All published articles of this journal are available on ScienceDirect.
The Magnitude of Suicidal Behavior among People Living with Diabetes Mellitus Attending an Outpatient Department of Alamata General Hospital, Mekelle, Tigray, Ethiopia 2019: A Cross-Sectional Study
Workua Mokenen Metekiya1
Birhane Gebrehiwot1
Authors Info & Affiliations
Abstract
Background:
Globally, suicide accounts for 75% in low and middle-income countries (LMICs). Though the magnitude of suicidal behavior in High-income Countries (HIC) is higher relative to the general population, limited studies had explored suicidal behavior among medical outpatients in LMICs including this locality. Suicidal behaviors among people with the chronic medical illness are one of the commonest psychiatric emergencies that demand a major health concern by researchers and mental health task forces. People with chronic medical illnesses show suicidal ideation and attempt which are fatal problems to end life. Therefore, this study will address the gaps by determining the magnitude of suicidal behavior among Diabetes Mellitus (DM) patients in an outpatient setting of Alamata General Hospital (AGH).
Methods:
Institutional based cross-sectional study was conducted among medical patients attending a chronic care clinic in Alamata general Hospital from May to June 2019. A sample of 146 DM patients who were attending an outpatient chronic care clinic was included in the study. Suicidal behavior was assessed by the World Health Organization (WHO) suicidal behavior assessment through software called Statistical Package for Social Science (SPSS) Version 25.
Results:
The magnitude of suicidal behavior among Diabetes Mellitus patients at AGH was 30.8%, 15.8% had suicidal ideation, 14.4% had a suicidal attempt and 15.1% of them had the plan to commit suicide.
Conclusion:
The prevalence of suicidal behavior was found to be significantly high in Diabetes Mellitus patients. Hence, it is important to conduct more interventions to assess the suicidal behavior symptoms among Diabetes Mellitus patients.
1. INTRODUCTION
Suicide is a fatal act of terminating one's own life and it is a complex process that involves a series of pathways and mechanisms from initiation of ideation, to planning, and finally to attempting suicide [1]. Some plan for days, weeks, or even years before the suicide attempt, while others take their lives without premeditation [2]. Suicidal ideation is an important phase in the suicidal process. Preceding attempted suicide is the major risk factor for completed suicide [2, 3]. It is known that suicidal ideas and suicide attempts occur more frequently in patients with DM than in the general population [4].
Diabetes Mellitus (DM) is a chronic disease that has become a serious public health problem worldwide [5]. Diabetes mellitus is a metabolic disorder of multiple etiological factors characterized by chronic hyperglycemia with disturbance of carbohydrate, fat, and protein metabolism which resulted from either insufficient Insulin secretion, resistance to the action of Insulin, or both [6]. It is a chronic disease that has become a serious public health problem worldwide [5, 7, 8]. Several studies support that DM has profound effects on physical and emotional health. Furthermore, DM has been associated with many adverse health effects including reduced life expectancy, increased risk of various complications, decreased quality of life, and even death [5, 7, 9, 10]. On the other hand, it has been reported that patients with chronic medical conditions such as type 2 diabetes mellitus (T2DM) are more likely to manifest psychiatric traits when compared with healthy individuals [11-14]. To date, the cause and effect relationship between these entities is not clear; we do not know whether diabetes increases the risk of psychiatric traits as depression, or whether psychiatric traits increase the risk of diabetes [15-17]. Nonetheless, differences have been reported between patients with type 1 or type II diabetes. For instance, a recent study of people who died by suicide in Finland reported that the proportion of patients with type II diabetes who completed suicide was twice the proportion of patients with type 1 diabetes who completed suicide [18]. The risk factors of suicidal behavior such as depression, anxiety, or hopelessness are present in patients with type 2 diabetes [13, 19]. Suicide is an important public health issue; which is the 10th leading cause of death worldwide, a second leading cause of death among those aged 15-29 [20]. An estimated number of more than 804,000 suicide bases occurred worldwide in 2012 and represent 1.8% of the global burden of disease [21]. It is estimated to contribute more than 2.4% of the global burden of disease by the year 2020 and the rate of death due to suicide will be increased to one every 20 seconds [3].
1.1. The Magnitude of Suicidal Ideation and Attempt
Suicidal ideation and attempt among people with DM had been reported as high as 58.5% [22] and 13.3% [23], respectively in the USA. Similarly, suicidal ideation was 51.4% in South Korea [24], and 20%, in Australia [17]. Suicidal behaviors in DM are higher in those with low-level education, female gender in Ethiopia [25], and alcohol use and cigarette smoking in the USA [26].
In Ethiopia, the magnitude of suicidal ideation and attempt among patients with severe mental illness is reported to be 23.3% for major depression, 23.8% for bipolar disorder, and 13.1% for schizophrenia [27]. Whereas another study reported in Ethiopia, Bahirdar that the lifetime prevalence of suicidal plan was 10.7%, and an attempt was found to be 7.6% [25].
The magnitude of suicidal behavior in patients with type 2 diabetes in Mexico was 11.6% [28] and diabetes increased the risk of death by suicide [14, 29]. The magnitude of previous suicide attempts in diabetic patients in Mexico was 11.4% [28]. It was found that suicide attempters with T2DM showed higher levels of blood glucose than those without a history of suicide. Old age can be a protective factor for suicidal behavior, as suicidal behavior magnitude decreases with age [30, 31] and younger individuals have more responsibilities and are under more stress than older individuals [28].
Deaths from hanging have increased from 16-24%. Most of the suicide deaths were reported in males of the 20 to 30 years age group. Deaths due to poisoning have increased from 4% to 28% over the past 8 years, with males predominating over females. Gunshot related death was also increased from 14% in 1996 to 25% in 2000. Over the same period, student university enrolment decreased from 7038 in 1996 to 3783 in 2000, a drop of nearly 50% [22].
In Ethiopia, mortality patterns from communicable and non-communicable diseases in Addis Ababa hospitals indicated that about 1.9% of deaths are attributed to suicide [23].
1.2. Associated Factors of Suicidal Ideation and Attempt
In South Korea, it was found that suicidal ideation was higher among those who are older than 50 years of age, less educated, divorce separated\widowed, and suffering from a co-morbid medical condition [24]. Similar studies from South Korea showed that insulin therapy duration in diabetes greater than or equal to 5 years of poor glycemic control [FBS≥126mg\dl] was significantly associated with suicidal attempt, whereas suicidal ideation was found higher among female patients (p=0.002). A history of alcohol abuse, other drug dependencies, and major depression past suicide attempts, and diagnosis of personality disorder have also been associated with suicide risk [32].
A cohort study in Sweden revealed the risk of suicide attempts in children with type 1 diabetes (hazard ratio [HR] 1.7 [95% CI 1.4–2.0]), and the risk for any psychiatric disorders was estimated to be HR 1.1 (95% CI 1.0–1.1) [9].
Suicidal ideation was found to be highly prevalent in those who had injected insulin in South Korea for more than 5 years [23].
Risk factor suicidality among diabetic patients in Canada was suicidal ideation (AOR=1.89, 95% CI: 1.36–2.63), attempted suicide AOR=1.45 (95% CI: 1.07–1.96), and completed suicide AOR= 1.85 (95% CI: 0.97–3.52) [33].
In Mexico, younger patients (AOR: 3.63, 95% CI: 1.29–10.19), having depression (AOR: 3.33, 95% CI: 1.13–9.76) and normal BMI (AOR: 3.14, 95% CI: 1.11–8.83), were predictive factors of suicide attempt [28].
In Italy, DM especially in patients with depressive symp- toms was found to be significantly associated with a marked increase in suicidal behaviors and suicidal ideation [13].
A study reported in Ethiopia, Bahirdar, found that being female (AOR = 2.14, 95% CI:1.10,5.65), poor social support (AOR = 3.21, 95% CI:1.26,8.98), comorbid depression (AOR = 6.40,95% CI:2.56,15.46) and poor glycemic control (AOR = 4.38,95% CI:1.66,9.59) were factors associated with lifetime suicidal attempt in diabetic patients [25].
Despite the high magnitude of suicidal ideation and attempt among the different chronic medical populations in different countries, no data showed such impact in Ethiopia. Therefore, the study aims to assess the magnitude of suicidal behavior among patients with DM visiting chronic illness clinic at AGH, Tigray, Ethiopia.
2. MATERIALS AND METHODS
2.1. Study Design and Period
The institution-based cross-sectional study was conducted from May to June 2019 G.C.
2.2. Study Area
Alamata is one of the eight weredas in the south Tigray, located 187 km by Korem line, and 174 km by Mehoni line from Mekelle. Total population of the town is 62, 735 (male = 30,488 & female=32,247), total household is 18,451 and the annual increment of the population is 4.61%. Alamata General Hospital was founded in 1983E.C and 255 staff are enrolled. It gives service to a population of 165,205.
2.4. Study Population
All DM patients who were attending Alamata General Hospital during the study period were included.
2.5. Inclusion and Exclusion Criteria
2.6. Sample size and Sampling Procedure
2.6.1. Sample Size Determination
The sample size was determined using a single population proportion formula by:
n= (z ᾳ/2)2p (q)
d2
Where n= sample size
z= confidence interval of 95%=1.96
d= marginal error; 0.05
n=1.962(0.5)(1-0.5)/0.052
=384
Since our sample population was less than 10,000 which was 200 we merged it to Nf using the correction formula
N= 200
Nf=n/1+n/N
=384/1+384/200
=132
Then, we added 10% of the Nf for the non-response rate.
132+14=146 is the sample size we used to conduct our study.
2.7. Sampling Techniques
Convenience random sampling technique was used to select each study participants from the AGH and the data was collected until the quota was full.
2.9. Independent Variables
2.10. Data Collection Method and Instruments
The study questionnaire had five components. Socio-demographic characteristics were collected by structured socio-demographic questionnaires. Depression and anxiety were collected by using HADS of seven stage questions and substance-related factors were collected by substance-related questionnaires. Social support was assessed by the Oslo-3 item social support scale, which is 3 item questionnaires, commonly used to assess social support and it has been used in several studies,
Outcome variables and suicidal behavior were assessed using a structured questionnaire. The data were collected through face-to-face interviews. There were 2 data collectors from fourth-year psychiatric nursing students. Data was collected in the AGH from May to June 2019.
2.11. Data Quality Control
To control the quality of the data, the questionnaire was designed and modified appropriately. The questionnaire was pre-tested one week before the actual data collection on 5% (8) of patients at AHC and was not included in the main survey. The data collectors were supervised daily and the field questionnaires were checked daily by the supervisors and principal investigator. When there was any problem the solution was given by discussion with the advisors and data collectors.
2.12. Data Analysis Technique
Data analysis technique and the collected data were analyzed using SPSS 23 software program. For this operation, proper data categorization and coding were used. For the analysis of obtained data, simple descriptive statistics (mean, percentage, frequencies, and standard deviation) and cross-tabulation were used to determine the magnitude of suicidal behavior among DM patients in AGH, Tigray, Ethiopia, 2019 G.C.
2.13. Operational Definition
Suicidal Behavior is defined if the respondent ever had either suicidal ideation or attempt.
Suicidal Ideation is defined as if the respondent answers the question have you ever seriously thought about committing suicide? If yes, the patient has suicidal ideation.
Suicidal Attempt is defined as if the respondent answers the question have you ever attempted suicide? If yes, the patient is likely to have a suicidal attempt.
Depression was assessed in individuals who scored≥8 on the depression subscale 7-items by assessing HADS.
Anxiety It was found in individuals, who scored ≥8 on anxiety subscale 7-items by assessing HADS.
Social Support: The presence of poor social support among medical patients was considered by the sum of the Oslo 3-item score of 3-8.
2.14. Ethical Consideration
Ethical clearance was obtained from the ethical review committee of Mekelle University. A formal letter of permission was obtained from MU and was submitted to AGH. The confidentiality of respondents was maintained. Informed written consent was obtained from each respondent and anyone not willing to take part in the study had full right to do so. Participants in the study who are highly suicidal were referred to the hospital for the assistant.
2.15. Dissemination and Utilization of a Result
After the data was analyzed, based on the findings obtained, conclusions and recommendations were made. Then the results of the study were submitted to AGH, Mekelle University, College of health science for future planning in routine assessments of mental health particularly suicide for a patient with DM in AGH. Moreover, the findings of the study were disseminated to the regional health bureau.
3. RESULTS
3.1. Socio-demographic Characteristics
The study compromised 146 diabetic patients, of which 113(77) were males. The majority 46(31.5) of the patients were between the age of 46-55 and the mean age was 44.91, 100(68.5) of them were orthodox Christian followers, 55 (37.7) have finished secondary school; more than half of the participants 79 (54.1) were married at the time of the study, nearly 1/3 of them, i.e., 45 (30.8) were a private employee; more than 3/4 of them, i.e., 113(77.4) were living with their family and most of them, i.e., 67(45.9) were earning more than 3000 ETB per month (Table 1).
| Sr. No. | Variables | Response Category | Frequency | % |
|---|---|---|---|---|
| 1 | Sex | Male | 113 | 77.4 |
| Female | 33 | 22.6 | ||
| 2 | Age | 18-25 | 12 | 8.2 |
| 26-35 | 22 | 15.1 | ||
| 36-45 | 44 | 30.1 | ||
| 46-55 | 46 | 31.5 | ||
| >55 | 22 | 15.1 | ||
| 3 | Religion | Orthodox | 100 | 68.5 |
| Muslim | 46 | 31.5 | ||
| 4 | Marital status | Single | 34 | 23.3 |
| Married | 79 | 54.1 | ||
| Widowed | 33 | 22.5 | ||
| 5 | Educational status | Unable to read & write | 34 | 23.3 |
| Able to read & write | 22 | 15.1 | ||
| Primary school | 25 | 17.12 | ||
| Secondary school | 55 | 37.7 | ||
| Diploma and university degree | 10 | 6.82 | ||
| 6 | Occupation | Unemployed | 34 | 23.3 |
| Government employee | 45 | 30.8 | ||
| Private employee | 45 | 30.5 | ||
| Farmer | 11 | 7.5 | ||
| Student | 11 | 7.5 | ||
| 7 | With whom are you living now | With family | 113 | 77.4 |
| Alone | 33 | 22.6 | ||
| 8 | Average monthly income | Less than 1000 | 12 | 8.2 |
| 1001-2000 | 55 | 37.7 | ||
| 2001-3000 | 12 | 8.2 | ||
| Greater than 3000 | 67 | 45.9 |
| Sr. No. | Variables | Response category | Suicidal behaviour | |
|---|---|---|---|---|
| Yes | No | |||
| 1 | Sex | Male | 23(20.4 | 90(79.6) |
| Female | 22(66.7) | 11(33.3) | ||
| 2 | Age | 18-25 | 0 | 12(100%) |
| 26-35 | 0 | 22(100%) | ||
| 36-45 | 11(25) | 33(75) | ||
| 46-55 | 12(26.1) | 34(73.9) | ||
| >55 | 22(100%) | 0 | ||
| 3 | Religion | Orthodox | 45(100%) | 0 |
| Muslim | 0 | 46(100%) | ||
| 4 | Marital status | Single | 0 | 34(100%) |
| Married | 12(15.2) | 67(84.8) | ||
| Widowed | 33(100%) | 0 | ||
| 5 | Educational status | Unable to read & write | 23(67.6) | 11(32.4) |
| Able to read & write | 11(50) | 11(50) | ||
| Primary school | 0 | 35(100%) | ||
| Secondary school | 11(20) | 44(80) | ||
| Diploma and university degree | 0 | 10(100%) | ||
| 6 | Occupation | Unemployed | 22(64.7) | 12(35.3) |
| Government employee | 0 | 11(100%) | ||
| Private employee | 0 | 45(100%) | ||
| Farmer | 11(100%) | 0 | ||
| Student | 12(26.7) | 33(73.3) | ||
| 7 | With whom are you living now | With family | 34(30.1) | 79(69.9) |
| Alone | 11(33.3) | 22(66.7) | ||
| 8 | Average monthly income | Less than 1000 | 12(100%) | 0 |
| 1001-2000 | 22(40) | 33(60) | ||
| 2001-3000 | 0 | 12(100%) | ||
| Greater than 3000 | 11(16.4) | 50(83.6) | ||
| 9 | Depression | Yes | 22(33.3) | 44(66.7) |
| No | 23(28.8) | 57(71.2) | ||
| 10 | Anxiety | Yes | 22(33.3) | 44(66.7) |
| No | 33(75) | 11(25) | ||
| 11 | Poor social support | Yes | 33(75) | 11(25) |
| No | 12(11.8) | 90(88.2) | ||
| 12 | Substance use | Yes | 45(36.3) | 79(63.7) |
| No | 0 | 22(100%) | ||
3.2. The Magnitude of Suicidal Behavior
Of 146 participants with DM, 45(30.8%) had suicidal behavior whereas 69.2% of participants had no such behavior. Of the study participants who had suicidal behavior, 22(48.88) were above the age of 55, and 23(51.12) participants were male, 33(73.3) of them were widowed, 22(48.88) of them had depression, 23(51.12) had anxiety, 33(73.3) of them had poor social support, and all of them, i.e., 45(100%) were substance users (Table 2).
4. DISCUSSION
Suicide is a fatal act of terminating one's own life and it is a complex process that involves a series of pathways and mechanisms from initiation of ideation, to planning, and finally to attempting suicide [1]. The finding of this study showed that the overall magnitude of suicidal behavior among diabetes mellitus patients in Alamata General Hospital was 30.8%, which is within the range of study performed in the USA which was 58.5% [22] and 13.3% [23], respectively in the USA. A cross-sectional study was conducted in Korea to assess a one-year magnitude of suicidal ideation in adults older than 20 years of age with DM and depression by taking 17065 subjects. Suicidal ideation was assessed by the self-administered question have you ever had any suicidal thoughts over the past year? and the result showed that 51.4% of patients had suicidal ideation [24] which was significantly higher than our finding which might be due to cultural differences, lack of religiosity which is one of the protective factors of suicide, lifestyle, and psychosocial stressors. Our finding was significantly different from a finding in Australia which was 20% and this might be due to better living standards, better management of the illness, and less psychosocial stressors compared to Ethiopia. One study observed that the magnitude of suicidal behavior in patients with type 2 diabetes was 11.6% [34] which is significantly different to our finding and this might be due to difference in the study design which was case-control and it might also be due to the fact that study only included T2DM patients. Suicidal behaviors in DM are higher in those with a low level of education, female gender in Ethiopia [32], alcohol use, and cigarette smoking in the USA [26], which was similar to our finding.
Another cross-section study conducted in South Korea assessed the one-year magnitude of suicidal ideation in adults older than 20 years of age with both diabetes and depression. It was found that suicidal ideation was higher among those who are older than 50 years of age, less educated, divorced, and separated/widowed [24], which is similar to our finding.
In this study, 66.7% of females had suicidal behavior and100% of those who were above 55 years of age had suicidal behavior. All widowed participants showed suicidal behavior, and 67.6% of uneducated DM patients had suicidal behavior. Out of those who had poor social support, 75% had suicidal behavior, 33% of depressed DM patients had suicidal behavior and also 36.3% of substance users had suicidal behavior. Out of participants that had suicidal behavior, 15.8% had suicidal ideation, 14.4% had a suicidal attempt and 15.1% had the plan to commit suicide.
CONCLUSION
This study demonstrated that suicidal behavior problem in diabetic patients is of high magnitude especially in individuals with sociodemographic factors like being female, widowed, and poor social support.
LIST OF ABBREVIATIONS
| AGH | = Alamata General Hospital |
| CI | = Confidence Interval |
| CMI | = Co-morbid Medical Illness |
| DM | = Diabetes Mellitus |
| ETB | = Ethiopian Birr |
| GC | = Gregorian Calendar |
| HIC | = High Income |
| LMIC | = Low and Middle Income |
| MU | = Mekelle University |
| OPD | = Out-patient Department |
| SBQR | = Suicidal Behavior Questionnaire-Revised |
| SB | = Suicidal Behavior |
| SPSS | = Statistical Package for Social Science |
| WHO | = World Health Organization |
| WMH | = World Mental Health |
AUTHORS’ CONTRIBUTION
The principal investigator was AT. Study conception and design, material preparation, data collection, and analysis were executed by AT, WM, and BG. All authors read and approved the final manuscript.
ETHICAL APPROVAL AND CONSENT TO PARTICIPATE
This study was carried out after obtaining ethical approval from Mekelle University, Ethiopia, College of Health Science office of Health Research Ethics Review Committee (HRERC) with the reference number of Notification of Expedited Approval ERC 1319/2019 and certifies that the study was executed by the ethical standards.
HUMAN AND ANIMAL RIGHTS
Not applicable.
CONSENT FOR PUBLICATION
Verbal consent was obtained from each participant before starting data collection. Study participants had the right to withdraw from the study at any time and information was recorded anonymously.
AVAILABILITY OF DATA AND MATERIALS
The authors confirm that the data supporting the findings of this study are available within the article with approval no. 1319/2019.
FUNDING
None.
CONFLICT OF INTEREST
The authors declare no conflict of interest, financial or otherwise.
ACKNOWLEDGEMENTS
Our sincere thanks go to Mekelle University College of Health Sciences, department of internal medicine, and psychiatry for its provision of the chance to conduct this study. We would like to thank Alamata General Hospital staff for the information given on diabetic outpatient units and data used for this research, data collectors, and participants without whom this thesis would not have been realized.
