MANAGEMENT OF ENDOCRINE DISEASE: Suicide risk in patients with diabetes: a systematic review and meta-analysis

in European Journal of Endocrinology
Correspondence should be addressed to J Zhang; Email: zhangjinan@hotmail.com

Background

Previous studies investigating the risk of suicide in diabetes patients reported controversial findings. We did a systematic review and meta-analysis to comprehensively estimate the risk and incidence rate of suicide in diabetic patients.

Methods

PubMed, EMBASE and PsycINFO were searched for eligible studies. Random-effects meta-analysis was used to calculate the relative risk (RR) and the incidence rate of suicide in diabetes patients. We also calculated the proportion of deaths attributable to suicide among diabetes patients.

Results

54 studies were finally included, including 28 studies on the suicide risk associated with diabetes, 47 studies on the incidence rate of suicide and 45 studies on the proportion of deaths attributable to suicide. Meta-analysis showed that diabetes could significantly increase the risk of suicide (RR = 1.56; 95% CI: 1.29–1.89; P < 0.001). Subgroup analysis showed that the RR of suicide associated with type 1 diabetes was 2.25 (95% CI: 1.50–3.38; P < 0.001). The pooled incidence rate of suicide in patients with diabetes was 2.35 per 10 000 person-years (95% CI: 1.51–3.64). The pooled proportions of long-term deaths attributable to suicide in type 1 diabetes patients and type 2 diabetes patients were 7.7% (95% CI: 6.0–9.8) and 1.3% (95% CI: 0.6–2.6), respectively.

Conclusion

This meta-analysis suggests that diabetes can significantly increase the risk of suicide. Suicide has an obvious contribution to mortality in diabetic patients, especially among type 1 diabetes patients. Effective strategies to decrease suicide risk and improve mental health outcomes in diabetes patients are needed.

Abstract

Background

Previous studies investigating the risk of suicide in diabetes patients reported controversial findings. We did a systematic review and meta-analysis to comprehensively estimate the risk and incidence rate of suicide in diabetic patients.

Methods

PubMed, EMBASE and PsycINFO were searched for eligible studies. Random-effects meta-analysis was used to calculate the relative risk (RR) and the incidence rate of suicide in diabetes patients. We also calculated the proportion of deaths attributable to suicide among diabetes patients.

Results

54 studies were finally included, including 28 studies on the suicide risk associated with diabetes, 47 studies on the incidence rate of suicide and 45 studies on the proportion of deaths attributable to suicide. Meta-analysis showed that diabetes could significantly increase the risk of suicide (RR = 1.56; 95% CI: 1.29–1.89; P < 0.001). Subgroup analysis showed that the RR of suicide associated with type 1 diabetes was 2.25 (95% CI: 1.50–3.38; P < 0.001). The pooled incidence rate of suicide in patients with diabetes was 2.35 per 10 000 person-years (95% CI: 1.51–3.64). The pooled proportions of long-term deaths attributable to suicide in type 1 diabetes patients and type 2 diabetes patients were 7.7% (95% CI: 6.0–9.8) and 1.3% (95% CI: 0.6–2.6), respectively.

Conclusion

This meta-analysis suggests that diabetes can significantly increase the risk of suicide. Suicide has an obvious contribution to mortality in diabetic patients, especially among type 1 diabetes patients. Effective strategies to decrease suicide risk and improve mental health outcomes in diabetes patients are needed.

Introduction

Suicide is one common cause of death with almost 1 million people dying from suicide worldwide each year, and it is a serious problem for global public health (1). Identification of risk factors of suicide is very important for the development of effective prevention strategies for suicide (2). Diabetes is also a public health issue worldwide and its global pandemic is still increasing obviously (3, 4). There are over 400 million people suffering from diabetes worldwide, and its prevalence is still increasing every year (4). Diabetes is also associated with increased risks of many subsequent diseases, such as cardiovascular diseases, cancers and chronic kidney disease (5, 6, 7, 8).

Mental diseases are common among patients with diabetes, such as depression and suicidal ideation, but mental health is still a neglected comorbidity in patients with diabetes (9, 10, 11). Recent studies have suggested possibly increased risk of externally caused death such as suicide in patients with diabetes (12, 13, 14, 15). However, the risk of suicide associated with diabetes has not received enough attention. To date, it is still unclear whether diabetes can result in increased risk of suicide. There was also lack of a comprehensive estimation of suicide risk associated with diabetes. Therefore, to provide a comprehensive estimation of the risk of suicide in diabetes patients, we carried out a systematic review and meta-analysis of published studies. In this meta-analysis, to appropriately evaluate the risk of suicide in diabetes patients, we investigated the risk of suicide associated with diabetes, the incidence rate of suicide in diabetic patients and the contribution of suicide to mortality in diabetic patients. This systematic review was registered at PROSPERO (CRD42016039709).

Methods

Search strategy and selection criteria

PubMed, EMBASE and PsycINFO were searched to identify eligible studies. Databases were searched from January 1980 to October 15, 2016. We updated the literature search in PubMed on February 16, 2017. We used keywords and MeSH searches with the combination of the following terms: (suicide OR suicides OR suicidal OR suicidality) AND (diabetes OR diabetic OR T1DM or T2DM OR NIDDM OR IDDM). Language restriction was not used. The reference lists of relevant articles or reviews were also retrieved to for more eligible studies.

The following predefined inclusion criteria were used: (a) cohort studies; (b) for studies assessing the risk of suicide associated with diabetes, the exposure of interest was diabetes and for studies on the rate of suicide, diabetes patients were the population of interest; (c) studies recruiting either type 1 diabetes mellitus (T1DM) patients or type 2 diabetes mellitus (T2DM) patients, or studies mixing T1DM patients and T2DM patients together were all considered eligible; (d) the outcomes of interest were the incidence rate of suicide, the proportion of deaths attributable to suicide and the risk estimates of suicide associated with diabetes, such as standardized mortality ratio (SMR), hazard ratio (HR), relative risk (RR) and odds ratio (OR); (e) reporting data on the outcomes of interests. Since a few studies did not report the risk or the incidence rate of suicide in either T1DM or T2DM patients separately, but reported outcomes in a population, which mixed T1DM patients and T2DM patients together, we thus firstly assessed the risk of suicide in diabetes patients regardless of its types. However, to make it more appropriate, we would then stratified the studies by the types of diabetes and evaluated the risk or the incidence rate of suicide in patients with either T1DM or T2DM separately. Studies reporting suicide attempts or suicidal ideation were excluded.

Data extraction and quality assessment

A standardized extraction form was used, and disagreements were resolved by discussion. Data were independently extracted by two reviewers from the full-texts of included studies. Study characteristics, such as first author’s name, number of participants, demographic characteristics, the number of person-years, incident cases of suicide, time of follow-up, completeness of follow-up, method of standardization of SMR and risk estimates (SMR, HR, RR or OR), were all extracted. Newcastle–Ottawa Scale was utilized to evaluate the bias risk of observational studies with comparison groups, which assessed studies on three domains including the selection of participants, the comparability of exposure group and non-exposed group, and the ascertainment of the outcomes (16). The comparability was mainly assessed by the adjustment for confounding factors or the standardization of SMRs. Four, two and three stars were scored for those three domains, respectively. Studies with less than 6 stars were deemed to have high risk of bias, and studies with one star for the selection of participants or outcome ascertainment, or zero for any domain were all deemed to have high risk of bias (17, 18). For studies without a comparison group, which reported data on the incidence rate of suicide or the proportion of deaths attributable to suicide among diabetes patients, the risk of bias was assessed by a modified version of Newcastle–Ottawa scale proposed by Rotenstein et al. (19). The modified scale assessed the risk of bias of those studies without a comparison group on five items including representativeness of participants, sample size, comparability between respondents and nonrespondents, ascertainment of suicide and thoroughness of descriptive statistics. One point was scored for each item in the modified bias scoring scale, and studies were judged to have low risk of bias (≥3 points) or high risk of bias (<3 points).

Statistical analysis

Meta-analyses were performed on the incidence rate of suicide, the risk estimate of suicide associated with diabetes and the proportion of deaths attributable to suicide, respectively. Because the incidence of suicide in diabetes patients was relatively low, OR and HR were approximately close to RR and were assumed to be the same measure and a common estimate of RR, and this strategy had been used in many published meta-analyses of observational studies (20, 21, 22, 23). Besides, SMR was in line regarded as one form of RR in epidemiological researches and was thus conventionally used as the same measure of RR in the present meta-analysis. As the results were presented separately for men and women in some studies, we firstly obtained a within-study summary RR using a fixed-effect meta-analysis. The pooled estimates with 95% confidence intervals (95% CIs) were calculated using a random-effects model (24). This meta-analysis used both the I2 statistic and the Cochran’s Q statistic to evaluate the heterogeneity (25, 26). I2 more than 50% indicated substantial heterogeneity among included studies. The proportion of deaths attributable to suicide was converted by the method of Freeman–Tukey double arcsine transformation before data synthesis, and then was back-transformed to the original pattern, and it was calculated using the METARATE command in the ‘meta’ package for R. The incidence rates of suicide were pooled using the ‘metan’ module of STATA. The commands above in detail were shown in Supplementary Table 1 (see section on supplementary data given at the end of this article). Sensitivity analysis was performed by excluding studies with high risk of bias. The estimates were also stratified by gender and types of diabetes. Publication bias was evaluated by funnel plot and Egger’s test (27). Meta-analysis was carried out using Stata (version 12, Stata Corp.) and R Studio (version 3.0.2). A P value less than 0.05 suggested statistical significance.

Results

Study selection and characteristics

1391 unique records were identified from literature search (Fig. 1). 127 articles with full-texts were screened and 54 studies were finally included into the meta-analysis (12, 13, 14, 15, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77). There were 28 studies on the suicide risk associated with diabetes (12, 13, 14, 15, 28, 29, 31, 32, 33, 34, 36, 38, 40, 41, 42, 43, 46, 47, 48, 50, 52, 54, 56, 60, 62, 63, 70, 76), 47 studies on the incidence rate of suicide in patients with diabetes (12, 14, 15, 28, 30, 31, 33, 35, 36, 37, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 70, 71, 73, 74, 75, 76, 77) and 45 studies on the proportion of deaths attributable to suicide in patients with diabetes (12, 14, 15, 28, 30, 31, 33, 35, 36, 37, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77).

Figure 1
Figure 1

Flow chart of study selection in the meta-analysis.

Citation: European Journal of Endocrinology 177, 4; 10.1530/EJE-16-0952

Table 1 described the main characteristics of those 28 studies on the suicide risk associated with diabetes (Table 1). Most studies were from North America and Europe, and all studies were published in English (Table 1). The mean or median time of follow-up ranged from 4.4 years to 28.7 years (Table 1). 14 studies reported SMRs of suicide in patients with diabetes, 6 studies reported HRs of suicide associated with diabetes and the other 8 studies reported RRs of suicide associated with diabetes (Table 1). 13 studies mixed T1DM patients and T2DM together and did not assess the risk of suicide in T1DM patients and T2DM patients separately (Table 1). Among those 28 studies, 11 studies reported data on risk estimates of suicide associated with T1DM, and 7 studies reported data on risk estimates of suicide associated with T2DM. According to the Newcastle–Ottawa scale, 24 studies had low risk of bias, while the other 4 studies had high risk of bias (Table 1). The scores for each term were shown in detail in Supplementary Table 2. Table 2 described the main characteristics of 47 studies on the incidence rate of suicide in diabetic patients (Table 2). According to the modified version of Newcastle–Ottawa scale, 37 studies had low risk of bias, while the other 10 studies had high risk of bias, which was shown in detail in Supplementary Table 3. Table 3 described the main characteristics of 45 studies on the proportion of deaths attributable to suicide in patients with diabetic patients (Table 3). According to the modified version of Newcastle–Ottawa scale, 36 studies had low risk of bias, while the other 9 studies had high risk of bias, which was shown in detail in Supplementary Table 4.

Table 1

Characteristics of 28 cohort studies assessing the relative risk of suicide associated with diabetes.

StudyCountryType of diabetesParticipantsFollow-up (years)SuicideOutcomesConfounding factors or variables used to standardize SMRBias risk
(76)UKT2DMT2DM: 187 968;Control: 908 0165805HRAge, sex, ethnicity, deprivation, calendar yearLow risk
(28)NorwayT1DM7871 patients16.820SMRAgeLow risk
(15)JapanMixed105 408: Normal; 4898: with diabetes19.4618RRAge, area, alcohol-drinking, BMI, cohabitation, employment status, hours of sleep, frequency of physical exercise, stress level, and history of major physical illnessesLow risk
(14)AustraliaT2DM1413: T2DM adults5660: non T2DM13.114HRAge, genderHigh risk
(29)DenmarkMixed1 849 110 older adults204792RRPeriod, age group, conjugal status, income level, comorbidity, psychiatric disordersLow risk
(13)SwedenMixed252 191: Diabetic 1 260 214 non-diabeticUnclear1126RRAge, sex, county of birthLow risk
(31)SwedenMixed7 140 589 adults88721HRAge, marital status, country of birth, education, employment status, income and urban/rural status, psychiatric and somatic disordersLow risk
(32)UKMixed4 005 640Unclear626RRAge, region of residence and deprivation score associated with patients’ area of residenceLow risk
(33)DenmarkMixed4 734 703111013RRAge, calendar year, diagnosis of severe mental illnessLow risk
(36)AustraliaT1DM1309: T1DM; 6451: age- and sex-matched controls7.66SMRAgeLow risk
(34)USAMixed4 005 662810 169HRAge, gender, BMI, psychiatric comorbidity, medical comorbidityLow risk
(38)UKT2DM849 434NA75RRAge, smoking status, hypertension, dyslipidemiaLow risk
(40)KoreaT2DM1 329 52514472HRExercise, smoking status, alcohol consumption, body mass index, height, blood pressure blood cholesterolLow risk
(41)USAMixed1 053 831 adults26464RRAge, education, BMI, smoking, alcohol intake, vegetable intake, red meat intake, physical activity, aspirin useLow risk
(42)FinlandT1DM17 30621.4110SMRAgeLow risk
(43)SwedenMixed879: patients;837 control152HRGender, ageHigh risk
(46)TaiwanMixed1792 patients72SMRAgeHigh risk
(47)UKT1DM4246 patients126SMRAgeLow risk
(48)13 countriesT1DM28 887 patients7.611SMRAgeLow risk
(50)NorwayT1DM1906 patients2413SMRAgeLow risk
(52)SwedenT1DM10 200 patients87SMRAge, genderLow risk
(54)TaiwanMixed256 036 patients4.4361SMRAgeLow risk
(56)SwedenMixed4097 patients510SMRAgeLow risk
(60)FinlandT2DM11 215 patients577SMRAgeLow risk
(62)SwedenT1DM91: T1DM28.72RRNoneHigh risk
(63)UKMixed578315.717SMRAgeLow risk
(12)DenmarkT1DM1682: patients1312SMRAgeLow risk
(70)USAT1DM or T2DM1200: T1DM; 1772: T2DM8.55SMRAge, genderLow risk

BMI, body mass index; HR, hazard ratio; NA, not available; RR, relative risk; SMR, standardized mortality ratio; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus.

Table 2

Characteristics of 47 cohort studies investigating the incidence rate of suicide in patients with diabetes.

StudyCountryType of diabetesParticipants, nFollow-up (years)SuicideBias risk
(76)UKT2DM187 968588Low risk
(77)TaiwanMixed92 54612327Low risk
(28)NorwayT1DM787116.820Low risk
(15)JapanMixed489819.441Low risk
(14)AustraliaT2DM141313.14Low risk
(30)USAT1DM1441277Low risk
(31)SwedenMixed7 140 58988721Low risk
(33)DenmarkMixed4 734 703111013Low risk
(36)AustraliaT1DM1309: T1DM; 6451 age- and sex-matched controls7.66Low risk
(35)USAT2DM4128106Low risk
(37)JapanT1DM1324357Low risk
(39)FinlandT1DM417496Low risk
(40)KoreaT2DM42 0821438Low risk
(41)USAMixed52 65526113Low risk
(42)FinlandT1DM17 30621.4110Low risk
(43)SwedenMixed879152High risk
(44)USAT1DM1075328Low risk
(45)USAT2DM41844.44Low risk
(46)TaiwanMixed179272High risk
(47)UKT1DM4246126Low risk
(48)13 countriesT1DM28 8877.611Low risk
(49)USAT2DM1234.31High risk
(50)NorwayT1DM19062413Low risk
(51)SwedenT1DM/ T2DM67718.517Low risk
(52)SwedenT1DM10 20087Low risk
(53)ItalyT1DM or T2DM31 264928Low risk
(54)TaiwanMixed256 0364.4361Low risk
(55)GermanyT1DM36741022Low risk
(56)SwedenMixed4097510Low risk
(57)USAT2DM710222Low risk
(58)UKT1DM18549.31Low risk
(59)FranceT2DM23752High risk
(60)FinlandT2DM11 215577Low risk
(61)AustriaT1DM118561High risk
(62)SwedenT1DM9128.72High risk
(63)UKMixed578315.717High risk
(64)UKT1DM845172Low risk
(65)SwedenT1DM491971Low risk
(12)DenmarkT1DM16821312Low risk
(66)USAT2DM35381Low risk
(67)SwedenT1DM or T2DM1467: T1DM; 334: T2DM2.63Low risk
(68)UKT1DM31081High risk
(71)NorwayT1DM1908102Low risk
(70)USAT1DM or T2DM1200: T1DM; 1772: T2DM8.55Low risk
(73)JapanMixed620.5–103High risk
(75)UKT1DM45182High risk
(74)USAMixed21 4472644Low risk

NA, not available; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus.

Table 3

Characteristics of 45 cohort studies investigating the proportion of deaths attributable to suicide in patients with diabetes.

StudyCountryType of diabetesParticipants, nFollow-up (years)Number of all deaths(suicide)Bias risk
(76)UKT2DM187 968540 286 (88)Low risk
(77)TaiwanMixed92 54612327Low risk
(28)NorwayT1DM787116.8241 (20)Low risk
(30)USAT1DM144127107 (7)Low risk
(33)DenmarkMixed4 734 7031155 424 (1013)Low risk
(36)AustraliaT1DM1309: T1DM; 6451: age- and sex-matched controls7.620 (6)Low risk
(35)USAT2DM4128101217 (6)Low risk
(37)JapanT1DM132435223 (7)Low risk
(39)FinlandT1DM41749455 (6)Low risk
(41)USAMixed52 6552638 559 (113)Low risk
(42)FinlandT1DM17 30621.41338 (110)Low risk
(43)SwedenMixed8791529 (2)High risk
(44)USAT1DM107532279 (8)Low risk
(45)USAT2DM41844.4581 (4)Low risk
(46)TaiwanMixed17927156 (2)High risk
(47)UKT1DM424612108 (6)Low risk
(48)13 countriesT1DM28 8877.6141 (11)Low risk
(49)USAT2DM1234.3110 (1)High risk
(50)NorwayT1DM190624103 (13)Low risk
(51)SwedenT1DM/ T2DM67718.5120 (17)Low risk
(52)SwedenT1DM10 200878 (7)Low risk
(69)UKT1DMUnclearUnclear98 (12)Low risk
(53)ItalyT1DM or T2DM31 26496749 (28)Low risk
(54)TaiwanMixed256 0364.416015 (361)Low risk
(55)GermanyT1DM367410251 (22)Low risk
(56)SwedenMixed4097558 (10)Low risk
(57)USAT2DM71022476 (2)Low risk
(58)UKT1DM18549.326 (1)Low risk
(59)FranceT2DM237512 (2)High risk
(60)FinlandT2DM11 215511 215 (77)Low risk
(61)AustriaT1DM118566 (1)High risk
(62)SwedenT1DM9128.712 (2)High risk
(63)UKMixed578315.73090 (17)Low risk
(64)UKT1DM8451744 (2)Low risk
(65)SwedenT1DM4919733 (1)Low risk
(12)DenmarkT1DM168213168 (12)Low risk
(66)USAT2DM353867 (1)Low risk
(67)SwedenT1DM or T2DM1467: T1DM; 334: T2DM2.610 (3)Low risk
(68)UKT1DM31087 (1)High risk
(71)NorwayT1DM19081020 (2)Low risk
(70)USAT1DM or T2DM1200: T1DM; 1772: T2DM8.51004 (5)Low risk
(72)USAMixedData from death certificatesUnclear233 (5)Low risk
(73)JapanMixed620.5–1025 (3)High risk
(75)UKT1DM45187 (2)High risk
(74)USAMixed21 447262474 (44)Low risk

NA, not available; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus.

Meta-analysis

Meta-analysis of 28 studies reporting risk estimates of suicide associated with diabetes showed that diabetes could significantly increase the risk of suicide (RR = 1.56; 95% CI: 1.29–1.89; P < 0.001, I2 = 93.2%) (Fig. 2). After excluding studies with high risk of bias, there was still an obvious association between diabetes and increased risk of suicide (RR = 1.56; 95% CI: 1.27–1.91; P < 0.001, I2 = 94.3%). Subgroup analysis by gender showed that diabetes was significantly associated with increased risk of suicide in both men (RR = 1.39; 95% CI: 1.23–1.57; P < 0.001, I2 = 54.6%) and women (RR = 1.36; 95% CI: 1.10–1.69; P = 0.004, I2 = 63.7%). Subgroup analysis by types of diabetes showed that the RR of suicide associated with T1DM was 2.25 (95% CI: 1.50–3.38; P < 0.001, I2 = 69.3%) (Fig. 3). However, there was a possible association between T2DM and increased risk of suicide (RR = 1.65, 95% CI: 0.95–2.85; P = 0.07, I2 = 95.0%). There was no obvious risk of publication bias in the meta-analysis of 28 studies reporting risk estimates of suicide associated with diabetes (PEgger’s test = 0.48).

Figure 2
Figure 2

Meta-analysis showed that diabetes could obviously increase the risk of suicide. 95% CI, 95% confidence interval; RR, relative risk; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus.

Citation: European Journal of Endocrinology 177, 4; 10.1530/EJE-16-0952

Figure 3
Figure 3

Forest plot in the subgroup analysis by types of diabetes. 95% CI, 95% confidence interval; RR, relative risk; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus.

Citation: European Journal of Endocrinology 177, 4; 10.1530/EJE-16-0952

Meta-analysis of 47 studies on the incidence rate of suicide showed that the pooled incidence rate of suicide in patients with diabetes was 2.35 per 10 000 person-years (95% CI: 1.51–3.64; I2 = 98.8%). After excluding studies with high risk of bias, the pooled incidence rate of suicide in patients with diabetes was 1.81 per 10 000 person-years (95% CI: 1.13–2.90; I2 = 99.0%). Subgroup analysis by types of diabetes showed that the pooled incidence rate of suicide in T1DM and T2DM patients was 2.25 per 10 000 person-years (95% CI: 1.66–3.05; I2 = 76.5%) and 2.56 per 10 000 person-years (95% CI: 1.56–4.19; I2 = 90.4%), respectively. Subgroup analysis by gender showed that the pooled incidence rates of suicide in male diabetic patients and female diabetic patients were 2.34 per 10 000 person-years (95% CI: 1.06–5.17; I2 = 99.3%) and 0.94 per 10 000 person-years (95% CI: 0.38–2.30; I2 = 98.5%) respectively.

Meta-analysis of 45 studies on the proportion of deaths attributable to suicide in diabetes patients showed that the pooled proportion was 3.5% (95% CI: 2.6–4.8; I2 = 97%). After excluding studies with high risk of bias, the pooled proportion of deaths attributable to suicide in diabetes patients was 3.5% (95% CI: 2.5–4.8; I2 = 98%). Subgroup analysis by types of diabetes showed that the pooled proportions in T1DM patients and T2DM patients were 7.7% (95% CI: 6.0–9.8; I2 = 67%) and 1.3% (95% CI: 0.6–2.6; I2 = 94%) respectively (Fig. 4). Subgroup analysis by gender showed that the pooled proportions of long-term deaths attributable to suicide in male diabetic patients and female diabetic patients were 2.9% (95% CI: 1.9–4.5; I2 = 98%) and 1.3% (95% CI: 0.8–2.2; I2 = 97%) respectively.

Figure 4
Figure 4

Meta-analysis of data on the proportion of deaths attributable to suicide in T1DM patients. 95% CI, 95% confidence interval; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus.

Citation: European Journal of Endocrinology 177, 4; 10.1530/EJE-16-0952

Discussion

Both diabetes and suicide are complex global health problems, and an accurate estimate of suicide risk associated with diabetes is crucial. However, the risk of suicide in diabetic patients is still not well understood and has not been comprehensively assessed. We therefore did a systematic review and meta-analysis to comprehensively estimate the risk and the incidence rate of suicide in diabetes patients. Our study is the first systematic review and meta-analysis investigating the risk of suicide in patients with diabetes. 54 studies were finally included, including 28 studies on the suicide risk in patients with diabetes, 47 studies on the incidence rate of suicide in patients with diabetes and 45 studies on the proportion of deaths attributable to suicide in diabetes patients. The findings suggested that diabetes was significantly associated with increased risk of suicide (RR = 1.56; 95% CI: 1.29–1.89; P < 0.001) (Fig. 2). The association was more profound in T1DM patients, and the RR of suicide associated with T1DM was 2.25 (95% CI: 1.50–3.38; P < 0.001). The pooled incidence rate of suicide in patients with diabetes was 2.35 per 10 000 person-years (95% CI: 1.51–3.64). The pooled proportions of deaths attributable to suicide in T1DM patients and T2DM patients were 7.7% (95% CI: 6.0–9.8) and 1.3% (95% CI: 0.6–2.6), respectively (Fig. 4). The above findings suggest an obvious risk of suicide in patients with diabetes.

Diabetes is a chronic metabolic disease, and there are about 400 million people with diabetes worldwide (4, 5, 78). This meta-analysis suggests that the incidence rate of suicide in patients with diabetes is 2.35 per 10 000 person-years, which indicates that there are about 94 000 completed suicides occurring in diabetes patients worldwide every year. In addition, the high proportion of deaths attributable to suicide in diabetes patients also shows a high risk of suicide in diabetes patients.

There are several possible explanations for the increased risk of suicide associated with diabetes. Depressive symptoms and suicidal ideation are common in patients with diabetes, both of which are associated with increased risk of completed suicide (10, 11, 79). Apart from depression, other mental health problems are also common in diabetes patients, which can also lead to increased risk of suicide (80, 81). In addition, patients with diabetes usually have functional disabilities and comorbidity, and as a result of these impairments, the low quality of life can further increase the severity of depression and the risk of suicide. However, the mechanism underlying the association between diabetes and suicide risk has not been well understood, and more studies are needed to further explore it.

This systematic review and meta-analysis has important implications for the management of diabetes. Given the increasing prevalence of diabetes worldwide and the increased risk of suicide associated with diabetes, it is urgent for us to develop effective solutions to decrease the suicide risk in diabetes patients, especially for T1DM patients. However, current guidelines for the treatment of diabetes provide few recommendations on the psychological monitoring and suicide prevention in diabetes patients. The findings from our meta-analysis suggest a great need to identify high-risk patients, and effective psychological support is necessary to decrease suicide risk among those patients (1, 2, 82, 83). Increased awareness of suicide risk associated with diabetes is also needed in clinical practice (11, 84). In addition, more studies are needed to develop effective strategies to decrease suicide risk and improve mental health outcomes in diabetes patients.

There is possibility of misclassification bias in the meta-analysis. Most included studies did not report whether they classified poisoning mortality into suicide, and some of those deaths listed as poisoning or drug overdose may be suicides, which may lead to an underestimation of the suicide risk in diabetes patients. The true proportion of deaths attributable to suicide in diabetes patients was also likely to be underestimated, because some suicide deaths may not be properly classified but were reported as poisoning or drug overdose in death certificates. Therefore, more population-based studies with high-quality data are urgently needed to estimate the magnitude of suicide risk in diabetes patients.

This meta-analysis had several other limitations. Firstly, there was obvious difference in the methodology across those included studies, which resulted in obvious between-study heterogeneity. Considering the obvious heterogeneity, the pooled estimates should be interpreted with caution. Secondly, the number of included studies was still relatively small in some subgroup analyses by types of diabetes or gender. More future studies with large number of participants are needed. Thirdly, deaths from suicide are usually difficult to ascertain. Misreporting of suicide deaths from families and relatives are also possible, which may result in imprecise estimates of suicide risk in diabetes patients (85). Finally, the outcomes from the subgroup analysis by types of diabetes showed that the risk of suicide associated with diabetes was likely to be more profound in T1DM patients, while there was a possible association between T2DM and increased risk of suicide (RR = 1.65, 95% CI: 0.95–2.85; P = 0.07). The non-significant association between T2DM and suicide risk may result from the small number of available studies, and more epidemiological studies are needed to further evaluate this association.

This meta-analysis suggests that diabetes can significantly increase risk of suicide. Suicide has an obvious contribution to mortality in diabetic patients, especially among T1DM patients. The findings from our meta-analysis suggest a great need to identify high-risk patients, and effective psychological support is necessary to decrease suicide risk among those patients. In addition, more epidemiological studies are needed to further evaluate the association between T2DM and suicide risk.

Supplementary data

This is linked to the online version of the paper at http://dx.doi.org/10.1530/EJE-16-0952.

Declaration of interest

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this review.

Funding

The present work was supported by grants from the National Natural Science Foundation of China (81471004).

References

  • 1

    TureckiGBrentDA. Suicide and suicidal behaviour. Lancet 2016 387 12271239. (doi:10.1016/S0140-6736(15)00234-2)

  • 2

    ZalsmanGHawtonKWassermanDvan HeeringenKArensmanESarchiaponeMCarliVHoschlCBarzilayRBalazsJ Suicide prevention strategies revisited: 10-year systematic review. Lancet Psychiatry 2016 3 646659. (doi:10.1016/S2215-0366(16)30030-X)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3

    MenkeACasagrandeSGeissLCowieCC. Prevalence of and trends in diabetes among adults in the United States, 1988–2012. JAMA 2015 314 10211029. (doi:10.1001/jama.2015.10029)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4

    NCD Risk Factor Collaboration (NCD-RisC). Worldwide trends in diabetes since 1980: a pooled analysis of 751 population-based studies with 4.4 million participants. Lancet 2016 387 15131530. (doi:10.1016/s0140-6736(16)00618-8)

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5

    PournarasDJle RouxCW. Type 2 diabetes: multimodal treatment of a complex disease. Lancet 2015 386 936937. (doi:10.1016/S0140-6736(15)00132-4)

  • 6

    TancrediMRosengrenASvenssonAMKosiborodMPivodicAGudbjornsdottirSWedelHClementsMDahlqvistSLindM. Excess mortality among persons with type 2 diabetes. New England Journal of Medicine 2015 373 17201732. (doi:10.1056/nejmoa1504347)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 7

    BrownriggJRHughesCOBurleighDKarthikesalingamAPattersonBOHoltPJThompsonMMde LusignanSRayKKHinchliffeRJ. Microvascular disease and risk of cardiovascular events among individuals with type 2 diabetes: a population-level cohort study. Lancet Diabetes and Endocrinology 2016 4 588597. (doi:10.1016/S2213-8587(16)30057-2)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8

    HuoXGaoLGuoLXuWWangWZhiXLiLRenYQiXSunZ Risk of non-fatal cardiovascular diseases in early-onset versus late-onset type 2 diabetes in China: a cross-sectional study. Lancet Diabetes and Endocrinology 2016 4 115124. (doi:10.1016/S2213-8587(15)00508-2)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9

    HandleyTEVenturaADBrowneJLRichJAttiaJRReddyPPouwerFSpeightJ. Suicidal ideation reported by adults with type 1 or type 2 diabetes: results from diabetes MILES-Australia. Diabetic Medicine 2016 33 15821589. (doi:10.1111/dme.13022)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 10

    PetrakFBaumeisterHSkinnerTCBrownAHoltRI. Depression and diabetes: treatment and health-care delivery. Lancet Diabetes and Endocrinology 2015 3 472485. (doi:10.1016/S2213-8587(15)00045-5)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 11

    The Lancet Diabetes Endocrinology. Poor mental health in diabetes: still a neglected comorbidity. Lancet Diabetes and Endocrinology 2015 3 393. (doi:10.1016/s2213-8587(15)00144-8)

    • Search Google Scholar
    • Export Citation
  • 12

    KyvikKOStenagerENGreenASvendsenA. Suicides in men with IDDM. Diabetes Care 1994 17 210212. (doi:10.2337/diacare.17.3.210)

  • 13

    WebbRTLichtensteinPDahlinMKapurNLudvigssonJFRunesonB. Unnatural deaths in a national cohort of people diagnosed with diabetes. Diabetes Care 2014 37 22762283. (doi:10.2337/dc14-0005)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 14

    DavisWAStarksteinSEBruceDGDavisTM. Risk of suicide in Australian adults with diabetes: the Fremantle Diabetes Study. Internal Medicine Journal 2015 45 976980. (doi:10.1111/imj.12853)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 15

    YamauchiTInagakiMYonemotoNIwasakiMAkechiTSawadaNIsoHNodaMTsuganeS. History of diabetes and risk of suicide and accidental death in Japan: the Japan Public Health Centre-based Prospective Study, 1990–2012. Diabetes and Metabolism 2016 42 184191. (doi:10.1016/j.diabet.2015.11.008)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16

    WellsGSheaBO’connellDPetersonJWelchVLososMTugwellP. The Newcastle-Ottawa Scale (NOS) for Assessing the Quality of Nonrandomised Studies in Meta-Analyses. Ottawa Health Research Institute2015. (Available at: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp)

    • Search Google Scholar
    • Export Citation
  • 17

    VialeLAlloteyJCheong-SeeFArroyo-ManzanoDMcCorryDBagaryMMigniniLKhanKSZamoraJThangaratinamS. Epilepsy in pregnancy and reproductive outcomes: a systematic review and meta-analysis. Lancet 2015 386 18451852. (doi:10.1016/S0140-6736(15)00045-8.)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 18

    SobhySZamoraJDharmarajahKArroyo-ManzanoDWilsonMNavaratnarajahRCoomarasamyAKhanKSThangaratinamS. Anaesthesia-related maternal mortality in low-income and middle-income countries: a systematic review and meta-analysis. Lancet Global Health 2016 4 e320e327. (doi:10.1016/S2214-109X(16)30003-1)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 19

    RotensteinLSRamosMATorreMSegalJBPelusoMJGuilleCSenSMataDA. Prevalence of depression, depressive symptoms, and suicidal ideation among medical students: a systematic review and meta-analysis. JAMA 2016 316 22142236. (doi:10.1001/jama.2016.17324)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    KivimakiMJokelaMNybergSTSingh-ManouxAFranssonEIAlfredssonLBjornerJBBorritzMBurrHCasiniA Long working hours and risk of coronary heart disease and stroke: a systematic review and meta-analysis of published and unpublished data for 603,838 individuals. Lancet 2015 386 17391746. (doi:10.1016/S0140-6736(15)60295-1.)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21

    KivimakiMVirtanenMKawachiINybergSTAlfredssonLBattyGDBjornerJBBorritzMBrunnerEJBurrH Long working hours, socioeconomic status, and the risk of incident type 2 diabetes: a meta-analysis of published and unpublished data from 222 120 individuals. Lancet Diabetes and Endocrinology 2015 3 2734. (doi:10.1016/S2213-8587(14)70178-0)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 22

    CornfieldJ. A method of estimating comparative rates from clinical data; applications to cancer of the lung, breast, and cervix. Journal of the National Cancer Institute 1951 11 12691275.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23

    KunutsorSKSeiduSKhuntiK. Statins and primary prevention of venous thromboembolism: a systematic review and meta-analysis. Lancet Haematology 2017 4 e83e93. (doi:10.1016/S2352-3026(16)30184-3)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 24

    DerSimonianRLairdN. Meta-analysis in clinical trials. Controlled Clinical Trials 1986 7 177188. (doi:10.1016/0197-2456(86)90046-2)

  • 25

    CochranWG. The combination of estimates from different experiments. Biometrics 1954 10 101129. (doi:10.2307/3001666)

  • 26

    HigginsJPThompsonSGDeeksJJAltmanDG. Measuring inconsistency in meta-analyses. BMJ 2003 327 557560. (doi:10.1136/bmj.327.7414.557)

  • 27

    EggerMDavey SmithGSchneiderMMinderC. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997 315 629634. (doi:10.1136/bmj.315.7109.629)

  • 28

    GagnumVSteneLCJenssenTGBerteussenLMSandvikLJonerGNjolstadPRSkrivarhaugT. Causes of death in childhood-onset Type 1 diabetes: long-term follow-up. Diabetic Medicine 2017 34 5663. (doi:10.1111/dme.13114)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 29

    ErlangsenAStenagerEConwellY. Physical diseases as predictors of suicide in older adults: a nationwide, register-based cohort study. Social Psychiatry and Psychiatric Epidemiology 2015 50 14271439. (doi:10.1007/s00127-015-1051-0)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 30

    OrchardTJNathanDMZinmanBClearyPBrillonDBacklundJYLachinJM. Association between 7 years of intensive treatment of type 1 diabetes and long-term mortality. JAMA 2015 313 4553. (doi:10.1001/jama.2014.16107)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 31

    CrumpCSundquistKSundquistJWinklebyMA. Sociodemographic, psychiatric and somatic risk factors for suicide: a Swedish National Cohort Study. Psychological Medicine 2014 44 279289. (doi:10.1017/S0033291713000810)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 32

    SinghalARossJSeminogOHawtonKGoldacreMJ. Risk of self-harm and suicide in people with specific psychiatric and physical disorders: comparisons between disorders using English national record linkage. Journal of the Royal Society of Medicine 2014 107 194204. (doi:10.1177/0141076814522033)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 33

    RibeARLaursenTMSandbaekACharlesMNordentoftMVestergaardM. Long-term mortality of persons with severe mental illness and diabetes: a population-based cohort study in Denmark. Psychological Medicine 2014 44 30973107. (doi:10.1017/S0033291714000634)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 34

    McCarthyJFIlgenMAAustinKBlowFCKatzIR. Associations between body mass index and suicide in the veterans affairs health system. Obesity 2014 22 269276. (doi:10.1002/oby.20422)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 35

    ColemanSMKatonWLinEVon KorffM. Depression and death in diabetes; 10-year follow-up of all-cause and cause-specific mortality in a diabetic cohort. Psychosomatics 2013 54 428436. (doi:10.1016/j.psym.2013.02.015)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 36

    CooperMNde KlerkNHJonesTWDavisEA. Clinical and demographic risk factors associated with mortality during early adulthood in a population-based cohort of childhood-onset type 1 diabetes. Diabetic Medicine 2014 31 15501558. (doi:10.1111/dme.12522)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 37

    MorimotoAOndaYNishimuraRSanoHUtsunomiyaKTajimaN. Cause-specific mortality trends in a nationwide population-based cohort of childhood-onset type 1 diabetes in Japan during 35 years of follow-up: the DERI Mortality Study. Diabetologia 2013 56 21712175. (doi:10.1007/s00125-013-3001-2)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 38

    GaoSJuhaeriJReshefSDaiWS. Association between body mass index and suicide, and suicide attempt among British adults: the health improvement network database. Obesity 2013 21 E334E342. (doi:10.1002/oby.20143)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 39

    AholaAJHarjutsaloVSaraheimoMForsblomCGroopPH. Purchase of antidepressant agents by patients with type 1 diabetes is associated with increased mortality rates in women but not in men. Diabetologia 2012 55 7379. (doi:10.1007/s00125-011-2347-6)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 40

    BattyGDKivimakiMParkISJeeSH. Diabetes and raised blood glucose as risk factors for future suicide: cohort study of 1 234 927 Korean men and women. Journal of Epidemiology and Community Health 2012 66 650652. (doi:10.1136/jech-2011-200464)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 41

    CampbellPTNewtonCCPatelAVJacobsEJGapsturSM. Diabetes and cause-specific mortality in a prospective cohort of one million U.S. adults. Diabetes Care 2012 35 18351844. (doi:10.2337/dc12-0002)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 42

    HarjutsaloVForsblomCGroopPH. Time trends in mortality in patients with type 1 diabetes: nationwide population based cohort study. BMJ 2011 343 d5364. (doi:10.1136/bmj.d5364)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 43

    TornCIngemanssonSLindbladUGudbjornsdottirS. Excess mortality in middle-aged men with diabetes aged 15–34 years at diagnosis. Acta Diabetologica 2011 48 197202. (doi:10.1007/s00592-011-0272-2)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 44

    SecrestAMBeckerDJKelseySFLaporteREOrchardTJ. Cause-specific mortality trends in a large population-based cohort with long-standing childhood-onset type 1 diabetes. Diabetes 2010 59 32163222. (doi:10.2337/db10-0862)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 45

    LinEHHeckbertSRRutterCMKatonWJCiechanowskiPLudmanEJOliverMYoungBAMcCullochDKVon KorffM. Depression and increased mortality in diabetes: unexpected causes of death. Annals of Family Medicine 2009 7 414421. (doi:10.1370/afm.998)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 46

    WuMCLeeWJTschenSMLinSYLeeITJengCYTsengLNLiuYMChenBYSheuWH. Predictors of mortality in hospitalized diabetic patients: A 7-year prospective study. Diabetes Research and Clinical Practice 2008 80 449454. (doi:10.1016/j.diabres.2007.11.020)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 47

    FeltbowerRGBodanskyHJPattersonCCParslowRCStephensonCRReynoldsCMcKinneyPA. Acute complications and drug misuse are important causes of death for children and young adults with type 1 diabetes: results from the Yorkshire Register of diabetes in children and young adults. Diabetes Care 2008 31 922926. (doi:10.2337/dbib7-2029)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 48

    PattersonCCDahlquistGHarjutsaloVJonerGFeltbowerRGSvenssonJSchoberEGyurusECastellCUrbonaiteB Early mortality in EURODIAB population-based cohorts of type 1 diabetes diagnosed in childhood since 1989. Diabetologia 2007 50 24392442. (doi:10.1007/s00125-007-0824-8)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 49

    BognerHRMoralesKHPostEPBruceML. Diabetes, depression, and death: a randomized controlled trial of a depression treatment program for older adults based in primary care (PROSPECT). Diabetes Care 2007 30 30053010. (doi:10.2337/dbib7-0974)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 50

    SkrivarhaugTBangstadHJSteneLCSandvikLHanssenKFJonerG. Long-term mortality in a nationwide cohort of childhood-onset type 1 diabetic patients in Norway. Diabetologia 2006 49 298305. (doi:10.1007/s00125-005-0082-6)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 51

    WaernbaumIBlohmeGOstmanJSundkvistGErikssonJWArnqvistHJBolinderJNystromL. Excess mortality in incident cases of diabetes mellitus aged 15 to 34 years at diagnosis: a population-based study (DISS) in Sweden. Diabetologia 2006 49 653659. (doi:10.1007/s00125-005-0135-x)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 52

    DahlquistGKallenB. Mortality in childhood-onset type 1 diabetes: a population-based study. Diabetes Care 2005 28 23842387. (doi:10.2337/diacare.28.10.2384)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 53

    GnaviRPetrelliADemariaMSpadeaTCartaQCostaG. Mortality and educational level among diabetic and non-diabetic population in the Turin Longitudinal Study: a 9-year follow-up. International Journal of Epidemiology 2004 33 864871. (doi:10.1093/ije/dyh089)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 54

    TsengCH. Mortality and causes of death in a national sample of diabetic patients in Taiwan. Diabetes Care 2004 27 16051609. (doi:10.2337/diacare.27.7.1605)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 55

    MuhlhauserISawickiPTBlankMOvermannHRichterBBergerM. Reliability of causes of death in persons with Type I diabetes. Diabetologia 2002 45 14901497. (doi:10.1007/s00125-002-0957-8)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 56

    WibellLNystromLOstmanJArnqvistHBlohmeGLithnerFLittorinBSundkvistG. Increased mortality in diabetes during the first 10 years of the disease. A population-based study (DISS) in Swedish adults 15-34 years old at diagnosis. Journal of Internal Medicine 2001 249 263270. (doi:10.1046/j.1365-2796.2001.00802.x)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 57

    GuKCowieCCHarrisMI. Mortality in adults with and without diabetes in a national cohort of the U.S. population, 1971–1993. Diabetes Care 1998 21 11381145. (doi:10.2337/diacare.21.7.1138)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 58

    WarnerDPMcKinneyPALawGRBodanskyHJ. Mortality and diabetes from a population based register in Yorkshire 1978–93. Archives of Disease in Childhood 1998 78 435438. (doi:10.1136/adc.78.5.435)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 59

    Bourdel-MarchassonIDubrocaBDecampsARichard-HarstonSEmeriauJPDartiguesJF. Five-year mortality in elderly French subjects from the PAQUID epidemiologival survey: the burden of diabetes. Diabetic Medicine 1998 15 830835. (doi:10.1002/(SICI)1096-9136(199810)15:10<830::AID-DIA683>3.0.CO;2-3)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 60

    KoskinenSVReunanenARMartelinTPValkonenT. Mortality in a large population-based cohort of patients with drug-treated diabetes mellitus. American Journal of Public Health 1998 88 765770. (doi:10.2105/AJPH.88.5.765)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 61

    SchoberESchneiderUFriedlHPUnsinnK. Early mortality in childhood diabetes in Austria – a population based cohort study. European Journal of Pediatrics 1997 156 1517. (doi:10.1007/s004310050543)

    • Search Google Scholar
    • Export Citation
  • 62

    IngbergCMPalmerMAmanJLarssonS. Social consequences of insulin-dependent diabetes mellitus are limited: a population-based comparison of young adult patients vs healthy controls. Diabetic Medicine 1996 13 729733. (doi:10.1002/(SICI)1096-9136(199608)13:8<729::AID-DIA147>3.0.CO;2-A)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 63

    SwerdlowAJJonesME. Mortality during 25 years of follow-up of a cohort with diabetes. International Journal of Epidemiology 1996 25 12501261. (doi:10.1093/ije/25.6.1250)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 64

    McNallyPGRaymondNTBurdenMLBurtonPRBothaJLSwiftPGBurdenACHearnshawJR. Trends in mortality of childhood-onset insulin-dependent diabetes mellitus in Leicestershire: 1940–1991. Diabetic Medicine 1995 12 961966. (doi:10.1111/j.1464-5491.1995.tb00406.x)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 65

    SartorGDahlquistG. Short-term mortality in childhood onset insulin-dependent diabetes mellitus: a high frequency of unexpected deaths in bed. Diabetic Medicine 1995 12 607611. (doi:10.1111/j.1464-5491.1995.tb00550.x)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 66

    HanisCLChuHHLawsonKHewett-EmmettDBartonSASchullWJGarciaCA. Mortality of Mexican Americans with NIDDM. Retinopathy and other predictors in Starr County, Texas. Diabetes Care 1993 16 8289. (doi:10.2337/diacare.16.1.82)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 67

    NystromLOstmanJWallSWibellL. Mortality of all incident cases of diabetes mellitus in Sweden diagnosed 1983–1987 at age 15–34 years. Diabetes Incidence Study in Sweden (DISS) Group. Diabetic Medicine 1992 9 422427. (doi:10.1111/j.1464-5491.1992.tb01811.x)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 68

    BothaJLParkerHRaymondNTSwiftPG. Diabetes diagnosed before the age of 2 years: mortality in a British cohort 8–17 years after onset. International Journal of Epidemiology 1992 21 11321137. (doi:10.1093/ije/21.6.1132)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 69

    LaingSPJonesMESwerdlowAJBurdenACGatlingW. Psychosocial and socioeconomic risk factors for premature death in young people with type 1 diabetes. Diabetes Care 2005 28 16181623. (doi:10.2337/diacare.28.7.1618)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 70

    MossSEKleinRKleinBE. Cause-specific mortality in a population-based study of diabetes. American Journal of Public Health 1991 81 11581162. (doi:10.2105/AJPH.81.9.1158)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 71

    JonerGPatrickS. The mortality of children with type 1 (insulin-dependent) diabetes mellitus in Norway, 1973–1988. Diabetologia 1991 34 2932. (doi:10.1007/BF00404021)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 72

    BildDGeissLSTeutschSMGabellaBHudspethBASchubertRMGollmarCWKaplanDJConnellFAWillJC. Sentinel health events surveillance in diabetes. Deaths among persons under age 45 with diabetes. Journal of Clinical Epidemiology 1988 41 9991006. (doi:10.1016/0895-4356(88)90039-X)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 73

    ShibataMKishiTIwataH. Clinical study of complications in dialyzed diabetics. Tohoku Journal of Experimental Medicine 1983 141 (Supplement) 417425. (doi:10.1620/tjem.141.Suppl_417)

    • Search Google Scholar
    • Export Citation
  • 74

    KesslerII. Mortality experience of diabetic patients. A twenty-six-year follow-up study. American Journal of Medicine 1971 51 715724. (doi:10.1016/0002-9343(71)90299-3)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 75

    MacGregorM. Juvenile diabetics growing up. Lancet 1977 1 944945. (doi:10.1016/S0140-6736(77)92236-X)

  • 76

    WrightAKKontopantelisEEmsleyRBuchanISattarNRutterMKAshcroftDM. Life expectancy and cause-specific mortality in type 2 diabetes: a population-based cohort study quantifying relationships in ethnic subgroups. Diabetes Care 2017 40 338345. (doi:10.2337/dc16-1616)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 77

    TsengCH. Factors associated with cancer- and non-cancer-related deaths among Taiwanese patients with diabetes after 17 years of follow-up. PLoS ONE 2016 11 e0147916. (doi:10.1371/journal.pone.0147916)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 78

    NathanDM. Diabetes: advances in diagnosis and treatment. JAMA 2015 314 10521062. (doi:10.1001/jama.2015.9536)

  • 79

    BoltonJMGunnellDTureckiG. Suicide risk assessment and intervention in people with mental illness. BMJ 2015 351 h4978. (doi:10.1136/bmj.h4978)

  • 80

    HoltRIMitchellAJ. Diabetes mellitus and severe mental illness: mechanisms and clinical implications. Nature Reviews Endocrinology 2015 11 7989. (doi:10.1038/nrendo.2014.203)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 81

    HuangCJWangSYLeeMHChiuHC. Prevalence and incidence of mental illness in diabetes: a national population-based cohort study. Diabetes Research and Clinical Practice 2011 93 106114. (doi:10.1016/j.diabres.2011.03.032)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 82

    ChwastiakLAFreudenreichOTekCMcKibbinCHanJMcCarronRWisseB. Clinical management of comorbid diabetes and psychotic disorders. Lancet Psychiatry 2015 2 465476. (doi:10.1016/S2215-0366(15)00105-4)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 83

    SnoekFJBremmerMAHermannsN. Constructs of depression and distress in diabetes: time for an appraisal. Lancet Diabetes and Endocrinology 2015 3 450460. (doi:10.1016/S2213-8587(15)00135-7)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 84

    ThombsBD. Routine depression screening for patients with diabetes. JAMA 2014 312 24122413. (doi:10.1001/jama.2014.14771)

  • 85

    FuhrDCCalvertCRonsmansCChandraPSSikanderSDe SilvaMJPatelV. Contribution of suicide and injuries to pregnancy-related mortality in low-income and middle-income countries: a systematic review and meta-analysis. Lancet Psychiatry 2014 1 213225. (doi:10.1016/S2215-0366(14)70282-2)

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation

If the inline PDF is not rendering correctly, you can download the PDF file here.

 

     European Society of Endocrinology

Related Articles

Article Information

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 815 698 20
PDF Downloads 445 420 34

Altmetrics

Figures

  • View in gallery

    Flow chart of study selection in the meta-analysis.

  • View in gallery

    Meta-analysis showed that diabetes could obviously increase the risk of suicide. 95% CI, 95% confidence interval; RR, relative risk; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus.

  • View in gallery

    Forest plot in the subgroup analysis by types of diabetes. 95% CI, 95% confidence interval; RR, relative risk; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus.

  • View in gallery

    Meta-analysis of data on the proportion of deaths attributable to suicide in T1DM patients. 95% CI, 95% confidence interval; T1DM, type 1 diabetes mellitus; T2DM, type 2 diabetes mellitus.

Cited By

PubMed

Google Scholar