non-DM elderly patients with regard to the risk of
developing depression. This may have been due to
the different characteristics of research subjects and
the fact that our study employed a cross-sectional
design. Thus, it was not possible to determine
whether patients depression had been successfully
treated and which treatment modalities were used.
In the present study, BZD/non-BZD use signicantly increased risk of depression by 1.52-fold in
non-DM elderly patients. Insomnia is a major factor for
initiating BZD/non-BZD use,29 and the inappropriate
use of BZD/non-BZD has been shown to be independently associated with older age ( = 0.130) and
chronic illnesses ( = 0.120).30 Therefore, clinicians
treating insomnia in elderly patients should take all
medical conditions, including psychiatric illness, substance abuse, and sleep disorders, into consideration
if there is a possibility that they may cause or exacerbate insomnia. Reasonable approaches for elderly
insomnia include behavioural therapy, medication, or
both.26 Above all, pharmaceutical care should avoid
PIM and substitute more appropriate and safer medicines for them. For example, ramelteon, a melatonin
agonist,31,32 was approved by the US Food and Drug
Administration. It was demonstrated to provide a subjective benet with persistence of the effect for at
least 6 months, resulted in improvement in older
adults,3336 and had fewer side-effects associated with

97

M-L. Tsai et al.

Table 2 Multivariate logistic regression (BZD/NonBZD)
Univariate analysis

Multivariate analysis

95%CI
OR
Diabetes
No
Yes
Age
Gender
Female
Male
Urbanization
1 (Highest)
2
3
47
BMI
<24
24
Education
Elementary school
Junior school
High school
College
Marital status
No
Yes
Household monthly income
<NT$30 000
NT$30 000NT$49 999
NT$50 000
Exercise
No
Irregular
Regular
Drinking
No
Yes
Smoking
No
Yes
Betel chewing
No
Yes
CCI
0
1
2
Hypertension
No
Yes
Hyperlipidaemia
No
Yes
ADL
Unrestricted
Restricted
IADL
Unrestricted
Restricted
Antidepressants
No
Yes
BZD/non-BZD
No
Yes

95%CI

Lower

Upper

OR

Lower

Upper

1
1.19
1.02

0.73
0.99

1.94
1.06

1
0.81
0.96

0.47
0.93

1.41
1.00

1
0.39**

0.28

0.54

1
0.84

0.50

1.41

1
0.76
1.05
1.25

0.38
0.52
0.80

1.51
2.11
1.95

1
0.93
1.26
1.18

0.41
0.50
0.64

2.08
3.13
2.18

1
0.81

0.60

1.10

1
0.75

0.48

1.17

1
0.35
0.75
0.41*

0.11
0.39
0.17

1.07
1.42
1.00

1
0.72
1.70
1.02

0.19
0.88
0.34

2.81
3.25
3.04

1
0.51**

0.36

0.72

1
0.68

0.39

1.17

1
0.66
0.60*

0.43
0.37

1.03
0.95

1
0.52*
0.72

0.31
0.39

0.87
1.33

1
0.55**
0.27**

0.37
0.14

0.81
0.53

1
0.49*
0.31**

0.29
0.14

0.84
0.67

1
0.32**

0.18

0.58

1
0.73

0.43

1.25

1
0.59

0.33

1.05

1
1.04

0.49

2.23

1
0.50

0.19

1.32

1
0.29*

0.10

0.85

1
1.82*
1.27

1.13
0.53

2.93
3.03

1
1.04
0.71

0.70
0.31

1.55
1.63

1
1.28

0.85

1.92

1
0.91

0.53

1.58

1
1.50*

1.08

2.09

1
1.52*

1.02

2.27

1
7.28**

4.02

13.16

1
3.49**

1.80

6.75

1
7.55**

3.88

14.69

1
4.79**

2.45

9.39

1
4.22**

3.03

5.89

1
2.77**

1.75

4.37

1
2.64**

1.93

3.62

1
1.66*

1.10

2.51

*P < 0.5, **P < 0.01.

ADL, activities of daily living; BMI, body mass index; BZD, benzodiazepines; CCI, Charlson comorbidity index; CI, condence interval; IADL, instrumental
activities of daily living; non-BZD, nonbenzodiazepines; OR, odds ratio.

98

2015 The Authors

Psychogeriatrics 2015 Japanese Psychogeriatric Society

BZD/non-BZD for depression in DM elderly

Table 3 Logistic regression (BZD/NonBZD)
95%CI
DM

BZD/non-BZD PIM

OR

Lower

Upper

No
No
Yes
Yes

No
Yes
No
Yes

1
1.52*
0.58
1.40

1.06
0.19
0.80

2.19
1.79
2.43

*P < 0.5. Adjusted for age, sex, urbanization, body mass index, education,
marital status, monthly household income, exercise, drinking, smoking, betel
chewing, Charlson comorbidity index, hypertension, hyperlipidaemia, activity
of daily living, instrumental activity of daily living, and antidepressants.
BZD, benzodiazepines; CI, condence interval; DM, diabetes mellitus; nonBZD, nonbenzodiazepines; OR, odds ratio; PIM, potentially inappropriate
medications.

hypnotic side-effects (e.g. next-day residual performance decits), withdrawal, or rebound insomnia in
several randomized trials.37
Hermans and Evenhuis reported that increased
depressive symptoms were positively associated with
the number of life events during the past year and the
occurrence of chronic diseases (heart failure, stroke,
chronic obstructive pulmonary disease, coronary
artery disease, DM, and malignancy) in the previous
5 years; they also indicated that depressive symptoms were negatively associated with IADL.38
Pawaskar et al. also reported that increased risk of
depression was associated with lower health-related
quality of life and higher impairments in IADL in elderly
patients.39 In addition, our results showed that physical function disability and history of chronic disease
such as hyperlipidaemia were associated with
depression. Older people with both DM and physical
function impairments may be less capable of managing the disease, thus increasing their risk of depression. These ndings are consistent with the results of
the study by Pawaskar et al.
An interesting nding in the present study was that
chewing betel nut was associated with a reduced risk
of depression (OR = 0.29, 95%CI = 0.100.85). In previous studies, arecoline, the main areca alkaloid of the
betel nut, was reported to have antidepressant effects
in animal models,40,41 and arecoline has also been
demonstrated to possess an antidepressant property
via Monoamine oxidase A inhibition.42 Our results are
certainly compatible with those ndings. However, the
association in the present study was only found in the
univariate analysis. Whether chewing betel affects
depression remains a controversial issue.
Exercise, especially regular exercise, which has
psychological benets such as enhanced mental well 2015 The Authors
Psychogeriatrics 2015 Japanese Psychogeriatric Society

being and self-esteem, can promote mental and

physical health in the elderly.43 Moreover, systematic
reviews of controlled trials indicate that physical exercise is benecial for depressed patients 60 years and
older.4446 In our results, regular (OR = 0.31, 95%CI =
0.140.67) and irregular (OR = 0.49, 95%CI = 0.29
0.84) exercise appeared to be protective against
depression. Therefore, maintaining an exercise programme could be an effective method of relieving the
symptoms of depression in elderly people.
One of the major limitations of this research was
that half of the initial sample was not included in the
analysis. The sample size of the DM group (n = 288)
was relatively small compared with that of the non-DM
group (n = 1043), and the patients demographic characteristics were not comparable. Also, because this
investigation was a cross-sectional study, cause-andeffect relationships were not clear.
Conclusion
Depression in non-DM elderly patients in Taiwan was
found to be associated with BZD/non BZD use, and
depression in DM elderly patients in Taiwan was not
found to be associated with BZD/non-BZD use. Exercise, good lifestyle habits, appropriate treatment, and
suitable medications for older persons may promote
mental and physical health. Health providers should
avoid prescribing PIM, such as BZD/non-BZD, for
elderly patients and select safer alternatives that do
not increase risk of developing depressive symptoms.

ACKNOWLEDGMENTS
The authors thank Chung Shan Medical University
Hospital and Yu-Hsun Wang for their support and
contributions.

REFERENCES
1 Yen CH, Yeh CJ, Wang CC et al. Determinants of cognitive
impairment over time among the elderly in Taiwan: results of the
national longitudinal study. Arch Gerontol Geriatr 2010; 50
(Suppl 1): S53S57.
2 Birrer RB, Vemuri SP. Depression in later life: a diagnostic and
therapeutic challenge. Am Fam Physician 2004; 69: 23752382.
3 Cole MG, Dendukuri N. Risk factors for depression among
elderly community subjects: a systematic review and metaanalysis. Am J Psychiatry 2003; 160: 11471156.
4 Pigeon WR, Hegel M, Unutzer J et al. Is insomnia a perpetuating
factor for late-life depression in the IMPACT cohort? Sleep
2008; 31: 481488.
5 Bloom HG, Ahmed I, Alessi CA et al. Evidence-based recommendations for the assessment and management of sleep disorders in older persons. J Am Geriatr Soc 2009; 57: 761789.

99

M-L. Tsai et al.

6 Buscemi N, Vandermeer B, Friesen C et al. The efcacy and
safety of drug treatments for chronic insomnia in adults: a
meta-analysis of RCTs. J Gen Intern Med 2007; 22: 13351350.
7 Holbrook AM, Crowther R, Lotter A, Cheng C, King D. Metaanalysis of benzodiazepine use in the treatment of insomnia.
CMAJ 2000; 162: 225233.
8 American Geriatrics Society Beers Criteria Update Expert Panel.
American Geriatrics Society updated Beers Criteria for potentially inappropriate medication use in older adults. J Am Geriatr
Soc 2012; 60: 616631.
9 Glass J, Lanctot KL, Herrmann N, Sproule BA, Busto UE. Sedative hypnotics in older people with insomnia: meta-analysis of
risks and benets. BMJ 2005; 331: 1169.
10 Grad RM. Benzodiazepines for insomnia in community-dwelling
elderly: a review of benet and risk. J Fam Pract 1995; 41:
473481.
11 Huang MC, Hung CH. Quality of life and its predictors for
middle-aged and elderly patients with type 2 diabetes mellitus.
J Nurs Res 2007; 15: 193201.
12 Gavard JA, Lustman PJ, Clouse RE. Prevalence of depression in
adults with diabetes. An epidemiological evaluation. Diabetes
Care 1993; 16: 11671178.
13 Raval A, Dhanaraj E, Bhansali A, Grover S, Tiwari P. Prevalence
& determinants of depression in type 2 diabetes patients in a
tertiary care centre. Indian J Med Res 2010; 132: 195200.
14 Nouwen A, Winkley K, Twisk J et al. Type 2 diabetes mellitus as
a risk factor for the onset of depression: a systematic review and
meta-analysis. Diabetologia 2010; 53: 24802486.
15 Lee CT, Yeh CJ, Lee MC et al. Social support and mobility
limitation as modiable predictors of improvement in depressive
symptoms in the elderly: results of a national longitudinal study.
Arch Gerontol Geriatr 2012; 55: 530538.
16 Lue BH, Chen LJ, Wu SC. Health, nancial stresses, and life
satisfaction affecting late-life depression among older adults: a
nationwide, longitudinal survey in Taiwan. Arch Gerontol Geriatr
2010; 50 (Suppl 1): S34S38.
17 Cheng ST, Chan AC. Detecting depression in Chinese adults
with mild dementia: ndings with two versions of the Center for
Epidemiologic Studies Depression Scale. Psychiatry Res 2008;
159: 4449.
18 Cheng ST, Chan AC. The Center for Epidemiologic Studies
Depression Scale in older Chinese: thresholds for long and short
forms. Int J Geriatr Psychiatry 2005; 20: 465470.
19 Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new
method of classifying prognostic comorbidity in longitudinal
studies: development and validation. J Chronic Dis 1987; 40:
373383.
20 Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity
index for use with ICD-9-CM administrative databases. J Clin
Epidemiol 1992; 45: 613619.
21 Shen HN, Lu CL, Li CY. Epidemiology of rst-attack acute
pancreatitis in Taiwan from 2000 through 2009: a nationwide
population-based study. Pancreas 2012; 41: 696702.
22 Berkman LF, Berkman CS, Kasl S et al. Depressive symptoms
in relation to physical health and functioning in the elderly. Am
J Epidemiol 1986; 124: 372388.
23 Ormel J, Rijsdijk FV, Sullivan M, van Sonderen E, Kempen GI.
Temporal and reciprocal relationship between IADL/ADL disability and depressive symptoms in late life. J Gerontol B Psychol
Sci Soc Sci 2002; 57: P338P347.
24 Wells KB, Stewart A, Hays RD et al. The functioning and wellbeing of depressed patients. Results from the Medical Outcomes Study. JAMA 1989; 262: 914919.

100

25 Jiang J, Tang Z, Futatsuka M, Zhang K. Exploring the inuence

of depressive symptoms on physical disability: a cohort study of
elderly in Beijing, China. Qual Life Res 2004; 13: 13371346.
26 Vallieres A, Morin CM, Guay B. Sequential combinations of drug
and cognitive behavioral therapy for chronic insomnia: an
exploratory study. Behav Res Ther 2005; 43: 16111630.
27 Egede LE, Zheng D, Simpson K. Comorbid depression is associated with increased health care use and expenditures in individuals with diabetes. Diabetes Care 2002; 25: 464470.
28 Bai YL, Chiou CP, Chang YY, Lam HC. Correlates of depression
in type 2 diabetic elderly patients: a correlational study. Int J
Nurs Stud 2008; 45: 571579.
29 Manthey L, Giltay EJ, van Veen T et al. Determinants of initiated
and continued benzodiazepine use in the Netherlands study of
depression and anxiety. J Clin Psychopharmacol 2011; 31: 774
779.
30 Manthey L, van Veen T, Giltay EJ et al. Correlates of (inappropriate) benzodiazepine use: the Netherlands Study of Depression and Anxiety (NESDA). Br J Clin Pharmacol 2011; 71:
263272.
31 Turek FW, Gillette MU. Melatonin, sleep, and circadian rhythms:
rationale for development of specic melatonin agonists. Sleep
Med 2004; 5: 523532.
32 Kato K, Hirai K, Nishiyama K et al. Neurochemical properties of
ramelteon (TAK-375), a selective MT1/MT2 receptor agonist.
Neuropharmacology 2005; 48: 301310.
33 Mini LJ, Wang-Weigand S, Zhang J. Self-reported efcacy and
tolerability of ramelteon 8 mg in older adults experiencing
severe sleep-onset difculty. Am J Geriatr Pharmacother 2007;
5: 177184.
34 Zammit G, Erman M, Wang-Weigand S et al. Evaluation of the
efcacy and safety of ramelteon in subjects with chronic insomnia. J Clin Sleep Med 2007; 3: 495504.
35 Roth T, Seiden D, Wang-Weigand S, Zhang J. A 2-night,
3-period, crossover study of ramelteons efcacy and safety in
older adults with chronic insomnia. Curr Med Res Opin 2007;
23: 10051014.
36 Roth T, Seiden D, Sainati S et al. Effects of ramelteon on
patient-reported sleep latency in older adults with chronic
insomnia. Sleep Med 2006; 7: 312318.
37 Richardson GS, Zammit G, Wang-Weigand S, Zhang J. Safety
and subjective sleep effects of ramelteon administration in
adults and older adults with chronic primary insomnia: a 1-year,
open-label study. J Clin Psychiatry 2009; 70: 467476.
38 Hermans H, Evenhuis HM. Factors associated with depression
and anxiety in older adults with intellectual disabilities: results of
the healthy ageing and intellectual disabilities study. Int J Geriatr
Psychiatry 2013; 28: 691699.
39 Pawaskar MD, Anderson RT, Balkrishnan R. Self-reported predictors of depressive symptomatology in an elderly population
with type 2 diabetes mellitus: a prospective cohort study. Health
Qual Life Outcomes 2007; 5: 50.
40 Kumarnsit E, Keawpradub N, Vongvatcharanon U,
Sawangjaroen K, Govitrapong P. Suppressive effects of
dichloromethane fraction from the Areca catechu nut on
naloxone-precipitated morphine withdrawal in mice. Fitoterapia
2005; 76: 534539.
41 Dar A, Khatoon S, Rahman G, Atta-ur-Rahman. Anti-depressant
activities of Areca catechu fruit extract. Phytomedicine 1997; 4:
4145.
42 Dar A, Khatoon S. Behavioral and biochemical studies of
dichloromethane fraction from the Areca catechu nut.
Pharmacol Biochem Behav 2000; 65: 16.

2015 The Authors

Psychogeriatrics 2015 Japanese Psychogeriatric Society

BZD/non-BZD for depression in DM elderly

43 Ruuskanen JM, Ruoppila I. Physical activity and psychological
well-being among people aged 65 to 84 years. Age Ageing
1995; 24: 292296.
44 Sjosten N, Kivela SL. The effects of physical exercise on
depressive symptoms among the aged: a systematic review. Int
J Geriatr Psychiatry 2006; 21: 410418.
45 Frazer CJ, Christensen H, Grifths KM. Effectiveness of treatments for depression in older people. Med J Aust 2005; 182:
627632.

2015 The Authors

Psychogeriatrics 2015 Japanese Psychogeriatric Society

46 Blake H, Mo P, Malik S, Thomas S. How effective are physical

activity interventions for alleviating depressive symptoms in
older people? A systematic review. Clin Rehabil 2009; 23: 873
887.

101