The burden and determinants of mental health disorders in Abu Dhabi, United Arab Emirates: a retrospective cohort

Abstract

Background

Assessing the burden of mental health disorders is critical for planning the best healthcare services and responding to community needs. This study aimed to assess the mental health burden in Abu Dhabi, United Arab Emirates and its important determinants.

Methodology

This retrospective cohort study included participants from a national cardiovascular screening program (Weqaya) in Abu Dhabi, United Arab Emirates, from 2010 to 2013. The data were collected through electronic medical records (EMRs). A total of 8699 subjects were included, with an average follow-up period of 9.2 years. Participants’ EMRs were reviewed in 2023 by physicians and nurses through electronic medical records for a diagnosis of mental health disorders or any psychiatric medication prescribed.

Results

The prevalence of mental health disorders among this cohort was 3.8%. Compared with males, females had a greater prevalence of mental illnesses. The highest prevalence of mental health disorders was among younger age groups (< 30 years old). Mood disorders had the highest prevalence. A history of psychiatric illnesses was significantly associated with the following baseline characteristics at screening: prediabetes mellitus status, smoking status, coronary heart disease status, dyslipidemia at screening, high-density lipoprotein (HDL) level and female sex. With regard to associated outcomes during the follow-up years, psychiatric illness was significantly associated with smoking at any time, chronic kidney disease, metabolic dysfunction-associated fatty liver disease (MUFLD), and diabetes mellitus. Moreover, logistic regression revealed that death, as a dependent variable, was associated with an increased risk of mortality. Such an increase in mortality compared with that in the general population was not found among patients with psychiatric illnesses who were using statins.

Conclusion

The prevalence of diagnosed and treated mental health disorders was not as high as that in other regions of the world, but when adjusted for age, it was close to the average worldwide prevalence. The discrepancy between questionnaire-based prevalence and our prevalence highlights the mental health burden that necessitates further studies. Furthermore, more studies are needed to estimate the prevalence in the total population of the UAE. The higher risk of mortality found in this study indicates that mental illness is either a risk factor for death or an outcome of higher-risk patients, which also needs to be confirmed by future studies.

Text box 1. Contributions to the literature
• There is limited evidence on the prevalence and risk factors of mental health conditions in our region.
• There was no supporting evidence in our region whether patients with chronic diseases are more susceptible to mental health conditions or not.
• Links between certain risk factors and health behaviors were found to be significantly associated with having mental health disorders.

Introduction

Mental and physical health are fundamentally linked, as both represent the core of health [1]. It has been established that there is a bidirectional association between mental and physical health. Moreover, many studies have identified a connection between mental distress and an increased risk of cardiovascular disease, cancer, type 2 diabetes, obesity, asthma, and stroke [2]. Mental distress is also associated with socioeconomic conditions, biological and psychological status, and lifestyle factors [2]. Mental disorders involve significant disturbances in cognition, emotional regulation, and behavior [3]. Although effective interventions and treatment options are available for many mental conditions, many barriers exist for patients seeking appropriate care [4]. Additionally, many people experience violations of human rights, discrimination, and stigma because of mental disorders, which affect their health-seeking behavior [5].

According to the latest update by the World Health Organization (WHO), one in every eight people in the world lives with a mental disorder, but regional differences exist [6]. In 2021, in the United States (US), the total prevalence of any mental health disorder was 22.8% [7]. In the Kingdom of Saudi Arabia (KSA), specifically in Riyadh city, the prevalence was 28.5% among primary healthcare patients [8].

In the United Arab Emirates (UAE), the prevalence of depression, for example, ranged widely from 12.5 to 28.6%. According to a recent study published in 2020, among the UAE population, 57.2% of participants suffered from at least one mental disorder [9]. In a recent report, an average of approximately 24,000 cases of mental health disorders were reported in the UAE in 2019, with depressive disorders and anxiety disorders being among the most common [10]. Seven out of 100,000 patients are admitted yearly for mental health disorders [11]. However, these estimates are cross-sectional and questionnaire-based, with no confirmatory psychiatric diagnosis.

Additionally, risk factors for mental health are best studied in community-based cohorts followed over time. To our knowledge, there are no studies in the UAE or the Arab world that have reported a community-diagnosis-based prevalence or cohort studies designed for that aim. Such regional epidemiological studies are important for better planning preventative strategies and effective psychiatric services at the policymaker level. Estimating the prevalence of different mental health conditions, their trends, and their determinants is as well critical to guide patients’ care screening and management guidelines.

This longitudinal retrospective study aimed to estimate the diagnosis-based prevalence of psychiatric illnesses, such as mood disorders, psychotic disorders, substance use disorders, and others in Abu Dhabi. In addition to determining the associations between psychiatric illnesses and other risk factors and medical co-morbidities such as chronic kidney disease, diabetes mellitus, obesity and smoking. As well as the association between mental health conditions and mortality was explored.

Methods

Study design and study setting

A retrospective cohort study was conducted in the United Arab Emirates (UAE) among the population of the Abu Dhabi Emirate. The sample included UAE nationals who joined a national screening program called Weqaya from 2010 to 2013. The UAE government initiated this program as a comprehensive screening program; it was assembled and monitored by the Department of Health of Abu Dhabi and carried out by the Abu Dhabi healthcare service company through ambulatory healthcare centers. The samples studied were a large segment of the Abu Dhabi population, representing 2.5% of the population at that time, 8699 participants. The program included self-reported indicators, anthropometric measures, and hematological parameters. Self-reported indicators were verified from the electronic medical records by the data collectors, who were nurses trained for Weqaya program by the government entity. Therefore, this can be a strength in data reliability. Regarding all lab-based indicators, Abu Dhabi Healthcare Services Medical Laboratory is the largest in the Abu Dhabi Emirate and implements high-quality assurance standards.

The study participants were UAE national individuals aged 18 years and above who had participated in the Weqaya program from 2010 to 2013. The average follow-up period was 9.2 years, with 4.2% lost follow-up after two years, and nearly 90% were followed for more than five years [12]. The retrospective assessment of this cohort aimed to study the prevalence, incidence, and determinants of a number of important health conditions such as cardiovascular diseases, diabetes, hypertension, and, among them, mental health conditions. Participants were reassessed in 2023 by physicians and nurses through electronic medical records. Charts of the subjects were reviewed for the diagnosis of mental health disorders and if any psychiatric medication was prescribed for any patient. The UAE has a countrywide program for monitoring controlled medications; psychiatric medications are part of this list. Therefore, if the charts did not show any clear diagnosis of mental health disorder, a search was conducted among the controlled medications list to increase the likelihood of identifying almost all diagnoses of psychiatric illnesses.

Furthermore, each prescribed medication has an International Classification of Diseases, Tenth Revision (ICD-10) diagnosis attached to it, which improved the accuracy of diagnosis, as it showed what type of mental illness the patient had to be prescribed a controlled medication. The data collected included baseline and follow-up vital signs, sociodemographic parameters, comorbid illnesses, mental disorders, cardiovascular risk factors, anthropometric parameters, hematological parameters, and death counts. The causes of death were documented on another electronic platform, so the causes of death were not included. For more details, refer to the main paper [12].

Sample size

The total sample size was 8699; all the subjects were included in our study and were not excluded.

Statistical analysis

All the statistical analyses were performed using SPSS software. Logistic regression was used to test the associations between multiple variables and mental health disorders. Cox regression was not possible as baseline psychiatric illnesses were not known, and therefore subjects free from mental health illnesses could not be determined at the start of follow up, and thus, logistic regression was used. The results are expressed as odds ratios (ORs). A P value < 0.05 was considered to indicate statistical significance.

Ethical approval and consent to participate

The study was approved by the Al Ain Human Ethics Committee, approval number 13/58, and Ambulatory Healthcare Services IRB 19-2022. All methods were carried out under relevant guidelines and regulations. The authors confirm that the study was conducted in accordance with the Helsinki Declaration.

Ethics approval and consent to participate were provided.

The IRBs waived informed consent because the study was designed to collect retrospective data gathered during patient care and anonymized at analysis.

Results

The prevalence of mental health disorders in this cohort was 3.8%. In our cohort, 169 patients (52%) were female, while 156 patients (48%) were male. Mood disorders (including depression and anxiety) had the highest score, with a prevalence of 1.8%, and it was more common in females than in males, followed by psychotic disorders, with a prevalence of 0.34%. Only four individuals had documented substance abuse; three were males, one was female, and the rest had other conditions.

In terms of age groups, the highest prevalence of mental health disorders was found in the young age group (< 30 years old), with 81 patients (24.9%), followed by the younger age group (40–49 years old), the younger age group (21.5%) and the younger age group (30–39 years old), with 68 patients (20.9%). Among patients with psychiatric illness, 99 (31.2%) had an intermediate education level, 76 (23.97%) had a university education, 65 (20.5%) were illiterate, 38 (11.98%) had a primary education level, 37 (11.6%) had a secondary education level, and finally, only 2 (0.6%) had postgraduate education. The highest numbers of psychiatric illnesses were observed in the in-paid employment category, with 99 (30.46%) patients, followed by housewives, with 89 (27.38%) patients. (Table 1).

Table 1 The burden and determinants of mental health disorders (2010 to 2013) in Abu Dhabi, UAE. Sociodemographic characteristics of participants

Acute coronary syndrome was found in 7 (2.15%) patients with psychiatric illness, compared to 112 (1.3%) patients without psychiatric illness (Table 2). Overall, out of 325 patients with psychiatric illnesses, 119 (36.6%) had obesity. According to the available data for 323 patients with psychiatric illnesses, 109 (33.7%) patients were overweight, and 6 (1.85%) patients were underweight.

Table 2 Demographics of different variables and risk factors in our targeted population

In total, among patients with psychiatric illness, 82 (25.23%) patients were diagnosed with hypertension. Compared to patients without psychiatric illness, 1508 (18.3%) patients had hypertension, and 6719 (81.7%) did not. Analysis of blood pressure readings in patients with psychiatric illness revealed that 139 (42.76%) patients out of our targeted psychiatric illness population had optimal blood pressure readings followed by normal blood pressure readings in 75 (23.07%) patients; 48 (14.7%) patients had stage I hypertension, and only 2 (0.6%) patients had stage II hypertension (Table 2).

Among the 266 patients with psychiatric illness, 119 (44.7%) had prediabetes mellitus. Compared to patients without psychiatric illness, prediabetes mellitus was detected in 2373 (33.95%) patients. In terms of diabetes, among patients with psychiatric illness, 97 (29.8%) had diabetes; compared to patients without psychiatric illness, diabetes was observed in 1749 (21.3%) patients.

Logistic regression using psychiatric illness diagnosis as the dependent variable in relation to baseline subjects’ characteristics assessed at screening revealed significant associations with prediabetes mellitus with an OR of 1.456 (p-value 0.011, 95% CI 1.089–1.948), smoking at the time of screening with an OR of 1.958 (p < 0.001; 95% CI 1.321–2.903), coronary heart disease at screening with an OR of 2.391 (p0.038; 95% CI 1.05–5.446), dyslipidemia diagnosis at screening with an OR of 1.733 (p0.003; 95% CI 1.205–2.492), higher HDL levels with an OR of 1.67 (p0.008; 95% CI 1.145–2.436), and gender, where psychiatric illnesses appeared to be more strongly associated with female gender than male gender with an OR of 0.645 (p 0.008; 95% CI 0.467–0.891) (Table 3). Age, hypertension, acute coronary syndrome, peripheral artery disease, stroke, statin use, and cancer were not significantly associated with psychiatric illness.

Table 3 Logistic regression with mental illness as the dependent variable

Other significant outcomes that were found to be associated with psychiatric illness as a dependent variable adjusted for baseline subject characteristics were smoking at any time, with an OR of 1.981 (p < 0.001; 95% CI 1.402–2.799); chronic kidney disease (old and new), with an OR of 1.538 (p0.022; 95% CI 1.065–2.222); metabolic dysfunction-associated fatty liver disease (MUFLD), with an OR of 1.854 (p < 0.001; 95% CI 1.345–2.556); and diabetes mellitus (old or new), with an OR of 1.499 (p-value 0.004, 95% CI 1.138–1.975).

With regard to the important outcomes assessed in this cohort of subjects, psychiatric illness showed interesting associations. Logistic regression, with death as the dependent variable, was found to be associated with an increased risk of mortality. (Table 4)

Table 4 Logistic regression with death as a dependent variable

Another important finding was that patients with psychiatric illnesses who were using statins did not have a greater risk of mortality (p-value 0.065, OR = 0.633; 95% CI 0.39–1.03). (Table 4).

Discussion

Our cohort study showed a confirmed psychiatric disorder prevalence of 3.8% in Abu Dhabi, UAE. This is lower compared to other countries, such as the United States of America (USA), with psychiatric disorders prevalence of 22.8%, 16.93% in Egypt, and 36.6% in Qatar, noting that these figures and not community-based and could be an overestimation of prevalence due to higher risk patients visiting healthcare facilities [7, 13, 14]. As well, the low prevalence in Abu Dhabi could reflect an underestimation of prevalence due to low screening rates in the community. An important area of research is to study social and familial determinants of mental health and the social support which may affect the need for pharmacological treatment or factors that can contribute to the low progression to severe psychiatric illness.

In the UAE, one recent cross-sectional study utilized a questionnaire-based assessment of prevalence, which was 57.2% compared to 28.5% in Saudi Arabia [8, 9]. This can reflect low utilization of healthcare or a less severe spectrum of mental health conditions. It is worth noting that a strength of this study is that it reports confirmed diagnosed patients and those treated for mental health conditions. It is also a longitudinal cohort study that reports all cases over time, and this methodology is better than questionnaire-based cross-sectional studies in identifying all confirmed diagnoses. As mentioned previously, in the UAE, there is a controlled psychiatric medication system integrated into patients’ electronic medical records; this system is governed by the Ministry of the Interior, which lists all psychiatric medications patients are taking; therefore, there is a very low possibility for patients treated for mental health disorders to be missed. Similarly, the age-adjusted prevalence of mental health conditions could not be identified in the literature and was 16.98% in this study. It is thought that this is a crucial measure for assessing the burden and formulating comparative assessments between countries.

As the cohort recruitment was random from the community and participation was required for the patients to obtain their health care services, this sample very much represents the community. Cross-sectional studies may describe those accessing healthcare more. Mental health screening at the time of the baseline assessment from 2010 to 2013 was not included, but screening in Abu Dhabi was introduced a few years later, which could have changed the trend in prevalence. Therefore, surveillance of the prevalence trend is needed; otherwise, those with the most severe conditions that are frequently encountered due to chronic illness or mental health disorders are included.

There are still factors affecting proper diagnosis of mental health disorders that need to be evaluated, such as stigma, fear of breaking confidentiality, and lack of practitioners’ knowledge about the importance of the diagnosis and treatment of these diseases. In a systematic review by Schnyder et al. of twenty-seven studies, participants’ own negative attitudes toward mental health help-seeking and their stigmatizing attitudes toward people with a mental illness were associated with less active help-seeking [15].

The distribution of risk factors in this study was similar to that in other international studies. In terms of sex, our population showed a slightly greater prevalence of mental health disorders in females than in males. In the US, a similar observation was noted, where females had a greater prevalence of mental health disorders than males, with a prevalence of 27.2% versus 18.1% respectively [7]. Compared to a study in the KSA in 2023, among young adults, gender differences were observed only in the prevalence of specific disorders. The prevalence of anxiety-social phobia (8.87% vs. 5.08%), obsessive-compulsive disorder (5.59% vs. 2.59%), and major depressive disorder (8.78% vs. 3.24%) was greater in females than in males [16]. Furthermore, more mental health diseases were found in young age groups, and similar findings were observed in both the US and KSA [7, 16]. One possibility is that care-seeking behavior is generally greater among women than among men. Health-seeking behavior was not assessed in this study, nor was the severity of the conditions studied, which can influence such differences in prevalence. Suicides tend to be more common in men, which may suggest that mental health is worse in men than in women [17]. Therefore, it is not recommended to give less priority to male patients; rather, such findings should be used to inform healthcare service planning and institute strong screening and management programs.

In 2020, according to the WHO, during and after the COVID-19 pandemic, the prevalence of anxiety and depression increased. The estimates showed a 26% and 28% increase in both anxiety and depression, respectively, within one year. It is also estimated that 4% of the global population is currently living with an anxiety disorder; that is, 301 million people in 2019 had an anxiety disorder, making it the most prevalent mental disorder of all mental disorders [3]. All these findings were similar to our findings on mood disorders, which are more than one-third of mental conditions reported.

In this cohort, a nonsignificant association was found between BMI and psychiatric illnesses. Compared to worldwide regions, obesity is linked to multiple psychiatric disorders. A review of association studies performed in 2017 showed a bidirectional association between obesity and depression that was clinically significant, especially among women, where they found that depression is associated with obesity and vice versa [18]. A possible explanation is that obesity is already a risk factor for diabetes and prediabetes mellitus, both of which are significantly associated with psychiatric illness.

Similarly, stroke was not associated with a higher incidence of psychiatric illnesses. However, a study published in Denmark in 2022 showed that there is a greater risk of mental disorders after having a stroke, especially mood disorders, which highlights the importance of mental health evaluation post-stroke [19]. Regardless of the lack of associations, screening is still needed in poststroke care, in which we can detect or even prevent the development of mental health disorders poststroke.

In this study, high HDL levels were significantly associated with psychiatric illnesses. Similar findings were reported in one study involving patients older than 18 years who were diagnosed with major depressive disorder. At the evaluation visit, serum lipid profiles, including total cholesterol, triglyceride (TG), low-density lipoprotein (LDL), high-density lipoprotein (HDL), and very low-density lipoprotein (VLDL) levels, and other blood parameters were measured. Those with recent suicide attempts or recent suicide status had significantly lower TG and higher HDL levels [20]. However, there was no clear scientific relationship between the two variables.

This study showed that patients on statins had better survival, which can indicate better access to care and better management of metabolic conditions. On the other hand, logistic regression revealed greater odds of death in patients with psychiatric illness. This was noted in a large population based cohort done in Denmark, where comorbid general medical conditions such as heart disease, diabetes, and cancer could contribute substantially to reduction in life [21].

There are other causes of premature death in patients with mental health disorders, including self-harm, substance abuse, and poor social determinants of health. Suicide is the fourth leading cause of death among patients aged 15–29 years in the US [22]. According to the Institute for Health Metrics and Evaluation (IHME), in the UAE, in 2019, substance abuse was the eighth leading cause of death, with a 57.8% increase in deaths attributed to substance use between 2009 and 2019, and self-harm was the tenth leading cause of death, with a 28.7% increase in deaths attributed to self-harm between 2009 and 2019 [23]. Additionally, other factors, such as immune-inflammatory disturbances and hypovitaminosis D, have been suggested [24].

Such differences in survival between patients with and without mental illness were also found to be attributed to social factors, family composition, and health-seeking behavior [2]. Psychiatric medication adherence and access to care may be important factors. According to a retrospective cohort study of adult Maryland Medicaid beneficiaries with schizophrenia and any antipsychotic use from 1994 to 2004, mental health visits were associated with decreased mortality, and adherence to medications was associated with reduced mortality [25].

The role of primary care through family medicine to provide continuity and early detection and management of chronic disease in patients with mental conditions rather than focusing on psychiatric management alone [26]. Family medicine promotes the integration of behavioral and physical models of illness and plays a vital role in providing both mental and behavioral health care services. The established relationship between a patient and their family physician creates a greater opportunity for continued support and care when navigating mental health concerns. Through this connection, primary care physicians may be able to recognize the onset of mental illness in their patients and have better preexisting knowledge of any relevant social, emotional, or environmental factors, as well as comorbidities and other considerations that can contribute to overall mental health. The long-term relationship between a patient and their family physician often creates a greater opportunity for continued support throughout the life cycle, provides continuity of care, and helps patients navigate mental health concerns. Additionally, primary care physicians are best equipped to take comorbid conditions and polypharmacy concerns into account when managing mental illnesses such as depression and anxiety in older adults [27,28,29]. On the other hand, antipsychotic treatment was found to alter immune function, affecting the response to infections [30]. In a systematic review and meta-analysis of 16 observational studies in 7 countries (Denmark, France, Israel, South korea, Spain, United Kingdom and United States ), mental health disorders were associated with increased COVID-19-related mortality [24].

According to the WHO official estimates of 2020, in the UAE, the age-standardized suicide mortality rate is 5.24 per 100,000 people [31]. This was similar to Saudi Arabia’s estimates of 5.43 per 100,000 people [32]. People with severe mental health conditions can die prematurely as much as two decades earlier due to their preventable physical conditions. In the UAE, the percentage of disability-adjusted life years lost (DALY) due to mental disorders in 2019 was 9% [23]. Additional studies are needed to clarify the risk factors contributing to this higher mortality rate. The UAE healthcare system is among the most resourceful and accessible healthcare systems; therefore, factors related to this system may include health-seeking behavior, associated physical conditions, and patient compliance. This inquiry is crucial for planning healthcare services for this population.

Research on mental health illnesses and mortality in the UAE is very limited. No studies have investigated the odds or relative risk of death in patients with mental illnesses or the causes of death in patients with mental health illnesses. In our collected data, we did not investigate the exact causes of death in patients with psychiatric illness, but this needs to be done in the future.

Strengths and limitations

Limitations

One limitation of this study is the lack of a baseline mental condition history at the time when the cohort subjects did the screening. Therefore, the incidence could not be determined. Another limitation is that Patients with milder psychiatric disorders who may not have been screened may underestimate the prevalence. Nevertheless, this study reports diagnosed conditions, which could be the tip of the iceberg. Only through prospective cohort studies can incidence and prevalence be accurately determined.

Additionally, the progression of psychiatric illnesses is important to study. Still, this study captures the diagnosis of mental health conditions during the follow-up period, and it does not study the natural progression of such conditions. This would be an excellent objective for a follow-up study in this cohort.

Strengths

This is a relatively large retrospective cohort with a long follow-up period. There is a high level of accuracy because it is an Electronic medical record-based study represented by more severe conditions that affected the patients and required medications.

Implications

This region’s unique mental health conditions and risk factors, with the cultural and social impact on their expression, make this study’s results a source for new important knowledge that is informative internationally and regionally, the Arab and Middle East. This is due to the rarity of long-duration cohort studies related to mental health in this region. Therefore, there is a need for well-designed studies to investigate further associations between mental health and other variables/conditions. As well as to study nonclinical data, such as social determinants of health and their association with mental disorders.

Investigating the impact of other factors on the prevalence of mental health disorders in our region (stigma, etc.) is another important area, as stigma can delay seeking medical care, resulting in a worse prognosis. Studying treatment responses, co-morbidities, and the mortality risk in patients with psychiatric illnesses in our population to estimate more accurate suicide rates is a key aim in future studies.

Conclusion

The prevalence of diagnosed and treated mental health disorders was not as high as that in other regions of the world, but when adjusted for age, it was close to the average worldwide prevalence. The discrepancy between questionnaire-based prevalence and our prevalence highlights the mental health burden that necessitates further studies. Furthermore, more studies are needed to estimate the prevalence in the total population of the UAE. The higher risk of mortality found in this study indicates that mental illness is either a risk factor for death or an outcome of higher-risk patients, which also needs to be confirmed by future studies.