Depression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen province

1 In Viet Nam, the updated data of WHOreported the prevalence of depressive disorders of 5.73%, and the rate ofsuicide in 2015 was 5.87 per 100,000 population.2 Without proper treated,de Depression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen provinceDepression-related factors and the effectiveness of community-based intervention with Stepped Care Model in Thai Nguyen province

The overview of depression 3

Definition of depression 3

The concept and classification of depression has evolved and changed significantly from the 18 th century 17 The term depression is extremely broad, variably defining an affect, mood states, disorders, or syndromes.

A depressive affect usually occurs in response to a specific situation and is defined as a transient feeling ‘depressed’, ‘sad’, or ‘blue’.

At a higher level, a depressive mood is more pervasive, more likely to be experienced as negative ideation, and may influence behavior It generally lasts minutes to days in non-clinical situations.

At the highest level, a depressive episode is generally distinguished by a longer duration (usually minimum duration of 2 weeks), more clinical and pathological features, and remarkable social impairment Additional clinical features about severity and sub-typing, especially social impairment criterion, cleaves ‘normal’ mood states from clinical depressive conditions.

In this study, we used the terms “depression” and “depressive disorders” with the same meaning that refers to the individuals with significant depressive symptoms which can range from a large spanning subclinical depression and adjustment disorders, to clinical depression To reach larger populations, the identification and intervention of depression focused on the long-term impact on depressive symptoms, rather than diagnosed depressive disorders This approach is considered as suitable for community intervention with limited resources and less burden for specialist levels in general 7,18

Some epidemiological factors of depression 4

Firstly, the prevalence of depression varies depending on studies with different classification methods This proportion ranges from 17% for diagnosis interviews to 31% based on screening tools, and 22% for combinations 19 The report from 30 countries found that utilizing self- reporting tools, studies in women, and nations with a medium human development index had considerably higher prevalence 20 Generally, a recent systematic review of depression prevalence represents a predominant upward trend within populations worldwide 21

Secondly, the depression rate depends on the demographic characteristics of the study population The prevalence in women is generally 1.5 to 3 times higher than that of men, and this figure of the 18-29-year-old group three times higher than that of people over 60 22 This figure may be higher in specific occupation with high work pressure such as doctors, nurses and teachers 23–25 Among 55 industries, the prevalence of clinical depression ranged from 6.9 to 16.2%, especially with high levels of stress, low levels of physical exercise, frequent or challenging interactions 26 A review in China observed a higher prevalence of depression in rural areas, 27 while a study in Canada noticed a higher prevalence of 18% in urban areas 28 A review detected that higher prevalence of depression in specific developing regions is affected by age, regional and ecological factors 29 A systematic review of depression in Africa and the Middle East found a predominant prevalence ranging from 4% to 53% with common risk factors including women, old, poverty, and chronic disease 30

Thirdly, a review of epidemiological studies demonstrated that estimates of depression's lifetime prevalence vary by nations 31 In UnitedKingdom (UK), the study indicated the point prevalence rate of 2.3% for depression 32 The prevalence of depression in China was estimated to be 1.6% currently, 2.3% over the course of a year, and 3.3% throughout the course of lifetime 27 The community surveys using the WHO Composite International Diagnostic Interview in 10 countries showed that the prevalence of depression in Japan is 3%, Turkey 6.3%, Czech Republic 7.8%, Mexico 8.1%, Canada 8.3%, Chile 9%, Brazil 12.6%, Netherlands 15.7%, and America 16.9% 33

In Vietnam, the rates of depression in the community differ depending on the time and locality studied Bui The Khanh (2001) investigated clinical epidemiology of some common mental illnesses in a ward in Buon Ma Thuot city and found that the depression rate was 2.1% 34 Tran Viet Nghi (2002) studied the clinical epidemiology of depressive disorders in some community populations in Thai Nguyen, found that the rate of depression in one commune was 8.35% and in one ward was 4.2% 35 Tran Huu Binh (2007) studied depressive disorders in Le Dai Hanh ward of Hanoi found that the rate of depression was 4.18% 36 Nguyen Thanh Cao (2012) found that the rate of depression in a ward in Bac Can town was 4.3% 37 The updated data of WHO reported the prevalence of depressive disorders along with 5.73% in Vietnam, and the rate of suicide in 2015 was 5.87 per 100,000 population 2

Recently, there are many studies in Vietnam about depression and related factors, especially during and after the COVID-19 pandemic A study among Vietnamese youth from 15 to 24 years old revealed the depression rate of 10% 38 A meta-analysis exhibited the pooled prevalence of depression was14.64% during the pandemic, especially in health care workers 39 Using thePatient Health Questionnaire-9 to screen for depressive symptoms among university students in Vietnam, it found that 46% of 302 students had depressive symptoms, particularly in individuals with low physical activity 40 The likelihood of depressive symptoms was proven to be higher in poor people, specifically in men and major ethnic groups 41 A study in 1,085 men in rural areas showed the depressive symptom rate of 6.39%, and related factors included older age, educational level less than high school, and low financial status 42 In general, the burden of depressive disorders is increasing in Vietnam which need more attention and effective management.

According to the WHO, by 2030 depression will become the third leading cause of the global burden of disease 1 The consequences of depression can be long-lasting or recurrent and can dramatically impact at all levels.

At an individual level, depression is related to various negative consequences in both physical and mental health Depression can lead to higher risk of various physical diseases, leading excess mortality with the overall relative risk of 1.52 43,44 Moreover, the comorbid physical conditions have poorer prognosis 45 According to the Global Burden of Disease Study, depression were the leading causes of disability-adjusted life years, with significantly increasing number of all-age years of healthy life lost due to disability during 2007-2017 46 Severe depression can lead to 20-fold higher risk of suicide, and the mortality was 1.8 times higher 47 The population attributable risk of depression was 12.7% and 4.8% for all-cause deaths and suicides respectively 48 It also account for 10% of all-cause death 49

At socio-economical level, depressive people are demonstrated to be less often married, more often unemployed and incapable in daily functioning.

Depression without proper treatment results in reduced productivity, and is estimated to lost 50 million years of life expectancy each year, causing a loss of about $925 billion 50 In UK, the cost is significant at £17 billion in lost output and direct health care costs to the economy annually, and a £9 billion impact on benefit payments and lost tax receipts 51

Recently, the burden of depression is exacerbated by the Corona Virus Disease of 2019 (COVID-19) pandemic, with the average proportion up to more than 34-36% in China and nearly 30% in other countries 52 A recent large- scale study notified a significant increase in depression prevalence globally, especially in women, healthcare workers, and COVID-19-infected patients 53 Basically, the depressive symptoms can be identified until more than 12 weeks following this infection with the frequency of about 28% 54 In sum, this increasing prevalence is attributed to the enhanced awareness of population, more access to mental health care, and more common diagnostic criteria and tools, not merely increase in incidence 55

In Vietnam, a study reviewed the burden of depression from 1990 to2019 demonstrated that depressive disorders comprised 2,629,000 estimated cases and 380,000 estimated DALYs, in particular among women 56 Although notable change in mental health services in recent years, the disease burden has been high and lack of integrating and promoting equity 57 Along with other common mental illness, depression risk was associated with social factors such as disadvantaged familial characteristics, low socio-economic status 58 These evidence suggest that a community-based strategy aimed at lowering home risk factors and offering helpful assistance could be a useful tactic to reduce the depression burden in Vietnam.

Etiology of depression 7

Depression is considered as multifactorial disorders with multiple etiological pathways 59,60 Therefore, it is of necessity to explore the depressive disorders with a biopsychosocial model instead of the biomedical model 3

● Biogenic amine hypothesis (monoamine theory)

The biogenic amine dysfunctions are traditional, evidence-based hypotheses with 3 major monoamine systems including serotonin, dopamine, norepinephrine It is the basis of using medication in depression treatment with different mechanisms 3,61 Firstly, serotonin controlling the affects, aggression, sleep, and appetite, is proved to contribute to the pathogenesis of depression 22 The serotonergic dysfunction range from reduced concentration in central nervous system (CNS) to decreased specific receptors, as well as lower number of transporter binding sites in the midbrain and amygdala 59,62 Secondly, dopamine is known as a important catecholamine for drive, pleasure, sex, and psychomotor activity 17 The dopaminergic alterations are supported by the high comorbidity of depression with Parkinson’s disease, and the dopaminergic properties of some antidepressants 3,17,59 Thirdly, the noradrenergic dysregulation is one of the main mechanisms of depression combining with anxiety, including low concentration of norepinephrine metabolites, increased β-adrenergic receptors in the CNS, and the stress response 59 Moreover, some research proposed the role of glutamate γ- aminobutyric acid, brain-derived neurotrophic factors, thyrotropin-releasing hormone, corticotropin - releasing factor, acetylcholinergic neurons in the etiology of depression 17,62–64 In general, it involves more complicated dysregulation than the single neurotransmitter hypothesis predicts.

● Neuroendocrine causes and stress responses

The most widely studied endocrine disturbance in depression is the dysfunction of the HPA (hypothalamic-pituitary-adrenal) axis, and increased concentrations of corticotropin-releasing factor in cerebrospinal fluid of depressive patients 3,17,59 According to the modern stress diathesis hypothesis, the excessive secretion of cortisol and associated hormones are critical The hyperactivity and impaired feedback regulation of the HPA axis; decreased corticotropin-releasing factor’s receptors were seen in the frontal cortex 3,59 In addition, the possible roles of growth hormone and thyroid-stimulating hormone was suggested in depression 17 It is reasonable to emphasize the role of hormones with evidence of higher depression rate in females, after using hormonal contraception among adolescents, and postpartum depression 3,65

The scientific findings pointed out one-third of depressed patients have elevated inflammatory biomarkers, even without medical conditions 66 Many crucial substrates in the depression aetio-pathogenesis such as monoamine neurotransmission, glucocorticoid receptor resistance, and hippocampal neurogenesis, can be affected by inflammatory mediators 3,17 Moreover, increased inflammatory cytokines relates to malfunction of the endocrine system, HPA axis, and glutamatergic activation, contributing to depression 59

Recent new imaging techniques identified related specific brain regions and pathways in the depression etiology On the one hand, the altered anatomy of the depressed brain is seen in the abnormalities of limbic system- cerebrocortical pathways The computed tomography scans represented enlarged pituitary and adrenal glands in depressive patients 17 Besides, the magnetic resonance images revealed the reduced size of hippocampus, caudate nucleus, and frontal cortical areas’s activities, and the increased activities of amygdala and other limbic sites, which is proved to be recovered by the treatment 59 On the other hand, the neurophysiology abnormalities in depression was proved with the neuronal hyper-excitability 17 The reduced neurogenesis and neuroplasticity especially in hippocampus were noticed 67

For a long time, the researchers have found that a large proportion of depressive cases run in families 17,55 The clinical evidence indicated that the heritability of depression is approximately 40% 3 The research has shown the significant roles of various polymorphisms in the genes for serotonin transporter, monoamine oxidase, brain derived neurotrophic factor, endogenous ꭒ-opioid receptor, and circadian clock gene in depression etiology 64 It is similar to genes of norepinephrine transporter, purinergic receptor 68–70 In general, the genetic heterogeneity might involve single or multiple gene families or subtypes with complicated mechanism 17

● The microbiota hypothesis and other biomarkers

Recently, the gut-brain axis and the gut microbiome was investigated in the etiology of depression 64,71,72 The altered microbiota was identified in depressive people with specific species 73,74 The vagus nerve, gut hormone signaling, immune system, tryptophan metabolism, and microbial metabolites involved in the microbiota-brain communication 75 Other biomarkers in the etiology of depression such as endocannabinoids, 76,77 neurotrophic factors, 78,79 polyunsaturated fatty acids, 80 hormones, 65,81 telomere length, 82 and vitamin D, 83 were suggested, however these association are inconsistent 84,85

There are several theories that may explain how psychological factors contribute to the development of depression such as psychodynamic theory, cognitive and behavioral theory, and attachment theory.

Firstly, psychodynamic theory which is Sigmund Freud’s propose suggests that depression is caused by unresolved conflicts or repressed emotions 17,86 While some evidence supporting psychodynamic theory and therapy for depression is considered not strong, some studies have shown that it can be effective in reducing depressive symptoms 87

Secondly, the cognitive theory revealed the cognitive triad about the self, the world, and the future, suggesting underlying negative and distorted thoughts, such as beliefs of hopelessness, helplessness, and worthlessness 88 People with depression are more likely to engage in the negative thinking patterns, such as all-or-nothing thinking and overgeneralization, triggering a downward spiral of negative emotions and depression 89

Thirdly, the behavioral theory suggests that depression is caused by a lack of positive reinforcement or experiences in life 17,90 It can cause a sense of hopelessness or helplessness, which can lead to negative thoughts, feelings, and behaviors Studies have shown that depressed people experience less pleasure from activities that they used to enjoy, and lack of social support, which enhance loneliness and isolation 91,92 This theory is the basis for applying the behavioral activation (BA) therapy to treat depression.

Last but not least, the attachment theory suggests an individual who experienced neglect or abuse in childhood may struggle with self-worth and relationships in adulthood A meta-analysis revealed that the insecure attachment is a predictable factors of depression’s emergence in youngsters 93

About 30% to 40% of depression risk comes from heritable factors, and the remaining contribution is attributed to socio-environmental factors 59

Basically, the early adversity and childhood maltreatment are of paramount importance in depression 3,17 Sexual, emotional, or physical abuse, a dysfunctional upbringing, parental separation or mental illness are the most frequent childhood challenges 55 At critical time points, the brain is more sensitive and responsive to socio-environmental factors, leading to abnormal neural activities 3 Not only does it have a negative impact on the psychological development, these early exposure to stress relates to cortical circuits and HPA axis, contributing to the emergence of depressive symptoms.

In addition, studies suggested the significant role of adverse life events and major life transitions in the etiology of depression, such as the death of a loved one, divorce, job loss, entering a new stage of life or a serious illness 3,94 The greater the number and severity of these events, the higher the risk of depression Social stressors are related to the onset of depression, and likely triggered by preceding temperamental instability 3,17

Furthermore, the quality and availability of social support influence the depression risk The individuals with stronger social networks are generally less vulnerable, while those with limited social support may be at greater risk.

The meta-analysis emphasized the moderately significant effect of loneliness on depression, particularly among patients, caregivers, students, and the elderly 95 Last but not least, studies proposed other sociocultural and environmental factors in the depression etiology 3,59,85,96 A study across 195 countries revealed potential correlates of depression, especially cultural individualism 97 In general, the relationship between socio-environmental stressors, the body's and mind's response in depression’s etiology is complex 55,94,98

Diagnostic criteria for depression 12

The diagnostic criteria of depression changed over time 3,88 Currently, two parallel classification systems are widely used in mental health practice including the Diagnostic and Statistical Manual of Mental Disorders (DSM) and the International Statistical Classification of Diseases and Related Health Problems (ICD) 17,22,60 The ICD-10 was developed by WHO in 1992 and used predominantly in Europe, whereas the DSM-V was composed by the American Psychiatric Association (APA) in 2013 and common in United States (US) These diagnostic criteria are listed in the Appendix of this thesis.

In both systems, the two core criteria including depressed mood(persistent feelings of sadness) and/or anhedonia (a lack of desire to engage in formerly pleasurable activities) In general, the diagnosis of depression is based not only on its severity but also on persistence, the presence of other symptoms, and the level of functional and social impairment However, there are some differences between the two diagnostic systems According to DSM-V, depression diagnosis requires the presence of at least one of the two core criteria and at least five of nine criteria in total The diagnostic code is based on whether this is a single or recurrent episode, current severity, presence of psychotic features, and remission status Most depression screening and assessment scales in use are based on DSM diagnostic criteria.

In Vietnam, the psychiatric systems mostly use the diagnostic criteria in ICD-10 to diagnose depression with three most typical symptoms and seven common symptoms The differentiation between mild, moderate, and severe depressive episodes depends on a complicated clinical judgment that involves the number, type, and severity of symptoms present Minimum duration of the whole episode is about 2 weeks, except particularly severe with rapid onset, it may be justified to make this diagnosis after less than 2 weeks Moreover, some of the above symptoms may be marked and develop special clinical characteristics called “somatic symptoms” Somatic syndrome was diagnosed with at least four of eight symptoms Recently, the ICD-11, the eleventh revision of the ICD, has been released from 2018, and officially endorsed by WHO in 2019 with some modifications One of the most significant was the introduction of cluster coding, which rebuilt the coding system and enabled ICD to be used digitally with the latest data 99

In general, no matter what diagnostic criteria are applied, the diagnosis of depression in the community still faces many common difficulties Firstly,it is especially difficult in the case of somatic depression with predominant somatic symptoms rather than psychological performance 100–102 Secondly,there are still myths about depressive disorders as a manifestation of weakness, the toxicity of antidepressants, or lifelong treatment merely with sedatives 6,103 These unfounded rumors may prevent people with depression from accessing proper treatment services.

Treatment of depression 13

The development of scientific treatments of depression is based on a comprehensive understanding of the biological, psychological and social etiology of depression In general, the treatment of depressive disorders consists of three main types: (1) medications; (2) evidence-based psychotherapy, and (3) somatic or socio-environmental intervention 17,104

Thanks to the state-of-art development of pharmacotherapy, there are a lot of effective medications for depression Basically, the therapeutic effects are caused by the impact on the neurotransmission and neurotransmitters 17,104 The physicians need to consider available options depending on the clinical symptoms, side effects of the drug, presence of medical illness, historical response, and financial costs, and predict the ability the adherence 55,88

On the one hand, antidepressants are often the first line of pharmacological treatment for depression based on the monoamine hypothesis 61 Traditional antidepressants such as monoamine oxidase inhibitors or tricyclic antidepressants are believed to regulate the mood effectively Recently there are many types of antidepressants with fewer side effects and targeted mechanisms such as selective serotonin reuptake inhibitors and serotonin–norepinephrine reuptake inhibitors Most antidepressants take time to reach the optimal effect The initial antidepressant might work for 50% to 65%, and remission can be achieved in 30-40% of cases 55 The proper pharmacotherapy in depression is effective to calm the biological abnormality, and normalize epigenetic alterations induced by psychological trauma 3,59 However, the patients need to be monitored closely for the risk of worsening depressive symptoms or suicidality, particularly at the beginning of pharmacotherapy 88

On the other hand, when a depressive patient has inadequate response to first-line treatment, the clinicians can consider following options: (1) increasing the dose until tolerance; (2) switching to a different antidepressant;

(3) adjunctive therapy with another agent such as typical antipsychotic,lithium, triiodothyronine or a second antidepressant with a different neurotransmitter action 61,88 However, there are several barriers in the pharmacotherapy of depression Firstly, the depressive symptoms, response and progression may vary between individuals, so there is no best optimal medication for all patients 55 Secondly, the non-adherence to medicine is a big issue for up to half of depressed patients, which is proved to link to a higher risk of relapse and hospitalization 3 Thirdly, pharmacotherapy has mainly been symptom- focused to limit the acute episodes of the illness.

Therefore, the pharmacotherapy is recommended to combine with the psychotherapy for better long-term outcomes 17 The medication may complement and synergize with psychotherapy in the treatment of depressive disorders To be specific, antidepressants may act on the amygdala, which is involved in the processing of negative feelings, whereas cognitive behavioral therapy (CBT) can improve the prefrontal function and thus cognitive control, which both lead to a more sustainable outcome of depression intervention 59

One of the mainstays of treatment for mood disorders has been psychotherapy in the form of individual or group format with different time- limited and manual-based methods Different types of psychotherapy are proven to be efficacious for depression including CBT, BA, Interpersonal Psychotherapy (IPT), and Problem-Solving Therapy (PST) 105

Firstly, the most traditional psychotherapy is the psychoanalytic and psychodynamic therapies which arose from psychodynamic theories depicting depression as a result of internalized anger and unconscious conflict It is usually time-consuming and focuses on identification and interpretation of past experiences The common techniques are free association and projection, which help to provide the insights of internal problems 17,106

Secondly, CBT is a type of psychotherapy that combines elements of cognitive and behavioral therapy In terms of cognition, it is critical to identify and challenge negative thoughts and beliefs, then replace them with more balanced and realistic ones Regarding the behavior, the negative emotion could be understood as the result of reinforcement withdrawal such as decreased engagement in activities that could be enjoyable or rewarding It emphasizes the role of homework assignments to improve personal mastery, confidence and experience in real-life situations 89,107,108 However, a typical CBT psychotherapy duration is around 6 to 12 sessions of 45-60 minutes therapy, which makes it sometimes difficult to apply broadly.

Besides, BA is an evidence-based approach that is based on behavioral theory and has been shown to be effective in treating depression BA focuses on increasing positive reinforcement by helping individuals identify and engage in pleasurable or rewarding activities The idea is that by increasing positive experiences, individuals can counteract the negative thoughts and behaviors that contribute to depression Studies have found that BA is as effective as other evidence-based treatments for depression, such as cognitive- behavioral therapy and medication 88,92,109 Compared to CBT, BA is a less structured and more flexible therapy, and can be delivered in more formats.

Thirdly, the 1980s saw the emergence of IPT, established by Klerman, Weissman, and associates IPT incorporates many of the more practical techniques used by social workers in case management to create a time- limited therapy that addresses the typical problematic patterns in relationships The interventions used in IPT are fairly eclectic and can include psychoeducation, nondirective questioning and empathic support, role playing and social problem solving, in addition to identifying the interpersonal difficulty areas that are most relevant to a specific patient 88,106

Last but not least, there are some other psychological interventions which may be useful in depression intervention, such as mindfulness-based interventions which highlight main skills including observing perceived sensations, accepting, and experiencing nonjudgmentally 17 The acceptance and commitment therapy emphasis on committing to living by one's core principles The dialectical behavior therapy focuses on mindfulness and acceptance and has been shown to reduce suicidal thoughts and behaviors 17,88

Overall, psychological treatments play an important role in the primary prevention and treatment for depression 110 When compared to antidepressant, psychotherapy increased the probability of attaining remission in individuals with an early life adverse experience by double 59 For patients with mild to moderate depression, the efficacy of psychological treatments is same as the pharmacotherapy 111 There may be advantages of psychotherapy over 23 pharmacological treatments in terms of relapse prevention 112,113

Brain stimulation is a psychiatric practice using electric current, or changing magnetic field to induce currents and alter neuronal firing with specific devices The common transcranial approaches are transcranial direct current stimulation (DCS), transcranial magnetic stimulation (TMS), and electroconvulsive therapy (ECT) with different side effects and contraindications ECT is an oldest and standard somatic psychiatric treatment in which electricity is delivered through the scalp to induce a therapeutic seizure based on broad mechanisms of action It is considered to be effective in refractory patients with response rate of 70% 17 With safer characteristics,TMS is referred to as electrical stimulation without an electrode and anesthesia The repetitive TMS is proved to be effective in at least 40% of depressed patients, especially in individuals with less severe symptoms and shorter episodes of depression 114 In general, the Food and DrugAdministration approved ECT and TMS for depression treatment with specific severe cases In addition, deep brain stimulation, vagus nerve stimulation, and implanted cortical stimulators are common surgical methods to treat depression However, the effectiveness and application of these techniques still need to be further studied and discussed 88

1.1.5.4 Social and community-based intervention

Despite the emergence of various evidence-based treatment methods in psychiatric practice, most patients with depression have a fair prognosis for recovery Three out of four depressed people have recurrences throughout their lives, with varied degrees of persistent symptoms between episodes 17 As a result, the modern psychiatry has brought about the refocusing of the lenses of psychiatric treatment from institutionalization to socialization 62 Even with people with significant impact of childhood maltreatment, depression can be mitigated by supportive environmental interventions 59 Therefore, the social and community interventions are considered as promising for depression intervention, especially when used in conjunction with other treatments.

Nowadays, the common social and community interventions for depression include support groups, family therapy, community programs, and online interventions These activities help depressive individuals connect with others, improve communication, reduce isolation and improve mood 115,116 Also, this can provide a sense of community, validation, and social support.

Depression-related factors

Biological factors 21

There is a lot of literature that emphasizes the higher prevalence of depression in women compared to men, the ratio of women to men is approximately 2:1 3,17,22 According to DSM-V, prevalence of this disease in women is 1.5 to 3 times higher than that of men in early adulthood 22 The study suggested gender as a related factors for depressive disorder 62,134 Although there have been many changes in women’s social positions over time, a recent systematic review revealed that the gender gap in depression has not significantly decreased among adults 135

To be specific, the higher risk of depression in women is influenced by a myriad of factors Women usually have more hazardous thyroid activities,unstable sexual hormones levels, which contributes to depression, especially in postpartum, premenstrual and menopause periods 17,65 Furthermore, women may have higher psychological susceptibility for depression including neuroticism, rumination, body shame and dissatisfaction 136 Also, the researchers found that big rapid social life transitions like childbirth, menopause, retirement, empty-nest transition, and midlife crises might trigger depression in women more severely 137 The meta-analyses demonstrated that there is a considerable negative correlation between depression and femininity, whereas masculinity acts as a protective factor for depression 138 The prevalence of depression was noticed to be higher in lesbian, gay and bisexual population compared to the heterosexual population 139–141

Although depressive disorders can appear at any age, the likelihood of the onset rises significantly during puberty In US, the peak incidence of depression is studied to be in the 20s 22 According to the APA, the 12-month prevalence of depressive disorders of age group from 18 to 29 is 3 times higher than that of the group aged 60 and over 142 Moreover, the literature indicating that the depression rate reaches a climax around age 13-15, and the distribution of onset age of major depression is bimodal with two points of thirties and fifties 3 Sadock suggested that the average age of onset of depression is about 40 years, with 50% of patients having an onset of between 20 and 50 years old 143 In the community, Williams found that typical depression is less common in older people than in young people 144

These high-risk age groups may be due to multiple mechanisms The dysregulated growth hormone and early puberty are demonstrated to be risk factors of depression, particularly in the childhood and adolescence 3 The puberty period marks lots of neurobiological and social changes in adolescents, from sex characteristics and body changes to interpersonal conflicts among peers In addition, the up-to-date contemporary life may cause various negative social interactions such as virtual harassment, bullying or isolation, leading to the development of depression in the youngster 135

1.2.1.3 Cortisol (HPA axis function) and neuroendocrine factors

There is evidence showing that HPA axis disorders are present in 50% to 75% of depressed cases, both anatomically and functionally 67 According to the modern stress diathesis hypothesis, the excessive secretion of cortisol and associated hormones are critical factors In particular, research showed the hyperactivity and impaired feedback regulation of the HPA axis; decreased corticotropin-releasing factor’s receptors in the frontal cortex 3,59 Moreover, the systematic review and meta-analysis of 75 prospective articles identified cortisol as the only biomarker that can be a potential predictor of depressive disorders’ onset, relapse or recurrence 84

1.2.1.4 CNS condition, the neuroamine and neurophysiological factors

The relationship between CNS condition and depression is proven in cases of Parkinson’s disease, post-stroke depression with higher rate of depression 3 Furthermore, the frontal lobe hypoactivation is proved to be a risk factor of depression in the period of infancy 3,145 Moreover, the biogenic amine dysfunctions which are consequences of CNS conditions may contribute to depression pathophysiology 146 The reduced serotonin availability is related to the rapid relapse of depressive symptoms 59 The monoamine transmitter is considered as an attribute to the higher risk of depression among females It can be explained that women have a greater level of monoamine oxidase enzyme compared to men 17

1.2.1.5 Physical condition (non-CNS condition) and the inflammatory factors

The depression-related physical factors are explained by the inflammatory theory The meta-analysis of 99 studies showed that elevated interleukin-6 was significantly related to the depression 147 A systematic review and meta-analysis of 73 studies noticed the relating abnormal blood chemokines, suggesting the potential implication of identifying depression based on inflammatory biomarker profile 148 An updated review highlighted the association between depressive symptoms and different immunological alterations, suggesting that inflammation is an important disease modifier 149 The inflammation was considered as a precipitating element that leads to depression, as well as a perpetuating factor that makes recovery difficult 66 Moreover, the influence of physical condition towards depression can be clearly seen in people with chronic diseases with biological and financial impairments The metabolically health status, and having at least 4 metabolic risk factors are proved to be risk factors for depression 4 In particular, unhealthy obesity was linked to a 30–83% higher risk of depression, while metabolically unhealthy non-obesity was linked to a 19-60% higher risk 150

1.2.1.6 Family history and the genetic factors

The genetic studies showed that depression risk is two to four times higher in first-degree relatives of those with the condition than in the general population Early onset and recurrent depression seems to have greater relative hazards, and neuroticism is a personality attribute that contributes significantly to this heritability 22 The latest genome-wide association meta- analysis also pointed out many promising genetic risk factors and architecture of depressive disorders in different populations 151,152

1.2.1.7 Gut microbiome and other biomarkers

The gut brain axis dysfunction is proved to relate to the metabolic and appetite disturbances, and functional gastrointestinal symptoms 72 The gut microbiota may regulate brain activities such as the HPA axis, immune system, and neurogenesis 71 Additionally, the evidence of mitochondrial dysregulation, nutritional factors, gut permeability and neuroprogression in the depression’s chronicity and treatment resistance suggest promising dietary modifications and microbiome for depressed people 55

Psychological factors 25

Depression is attribute to various psychological factors such as personality, emotional resilience, early trauma, and cognitive styles.

The different personality traits can differentially relate to depressive mood 153 The low self-esteem is more likely to cause an individual to sink into depression in a difficult environment 17 On the contrary, people with high self- esteem have been shown to have a lower possibility for developing depression, regardless of whether or not having narcissistic self- enhancement 154 Other common traits relating to depression are neuroticism, anxious, impulsive and obsessional, especially low extraversion and conscientiousness, and high neuroticism 3,155 A systematic review found that the temperament characteristics of high harm avoidance (anticipated fear of challenging events) and low self-directedness (the adaptive capacity to achieve personal goals) seem to be reliable indicators of an individual's susceptibility to depression, even predicting response to treatment 153

1.2.2.2 The early trauma and adverse childhood experiences

The literature showed that early adversity and childhood maltreatment are of paramount importance in depression, especially in vulnerable people 3,17 Childhood maltreatment includes separation, physical neglect or abuse,emotional or sexual abuse, and witnessing domestic violence, all occur before the age of 11 and lasting for a minimum 6 months 85 A meta-analysis about the role of childhood trauma towards adult depression emphasized that the neglect and emotional abuse are the strongest childhood risk factors 156 Moreover, these early maltreatments can be risk factors for depression with more severe, early- onset and treatment-resistant characteristics 157 The premature life adversities are considered to relate to low response of depression treatment, higher risk of relapse and more prolonged course 59

1.2.2.3 The emotional resilience and cognitive styles

The temperamental instability is proved to be a risk factor for depressive disorders 17 The unstable emotion which typically precedes clinical depressive episodes, appears to be a trigger for stressful events The thinking of self as helpless, interpreting life events negatively, and believing the future to be hopeless are common characteristics of depressive people, which can lead to misinterpretations of daily circumstances 88,143 The past experience of social powerlessness contributes to depressive symptoms through low capacity and inadequate good reinforcement 158 It is known that the positive affectivity and cognitive flexibility are protective factors towards depression.

The better coping skills and high ability to find meaning in difficulties helps individuals to overcome and prevent the relapse of depressive symptoms 59 These theories are the foundation for the new psychotherapeutic approaches such as CBT in depression treatment 3,17

Socio-environmental factors 26

Research shows that various socio-cultural and environmental factors related to depression from family to socio-environmental conditions 59,63,98

Firstly, the impact of family factors towards depressive disorders can be observed from childhood Sexual, emotional, or physical abuse, a dysfunctional upbringing, parental separation or mental illness are the most frequent childhood challenges 55 Not only does it have a negative impact on the psychological development of children, these early exposure to stress relates to cortical circuits and the HPA axis Maternal illness, family conflict or parental stress are common risk factors of depression, respectively 3

Secondly, the role of family is seen as the vulnerable adults with negative familial characteristics Marital status is considered to be an important factor relating to the development of depression in adults Being single, divorced, separated or widowed are common risk factors 4,17 Moreover, the systematic review and meta-analysis emphasized the moderate to strong relationship between intimate partner violence and depression in women, including physical and psychological threats in both short term and long term 159

Firstly, the role of social events was acknowledged in depression intervention 3,55 Interpersonal losses, such as grief or ending of relationships, are regular situations, especially people with vulnerable personalities, which can cause psychological trauma, sleep loss, leading to depression 17 The cumulative adverse life stress may precipitate depression, especially in at-risk people 94 The reason is considered as the damaged allostatic load of the body after prolonged exposure to adversities at sensitive periods 3

Secondly, the literature emphasizes the critical role of social support and interaction in depression, however, the frequency is more important than the amount The concurrent life events, resulting lifestyle changes, a lack of social support and skills are proved to be modulating variables of the loss’ impact in the development of depression 3 Other risk factors such as poor sleep, unpleasant social interactions, and stressful negative experiences may precede subjective negative affect In contrast, exercise and good social relationships have been proven to predict later depressive mood reductions 160

Thirdly, the patriarchy, gender-based oppression, vulnerable conditions of socio-economy, race, and disability are common social risk factors 62 People with lower socioeconomic conditions usually have lower levels of education, income, living standards, and higher risk of unemployment 17 The study of

Butterworth and Noori in the Australia and Canada found that people in more favorable socioeconomic regions with higher household incomes had a lower risk of depression 161,162 Additionally, the researcher suggested the varied impact of femininity to depression in the experience of women of color, lesbian women, and women from underdeveloped regions 62 The systematic review about social rank theory also found a negative relations between subjective perception of social position and depression or suicide 158

First and foremost, the working environment can contribute to depression in different ways A meta-analysis showed that job strain is one of the most convincing risk factors for depression across the lifetime 4 In specific, being out of work can have a greater impact for a short period of less than a year, while occupational instability is more likely to have a longer-term effect in youngsters 163 In addition, the long working hours is demonstrated to be strongly associated with the depressive symptoms, particularly in Asian 155 Correspondingly, the overtime work tends to be more significant in women, which is explained by heavier family responsibilities and effects of work- family conflicts 164 In contrast, positive peer groups can serve as evolving environments to shape the characters and prevent depression 3

Furthermore, the living environment factors can be a potential risk factor of depression The polluted environment often lead to the uptake of synthetic chemicals such as preservatives and hormones, and the prolonged exposure to continual noise and electrical pollution, which may cause depression among susceptible individuals 55 The relationship between depression and preventable indicators of the house environment was seen in poor housing quality and the lack of green areas 165 The modernization and globalization relate to depression with unhealthy diet, excessive technology use, and sedentary lifestyle 166 The consumption of poor-quality foods and an unhealthy diet were both linked to poor mental health 167,168

Last but not least, on a larger environmental scale, the seasonal and geographic factors affect the incidence of depression In many countries far from the equator, the seasonal affective disorders accounted for 25% of recurrent depression 17 The natural disasters (such as earthquakes, hurricanes) or man-made catastrophes (such as terrorism, war) are possible contributing factors to depressive symptoms, particularly among vulnerable person 55

Many different interactions between environmental stressors and personal genomes were identified to lead to stable alterations in specific gene structure and expression, and depressive symptoms 59,98 A systematic review identified the stress-associated epigenetic changes in different genes of glucocorticoid, serotonergic, and neurotrophin activities having correlation with depression among stressed patients 152 The diverse effects of combination are mediated by the epigenetic alterations at critical phases of development and single-nucleotide polymorphisms in certain genotypes 17

The community-based intervention with Stepped Care Model for

Several community-based interventions for depression and the origin of

After many years of depression intervention, specialists realized that traditional care systems were inadequate, the access to care was limited.

Despite various treatment methods in psychiatric practice, the treatment gap is noticeably significant in which up to 85% of people with mental problems had not been properly treated 169,170 The WHO reported that even in high-income countries there was up to 35 to 50% of people with depression received no treatment 171 Only one in three patients with depressive disorders received treatment, especially in scarce-resource regions 55 This increasing burden of disease can be explained by the limitation of specific treatments and lack of coordination, especially in LMICs like Vietnam The pharmacotherapy has limited effect on long-term outcomes, and psychotherapy requires time and money to improve the symptoms With the key target of increasing access to depression treatment with fewer resources, particularly in publicly funded healthcare systems like Vietnam, there are a number of community-based interventions for depression which were proven to be promising in the long term According to WHO, the community-based intervention can lower the burden of illness by 10% to 30%, but there are lots of barriers such as delayed diagnosis and treatment, stigmatization, and lack of evidence-based strategies 3,12 In general, depression intervention requires evidence-based interventions with appropriate resources in specific circumstance.

Recently, the community mental health care approaches have been emphasized by WHO and the Action Plan of the World Psychiatric Association, especially in LMICs with the shortage of resources 171,172 A meta- review identified several community interventions including awareness raising, psychoeducation, skills training, rehabilitation, and psychological treatments to increase accessibility and acceptability 8 The community-based psychological therapy was recommended as first line treatment for mild-to- moderate depression, particularly led by non-specialists 9,170 A systematic review concluded that the community interventions led by commune health workers were effective to reduce the depressive symptoms in underserved communities in both US and LMICs 173

In Vietnam, with a rapidly increasing population and changing socioeconomic status, mental health services lag far behind the need and largely focus on medication and severe mental illness in limited institutional settings 6,174 Significant barriers contribute to this treatment gap, including a lack of human resources, significant stigma, and limited knowledge in the general public about depression and mental health 12,175 To deal with this high disease burden, the government has focused more on depression care in the community Vietnam's National Health Target Program in 2002 targeted in lowering the prevalence of depression and suicide 57 In 2011, the government had a mental health plan to focus on community-based mental health care, through the Ministry of Health, Ministry of Labor, Invalids and Social Affairs (Decision 1215/QDTTG, July 22, 2011) However, treatment services for depression are mainly through networks of provincial psychiatric hospitals or psychiatric departments in district general hospitals, mainly focusing on schizophrenia and epilepsy The community services in mental health care has been in place since 1998, in which many primary health care centers provided basic medication for psychiatric disorders However, psychosocial care received less attention with 0.4 psychiatrists per 100,000 people compared to an average of 9.7 psychiatrists per 100,000 in European countries 176 Before this situation, several models of integrated community-based mental health care using task-shifting have been conducted in Vietnam to increase access and availability of depression care 177–179

In sum, the use of brief (‘low-intensity’) versions of psychological therapies, such as self-help and group therapy, had potential to improve access to depression care in the community 107 However, such treatments may be inappropriate and insufficient for some more severe patients who need more specialized treatment Therefore, the low-intensity psychotherapies have to be incorporated within a system for the allocation of different levels of treatment that meet the needs of different cases As a result, Stepped CareModel titrate the provision of low-cost intervention to the need of a larger population 172

The community-based intervention for depression with Stepped Care

Stepped care represents “a system of delivering and monitoring treatments, so that the most effective, least intrusive and least resource intensive treatments are delivered first” 180 In 2000, Haaga introduced a series of papers on the potential for and implementation of stepped care for common mental health disorders including depressive disorders 181 The authors reflected on how to define and implement stepped care as a system of treatment The core principles were that almost all patients start with an evidence-based treatment of low intensity; only patients who require further treatment ‘step up’ to a treatment of higher intensity 182 In practice, low- intensity psychotherapies are often used as a first step in stepped care treatment for depression, pharmacotherapy and high-intensity psychological therapies delivered in a more conventional form can follow.

With the above definition, the Stepped Care Model is appropriate with public health practice with two fundamental characteristics Firstly, the recommended treatment is the least restrictive, which means having optimal cost with the least amount of specialist input required Secondly, the SteppedCare Model should be self-correcting, which means that the interventions and stepping up have to be monitored systematically by the specialists and the health care system The individuals with less severe problems would be more likely appropriate to minimal interventions Basically, psychological sessions are recommended as 8-16 face-to-face sessions with potential efficiency In the context of publicly funded healthcare systems, the involvement of available non-specialists in such primary care is critically important to save the specialist time and resources 14 With above reasons, the Stepped Care model was recommended as one of the most suitable approaches to manage depression in the community with the updated WHO manual 172

Figure 1.2 The Stepped Care Model in the Management of depression.

Source: Depression: The treatment and management of depression in adults, National Collaborating Centre for Mental Health (UK), National Institute for Health and Clinical Excellence: Guidance (2010, updated 2017) 7

One of the most successful examples of the Stepped Care model being applied, researched, and developed into standard guidelines is the Improving Access to Psychological Therapies (IAPT) program in the UK This National Institute for Clinical Excellence (NICE) guideline for depression intervention recommended the stepped-care model as a framework to organize the delivery of services, and aid patients and practitioners in locating and gaining access to the most effective interventions (Figure 1.2) This model enables the least intrusive, most effective therapy provided for the largest population in need.

If a person does not benefit from the initial intervention or refuses an intervention, they would be offered another appropriate intervention in the following step This national guideline of the UK also emphasizes the important role of primary healthcare staff in the recognition, assessment,offering of basic intervention, and referral to higher levels of treatment There is only about 20% of depressed patients need to be referred to specialists with the main indications including failure of treatment in primary care, inadequate response, frequent episodes, having suicidal or psychotic symptoms At the highest level, only a few patients need inpatient treatment with 24-hour care and special interventions with specialists 7

To be specific, step 1 focuses on case identification, risk assessment, and monitoring by the trained non-specialists and validated measures In step 2, all recognized depression from mild to moderate severity are offered psychoeducation, low-intensity psychosocial interventions such as individual or group structured CBT, or drug treatment by trained practitioners In other words, the two first steps are mainly relied on non-specialists, the specialists can participate in training and supervise the progress in the minimal intervention if possible The patients who are identified to have severe symptoms or incomplete responses are referred to steps 3 and 4 for high- intensity psychological and/or pharmacological treatment by specialists.

Specifically, the patient's progress needs to be assessed to inform the next treatment step by validated symptom checklists; and the decision to end or continue treatment after low-intensity therapy is made using set criteria appraised at a pre-specified time interval To determine which patients ‘step up’, response to treatment is systematically evaluated 10 Its core principles of delivering low-burden treatments first, followed by careful patient progress monitoring to step patients up to more intensive treatment, lead to considerable implementation diversity.

In general, Stepped Care Model is described within the context of strategies such as collaborative care and task shifting that aim to increase access to health care through the improved coordination of care between primary and specialist services 13 It requires that all patients start with an evidence-based therapy of low-intensity; progress is monitored systematically and patients who do not respond adequately go on to receive high-intensity treatment 182 In fact, the Stepped Care Model has been used to manage some mental disorders such as eating disorders, drug problems, and panic disorders with significant outcomes 183–185 In a similar way, this model has been applied to depression intervention in the community worldwide In Netherlands, a Stepped Care program with five steps led by general practitioner under the supervision of specialists (psychiatrist, psychotherapist) helped to decrease the referrals in depression management 186 Depending on available resources, WHO recommends Stepped Care model can be started with simple interventions delivered by supervised non-specialists 172

In Vietnam, there were several community-based approaches applied to stepped care principles and the task-shifting for depression intervention Some programs implemented by Basic Needs Vietnam with the partnership of manyProvincial Psychiatric Hospitals, Vanderbilt University, and Research andDevelopment corporation, such as the Multicomponent Collaborative CareModel for Depression (MCCD), Vietnam Collaborative Care for DepressionProgram (CCDP), Livelihood Integration for Effective DepressionManagement (LIFE-DM), and Mental Health and Community Development(MHD) In particular, the MCCD project was conducted with the partnership of Da Nang and Khanh Hoa Psychiatric Hospital, Vanderbilt University andResearch and Development corporation 187 The trained commune health station (CHS) personnel took charge of screening, providing psycho- education, delivering guideline antidepressants, and/or 5-session individual psychotherapy under supervision This approach was considered to reduce stigma and pressure on the specialized health system, increase care access and the capacity of primary health care 188

The effectiveness of the Stepped Care Model for depression

With the simple core principles, many clinical guidelines worldwide endorse stepped care for depression such as National Institute for Health and Clinical Excellence and Dutch multidisciplinary guideline 7,15,16 With the potential to improve access to depression treatment, several authors have recommended that stepped care should be widely adopted 189,190 The treatment guidelines clearly recommend stratified stepped care involving low-intensity psychological therapy for patients with mild and moderate depression, followed by high-intensity psychotherapy and/or pharmacotherapy only if required; patients with moderate-severe depression start with high-intensity psychological therapy and/or pharmacotherapy 191 From NICE guideline 2004, the Stepped Care Model was applied to depression intervention in the community, especially the first 3 steps helped primary health care staff to overcome significant barriers and core skills in depression intervention, and improve community awareness 192 This format was proven to decrease the risk of the depression onset by 50% in older patients in primary care 193 Until now in the UK Stepped Care IAPT program has become the dominant model of treatment organization 194 The research demonstrated that the majority of patients received low-intensity care; about 30% were either assigned immediately to high-intensity psychological therapy or "stepped up" to it 195

To measure the effectiveness of the Stepped Care programs, many studies used a simple measure such as the nine-item Patient HealthQuestionnaire (PHQ-9) to evaluate the depressive symptoms at baseline and after a period of follow-up It is the most simple and broadly used screening tool in depression identification and management in the community 129,131 InNetherlands, the study about Stepped Care Model for depression in primary care showed that the 12-month-incidence of depressive disorders was reduced by 50% compared to the control group 194 In the UK, the studies about IAPT evaluating the clinical effects of Stepped Care services utilized the minimum data set on patient outcomes which included the PHQ-9 196 The research showed favorable effects which were observed over 2 years with uncontrolled pre-post treatment effect sizes of 1.07 196–198 Moreover, these evaluations reported better clinical outcomes for patients who completed treatment 197,199 Among patients who completed treatment, the recovery rates at Doncaster and Newham were in excess of 50% 198,199

In LMICs, the Stepped Care Model has been proven to be effective and appropriate In India, the MANAS program is the first collaborative stepped- care interventions including case management, psychosocial interventions and antidepressants led by non-specialists under the supervision of mental health specialists The study showed strong evidence of effectiveness in depression among patients in primary care facilities compared to the usual care group 177 In Chile, a stepped care program including group psychoeducation, behavioral activation therapy, and problem-solving techniques led by non-medical health workers demonstrated remarkable effectiveness in socially disadvantaged women with depression at 6 months 200

Not only in general depressed populations, the Stepped Care Model has also been proven to be effective in diverse populations The study about this model represented the remarkable effectiveness of integrated depression intervention in obstetric and gynecological care in US 201 A similar study presented a greater improvement in depression outcomes for socially disadvantaged female patients 202 In Dutch, the stepped care treatment with 4 steps including watchful waiting, guided self-help, problem-solving therapy,and referral in primary care showed no significant advantage compared to usual care However, the authors pointed out that the usual care patients were not a good control group because of high convenience and accessibility in the Netherlands, leading to lower actual benefit of the program 203 Based on this suggestion, a nurse-led stepped-care study among diabetes and heart disease patients with depressive symptoms demonstrated the usefulness in both improving and preventing depression in primary care 204

In Vietnam, the initial application of community intervention with stepped care principles and task-shifting also noticed potential outcomes The study about the MCCD program at 12 months showed the effectiveness in reducing depressive symptoms 12,103 Based on that project, the LivelihoodIntegration for Effective Depression Management (LIFE-DM) program was implemented in Da Nang with the support of psychiatric hospitals It combined with the livelihood support including microfinance loans and income generation skills training for disadvantaged depressed women aged18-65 187 From these positive results, the LIFE-DM program was expanded to more communes in Da Nang and Thua Thien Hue As a result, 40,000 people were screened for depression, 2,541 people with depressive symptoms were referred to the CHS, of which 914 people were diagnosed, 92% accepted treatment and 73% completed The mean PHQ-9 scores decreased from 13.8 to 5.1; combined with improved anxiety and quality of life scores, and these figures continued to improve at 12 months 205 A study comparing this program to antidepressant treatment among 475 subjects indicated significant improvements in mental health and functional performance 12 Another community-based stepped care program for depression was conducted in HaNam province in 2013-2014 with the participation of 21 CHS and a district hospital The screening reached 1,951 people, then the intervention including psychoeducation and 8-week-yoga was implemented with moderate depression and supervised by specialists The effectiveness was identified to be higher than standard care in reducing the depression with the mean decrease of 5.72 points in PHQ-9 206

The acceptability and feasibility of the Stepped Care Model for

Many stepped care programs were recognized as acceptable and feasible for the general public globally The study about IAPT in the UK which applied stepped care model for depression management in the community demonstrated a promising attendance rate of 75% 196–198 The study about the Stepped Care Model integrated into obstetric and gynecological care in US demonstrated that nearly 100% reported the need of depression intervention 201 In addition, some evidence of effects on cost-saving and employment was noticed Among patients attending Doncaster and Newham in year one, there was a 4% net increase in the number at work Across Pathfinder sites, the number of patients returning to work and off statutory sick pay increased by 16% 198,199 The studies about the cost-effectiveness of collaborative care interventions using Stepped Care Models demonstrated controversial outcomes 119,207 However, with less expensive initial delivered interventions, stepped care approach may be cost-effective overall 188 The cost-effectiveness is an important base to enhance the feasibility of the intervention in the community.

Furthermore, many stepped care community-based interventions for depression brought about benefits in terms of human resources, the public outcomes, and possible long-term maintenance In US, the Improving MoodPromoting Access to Collaborative Treatment model for late-life depression had been applied with the common role of behavioral health professionals 208 It was demonstrated to be more feasible and effective than usual care with greater depression reduction, patient satisfaction, and quality of life, and less functional impairment 209 Another stepped care in the US resulted in significantly improved depression compared to usual care 210 It also revealed a small to moderate functional enhancement and disability reduction among depressed patients with persistent symptoms in primary care compared to usual care after 6 to 8 weeks 211 The study about Stepped Care model in Chile emphasized the advantages of saving resources to improve depressive symptoms up to 6 months 200

In Vietnam, the stepped care model was introduced as community- based collaborative care with task-shifting for primary care and simple psycho- educational intervention with significant acceptability The report of the MCCD program emphasized the importance of raising community awareness and primary health care providers’ capacity for depression intervention 12,103 Similarly, the report about the LIFT-DM project also highlighted the high acceptance and capacity building of depression detection,management and referral in the locality 212 In general, the task-shifting approaches in which trained non-mental health workers do the screening and provide the simplest intervention for people with depression under the supervision of specialists, which is proved to be acceptable and feasible in scarce-resource settings 46,213,214 The study disseminating the task-shifting interventions in 8 communes in 2017 also suggested the effectiveness and feasibility of the Stepped Care Model based on community resources for depression intervention in the context of Vietnam 215,216

RESEARCH SUBJECTS AND METHODS 46

Study subjects

2.1.1 Objective 1: Screening for depression at home

- People aged between 18 and 65 years old.

- Settling in a certain commune of Thai Nguyen province.

- Agree to participate in the screening.

● Exclusive criteria: Having serious physical diseases or mental disorders that could interfere with the interview.

2.1.2 Objective 2: Intervention group at CHS

- Screened at CHS with PHQ-9 score ≥ 10.

- Agree to participate in the group intervention.

● Exclusive criteria: Having high risk of drug abuse, suicide, or psychosis which needed to be treated at psychiatric hospital.

- In-depth interview with 20 group subjects (2 people/commune) at 3 months and 12 months after the intervention.

- Group discussions at 6 months after the intervention with 6 psychiatrists of Thai Nguyen provincial psychiatric hospital who supervised the screening and intervention process, and 20 CHS staff who completed the training course and had capacity to deliver the group intervention.

Study time and sites

The study was conducted in 10 selected communes of Thai Nguyen city in Thai Nguyen province including Quyet Thang, Tan Thinh, Thinh Dan, TanLap, Trung Thanh, Gia Sang, Huong Son, Cam Gia, Tich Luong, Phu Xa.

- Time of research: From August 2020 to December 2022.

- Time of intervention: From September 2020 to February 2021.

- Time of follow-up: From March 2021 to March 2022.

Study design

- Objective (2): An intervention study, quasi-experimental design (pre- test and post-test studies) with quantitative data.

- Objective (3): A qualitative study including in-depth interviews and group discussions.

Sample size and sampling method

The sample size was calculated using the formula for estimating the sample size in the before-after study: 230

(𝐸𝑆)2 In which, n = minimum sample size of the intervention group.

C = 13 is a constant related to type I and type II error with α = 0.05, β 0.05. r = 0.6: the correlation coefficient.

ES = d / s: the effect size with d = 2 (desired mean effect of the intervention) and s = 6.12 (the standard deviation of the PHQ-9 score in the depressed patient group in Vietnam) 123

Thus, the minimum sample size of the intervention group is n = 97.

With the estimated rate of ineligibility and refusal to participate in the intervention group of 40%, our study is expected to recruit at least 162 people(97/0,6 = 161.67) with scores of PHQ-9 ≥ 10 in 10 communes of ThaiNguyen city According to Tran Viet Nghi (2002), the rate of depression in one commune of Thai Nguyen province was 8.35%, so that the minimum pre- people by using the PHQ-2 In fact, this study approached 1,689 people to screen for depression with the PHQ-2 in the community.

The research team contacted localities and Provincial Psychiatric Hospitals to select 10 communes in Thai Nguyen city that had appropriate and convenient CHSs to organize intervention and monitoring.

The trained collaborators (5 people/commune) randomly came to households and interviewed the eligible people who were present at home.

The selection interviews were conducted with structured paper forms until the sample size reached at least 140-150 subjects/commune.

Randomly selecting 20 subjects who participated in the group intervention (2 people/commune) and contacted in-person (at 3 months) or via mobile-phone (at 12 months follow-up) The researcher eliminated overlapping cases between two interview times.

During the study process, there were 20 CHS staff/10 communes and 6 provincial psychiatrists participating in the screening and group intervention at the CHS In total, 26 medical staff were recruited in the study at baseline.

In the follow-up period, one representative CHS staff/commune and all6 psychiatrists were invited to join qualitative study However, there was oneCHS staff member who was absent because of a personal problem at the specific time of interview Therefore, 15 medical staff (including 9 CHS staff and 6 psychiatrists) participated in group discussions after the intervention.

Research variables and indicators

The detailed description of variables was put on the Appendix 2.

- The outcome or dependent variable was the prevalence of depressive symptoms as measured by the PHQ-2 scale, which was categorized into a binary variable: 'No depression' for scores below 3, and 'Depression' for scores equal or greater than 3.

+ Age: This variable captures the chronological age of the subjects.

+ Sex: This categorical variable is coded as 1 for male and 2 for female subjects.

+ Educational level: This ordinal variable represents the highest level of education attained by the subjects, with categories coded as follows: 1 for Primary School, 2 for Secondary School, 3 for High School, and 4 for College/University.

+ Marital status: The current marital status was classified as 1 for being single/divorced/widowed, and as 2 for being married.

+ Medical insurance: This variable indicates the level of health insurance coverage, with 0 denoting no coverage, 1 for 80% coverage, 2 for 95% coverage, and 3 for 100% coverage.

+ Household income (Viet Nam Dong): This ordinal variable reflects the average monthly income of the entire family over the past 12 months, with categories as follows: 1 for less than 500,000 VND, 2 for 500,000 to less than 2,000,000 VND, 3 for 2,000,000 to less than 5,000,000 VND, 4 for 5,000,000 to less than 10,000,000 VND, and 5 for more than 10,000,000 VND.

The qualitative data included in-depth questions about the personal reasons of participation among group subjects This result explored clearer for

- The primary outcomes or dependent variables include:

+ Depression score: Assessed with the Patient Health Questionnaire (PHQ-9), which consists of 9 questions The total score can range from 0 to 36 In the range from 5 and above, higher scores indicated more severe depressive symptoms.

+ Depression reliable improvement/deterioration: An decrease/increase in the PHQ-9 score of at least 6 points from the initial assessment was considered a reliable improvement/deterioration, respectively.

+ Depression recovery: was determined by a reduction of PHQ-9 score at least 50% compared to baseline PHQ-9 scores to be less than the cut-off score of 10.

+ Quality of Life Score: represented by the Quality of Life Enjoyment and Satisfaction Questionnaire Short Form (Q-LES-Q-SF) with 14 questions.

The total score ranges from 0 to 70, with higher scores reflecting better quality of life.

+ Anxiety Score: measured by the Generalized Anxiety Disorder-7 items scale (GAD-7) Scores range from 0 to 21, with higher scores from 5 and above suggesting greater anxiety levels Note: this is not a diagnostic anxiety assessment.

+ Brief Resilient Coping Score: assessed by the Brief Resilient Coping Scale (BRCS) with 4 questions, total score ranges from 4 to 20 Higher scores indicated a more resilient coping capacity.

+ Employment rate: the percentage of the group subjects who had a job

The qualitative data included in-depth questions about the perceived effectiveness of group intervention in group subjects after 3 months and 12 months This result explained more clearly for quantitative data in objective 2.

Based on previous theoretical frameworks about the acceptability and feasibility of healthcare interventions in public health, this study assessed the following qualitative themes via in-depth interviews and group discussions: 231,232

● The acceptability of community-based group psychological intervention at CHS including:

+ Participatory and Burden: personal motivation and effort to participate, participatory rate of group intervention.

+ Affective attitude, Ethicality, and Intervention Coherence: personal feeling and impression about the intervention, research subjects’ understanding and application of learned skills; required costs and resources to participate in the intervention.

+ Perceived effectiveness and Self-efficacy: intervened subjects’ perceived changed emotion and skills, social network and quality of life; the change of personal awareness about mental health and depression after the intervention.

● The feasibility of community-based group psychological intervention at CHS including:

+ For group subjects: Perceived advantages and benefits of group psychotherapy at CHS (content of intervention, group organization, socio- cultural factors); Barriers and disadvantages of intervention (space, time,group activity, socio-cultural factors); the maintained activities and the need of identification and intervention, personal skills; the coordination with provincial specialists.

+ For provincial specialists: the change of personal capacity in community intervention and communication skills, the coordination with primary health care settings in depression intervention.

Some research assessment criteria in the study

2.6.1 Research measurements for quantitative data:

The study used a paper set of questionnaires including the general information and the psychometric tests The psychometric tests consisted of 5 measures to investigate the specific depression-related variables and factors.

The subjects were interviewed at baseline and at 3-month, 6-month, and 12- month follow-ups, all data was collected by interviewers These psychometric measures included:

- PHQ-9 (Patient Health Questionnaire, 9 questions): The PHQ-9 including nine questions was developed by Kroenke et al in 2001 based on the diagnostic criteria of depression in the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) 121 For each of nine diagnostic items, the responses can be “not at all,” “several days,” “more than half the days,” and “nearly every day,” scored as scores of 0, 1, 2 and 3, respectively.

Thus, the total score range is from 0 to 27 For subjects who answered positively to any of these nine items, item 10 asks about how difficult these problems made the response to do work, take care of things at home, or get along with others In general, a total score of 0–4 suggests no depression; scores of 5–9 shows mild depression; a score of 10–14 represents moderate depression; a score of 15-19 suggests moderately severe and a score ≥ 20 indicates severe depression It takes about 10 minutes to complete PHQ-9 cutoff scores of 10 had a sensitivity with the sensitivity of 91.7% and specificity of 78.3% which can be considered as good as clinician-administered instruments 122 In clinical practice, the PHQ-9 is usually used to screen, evaluate the severity of depression and follow the response of treatment This screener has been validated among Vietnamese populations 123

- PHQ-2 (Patient Health Questionnaire, 2 questions): The PHQ-2 comprises the first 2 items of the PHQ-9 about the frequency of depressed mood and anhedonia over the past 2 weeks With a score of 2 or above, the PHQ-2's sensitivity and specificity for diagnosing depression were 86% and 78%, respectively, and 61% and 92% with a score of 3 or higher 124 In other words, using the PHQ-2 cutoff score of 2 detected more cases of depression in the community than a cutoff score of 3 Therefore, in this study, we used the PHQ- 2 cutoff score of 2 to enhance the ability of depression detection in the community After the primary screening, the PHQ-2 cutoff of 3 was used to explore the related factors with 83% sensitive and 90% specific for detection of depression 234 In general, it takes about 2 minutes to complete.

- GAD-7 (Generalized Anxiety Disorder, 7 questions): The GAD-7 is a screening tool for anxiety with a recommended cut off score of 10 or greater.

The responders rate the frequency of each anxiety symptom on a four-point scale ranging from 0 (not at all) to 3 (almost every day) within 2 weeks A cut- off of 10 has been identified as the optimal point for sensitivity (89%) and specificity (82%) 235 It takes about 10 minutes to complete The scale's reliability has been evaluated within the Vietnamese population,demonstrating a Cronbach's alpha coefficient of 0.94, indicating high internal consistency 236 In this study, that scale was used to access any coexisting anxiety symptoms with depressive symptoms which was noticed the significant rate of 53% and

Questionnaire Short Form): This questionnaire evaluates the degree of enjoyment and satisfaction in daily functioning that are assessed in the longer form of the Q- LES-Q A total score is derived from 14 items with a maximum score of 70 with higher scores indicating greater life satisfaction and enjoyment 238

- BRCS (Brief Resilient Coping Scale): includes 4 items Total sum scores range from 4 to 20 Scores of 4-13 indicate low resilient coping, 14-16 indicate medium resilient coping and 17-20 indicate high resilient coping The scale focuses on the tendency to effectively use coping strategies in flexible, committed ways to actively solve problems despite stressful circumstances.

This scale is validated in Vietnamese and used to evaluate the resilience coping skills 239

- The index of depression improvement and recovery: Reliable and clinically significant change criteria, as defined by Jacobson and Truax, 240 are among the most commonly used methods of quantifying improvement in studies of psychological treatments At a conceptual level, clinically significant change defines the improvement as a move from a clinical to a non-clinical range Jacobson and Truax provided several operational definitions of cut-off points to distinguish between these ranges based on the central location and distribution of scores for a clinical and non-clinical group As an additional criterion, the change in scores must be greater than that which could be due to the inherent unreliability of the measure Patients were deemed to reliably recover if they initially scored above the clinical cut- off on the PHQ-9, then showed reliable improvement during treatment (at least a 50% reduction in baseline PHQ-9 scores), and finally scored below the clinical cut-offs at the end of treatment Similarly, reliable improvement was assessed using Jacobson and Truax reliable change criteria with a decrease of at least 6 points in the PHQ-9 research involving community-based depression intervention 241

2.6.2 Research measurements for qualitative data:

Qualitative study using instructions for in-depth interviews and group discussions was conducted by in-person interviews (at 3 months) and telephone interviews (6 and 12 months) because of COVID-19 pandemic.

Each in-depth interview with group subjects was conducted one-by-one by trained researchers in about 10-15 minutes Group discussions with health care staff were conducted online via Zoom for about 60 minutes with each group The qualitative data was recorded and taped during in-depth interviews and group discussion at a specific point of time (Figure 2.4) Instructions for specific qualitative interview content were clearly stated in the Appendix.

Study intervention

The study involves a Stepped Care Model and task-shifting for citizens, then refer the at-risk individuals to the nearest CHS for confirmation.

The second step consisted of psycho-education and group psychotherapy led by CHS staff under specialists’ supervision at CHS The third and fourth steps involved more specialist treatment in which the people with more severe depression or high-risk symptoms were referred to psychiatric hospitals This intervention process had five steps as following:

2.7.1 Training for healthcare staffs and collaborators

The training materials were developed based on the Manual for Group Facilitators in the Project entitled “Strengthening Health System Capacity on Community Based Care for Effective Depression Management in Thai Nguyen and Ninh Binh province, Vietnam” which was manualized and approved by mental health care experts.

2.7.1.1 Training at the provincial level:

- Duration: 3 days - Location: in the hall of the Provincial Mental Health Hospital.

- Trainer: experienced one psychiatrist and 2 psychologists.

- Material: presentations, handouts, and manualized books.

- Content: Introducing Stepped Care Model for depression and research;

How to use research tools; Group psychological intervention for depressed adults in the community; Necessary skills in supportively supervising CHS staff; Communication skills in the community.

- Duration: within 2 months Included 9 days (3 sessions, 3 days/session) of theoretical lectures Interspersed practice time (role playing, video recording, virtual supervision and comments for lessons learned)

- Location: Theoretical lectures at the hall of the Provincial Mental Health

- Trainee: 20 CHS staff (2 staff/commune) including 14 general practitioners, 4 nurses, one midwife, and one pharmacist And 3 officials from the city medical center as supervisors.

- Material: presentations, handouts, and manualized books.

+ An overview of the research and tools.

+ Practice using research tools and forms.

+ Content of 8-session group behavioral activation therapy.

+ Principles and skills of group facilitation and community activation including: how to start a conversation, how to ask questions and create interest, how to quiet people and use silence, how to invite to speak; how to use hands, body movements and facial expression, how to stop a conversation and manage people with strong personalities, how to manage anger, how to manage space and time, how to organize games

- Location: in the hall of the Provincial Mental Health Hospital.

- Trainer: experienced one psychiatrist and 2 psychologists, and 6 trained provincial psychiatrists.

- Trainee: 5 collaborators/commune, total of 50 people including volunteer and eligible village health workers and members of the local women's union.

- Material: presentations for education, leaflets and posters for communication.

- Content: Psychoeducation about mental health and depression; How to use communication materials and screening tool (PHQ-2) in the community.

● Method of the training and follow-up: using paper forms and direct reports Specialists including psychiatrists and psychologists supervised and provided indirect support through local staff.

2.7.2 Screening for depression, recruitment and referral - Screening depression in community and referring to the CHS

People living in the community were approached randomly at home and screened with the PHQ-2 by trained village health workers and local collaborators People having PHQ-2 score ≥ 2 received the information about the study and were referred to the CHS for the next screening at CHS.

-Screening depression at the CHS

People with PHQ-2 score ≥ 2 coming to the CHS for clinical screening were assessed (using the PHQ-9) by CHS staff and diagnosed by psychiatrists from Thai Nguyen Provincial Mental Health Hospital All research subjects having PHQ-9 score ≥ 10 were informed about the program and referred to the further assessment for risk of severe mental disorders All the individuals having PHQ-9 score < 10 were noted in the watchful list.

-Screening for high-risk cases and referral to higher level of care

CHS staffs and psychiatrists screened the subjects having PHQ-9 score ≥ 10 for the risk of severe mental disorders (psychotic symptoms, suicide, substance abuse) If any, referral to the provincial psychiatric hospital for higher level treatment.

- Recruiting the intervention group at CHSEligible depressed people (having PHQ-9 score ≥ 10 and after clinical assessment of psychiatrists) received simple psycho-education from CHS staff under the supportive supervision of psychiatrists and were invited to participate in group intervention at CHS.

People with depression who were voluntary to participate in group intervention received community-based depression care services including psychoeducation and group psychotherapy at the CHS by trained CHS staff under the supportive supervision of provincial psychiatrists, 90 minutes each week, during 8 sessions.

- Assessing personal perspectives about the mental health and depression, and readiness for change.

- Providing information about mental health and depression to decrease stigma, and educate the patients about benefits and availability of depression treatment in this program.

● Group behavioral activation therapy at CHS:

- CHS staff carried out group behavioral activation therapy with 14-20 people/group within 8 weekly sessions, about 60-90 minutes/session at CHS.

- Mental health staff from provincial psychiatric hospitals provided supportive supervision during the group intervention at CHS. and activity.

+ Session 2: Doing healthy activities; Motivation to overcome depression.

+ Session 3: Choosing appropriate activities; Keeping life balance.

+ Session 4: Goal setting; Activity planning.

+ Session 6: Importance of social support network; Link between social support and mood.

+ Session 7: Simple and effective communication skills to improve social connections.

+ Session 8: Relapse prevention and graduation.

- In particular, each session had five parts including (1) welcome and warm-up for 10 minutes; (2) Review the previous session for 10 minutes; (3)New content for about 30 minutes; (4) Summary and key messages; (5)Practice and see you again Especially, all the lesson contents were integrated into fun activities and team building games to increase group energy and learning efficiency, based on animation technique to increase group energy and learning efficiency 233

Researchers conducted follow up evaluation at the time of 3 months, 6 months and 12 months following the completion of intervention.

- Quantitative interviews used a paper-based questionnaire with all group subjects at baseline and at the time of 3 months, 6 months and 12 months after the intervention.

- Qualitative study was conducted by in-person interviews (at 3 months) and telephone interviews (6 and 12 months) because of COVID-19 pandemic.

The researcher randomly selected subjects to interview and invited to participate in group discussions from the list of research subjects Qualitative data collection was conducted in each commune in turn until the expected number of patients was collected.

+ In-depth interviews at 3 months and 12 months after finishing group activities with 2 patients/ commune The total was 20 patients/ 10 communes. communes) at 6 months after the group intervention.

Figure 2.4 Data collection and follow-up process.

Possible errors and error control measures

Selection bias might happen when the collaborators predicted the people with higher risk of depression (e.g, divorced women, people suffering domestic violation or chronic disease, poor people, caregivers of people with chronic severe diseases) and reached these vulnerable people on purpose We communicated and thoroughly trained for collaborators to randomly approach and screen subjects in the community.

Misclassification bias happened as incorrectly using PHQ-2 and PHQ-9 to identify the depressive patients to refer to CHS and invite to join group collaborators and CHS staff were structured with both theoretical and practical sessions (role play) to provide accurate screening skills for screeners before proceeding with the community.

The lack-of-experience interviewers put more attention on the negative factors of depression in some cases We trained carefully for interviewers and used a set of standardized questionnaires, and psychiatrists supervised the interview process at CHS Furthermore, all subjects who were eligible to join the group were re-screened with psychiatrists to confirm screening results and rule out risk of severe conditions.

The depressive patients usually have cognitive problems, reduced memory or concentration, leading to recall bias The interviewers were carefully trained to confirm the information many times during the communication and interview, and the personal data of patients were checked again at every point of follow-up to control this bias.

Statistical analysis

● Quantitative data was collected via paper questionnaires by the program coordinator at baseline and at 3-month, 6-month, and 12-month follow-ups Data was entered using Epi Data v3.1 software and analyzed by SPSS 20.0 software.

- Quantitative variables are tested for standard distribution The normal distribution variable was expressed as mean ± SD.

-Compare between two quantitative variables by Student T test (if normal distribution variable) or Mann-Whitney-U test (if distribution variable is not standard).

ANOVA to compare the differences of all groups If the variables did not meet the assumptions of the ANOVA variance analysis, the Kruskal-Wallis test was used instead.

-Logistic regression was performed to explore risk factors with subjects screened for depression using that PHQ-2 score.

-Cohen's d was determined by calculating the mean difference between two groups, and then dividing the result by the pooled standard deviation We used the following rule of thumb when interpreting Cohen's d:

+ A value of 0.2 represents a small effect size.

+ A value of 0.5 represents a medium effect size.

+ A value of 0.8 represents a large effect size.

-This study was not suitable for time series analysis for the following reasons Firstly, the most appropriate approach for a quasi-experimental public health intervention study without randomization and no control group was interrupted time series analysis 242 In this analysis, an important assumption was that the pre-intervention trend was linear However, this study had only one pre-intervention assessment that could not replicate the pre-intervention trend, thus violating that assumption 243 Studies also showed that fewer pre- intervention assessments reduced the strength of the model to perform time series models 244 Secondly, the time series analysis estimates have not been controlled for covariates, and there is no comparator against which to adjust the results for changes that should not be attributed to the intervention itself 242 Therefore, we use the generalized estimating equation (GEE) model to evaluate confounding factors that affect intervention effectiveness.

-GEE was used to estimate the parameters of a generalized linear model with a possible unmeasured correlation between observations from different time points The GEE model is well-suited for analyzing outcomes in robustness in yielding unbiased estimates of population-averaged effects, maintaining accuracy despite potential inaccuracies in the assumed correlation structure among repeated measurements.

-The level of significance was set at p value less than 0.05.

● The qualitative data was recorded and taped during the in-depth interviews and group discussion Thematic analysis was used to identify topics that highlight group subjects and health-care workers' responses to acceptability and feasibility of community intervention, the study subjects' benefit and experience Microsoft Word and Excel software were used to assist with data processing and analysis Qualitative variables were expressed in percentages and compared by the  2 test or Fisher Exact test.

STUDY RESULT 66

Several depression-related factors in screening population 71

High school or less 808 (58.0%) 586 (42.0%) 1394 (100.0%) College/University 161 (54.6%) 134 (45.4%) 295 (100.0%)

The prevalence of depression in people over 50 years old was higher when compared to people under 50 years old (36.3% and 46.4% respectively).

This difference was statistically significant with p-values less than 0.05.

Table 3.3 shows the result of logistic regression models exploring the association between depression and risk factors Individuals aged over 50 were

1.75 times as likely to exhibit symptoms of depression compared to those under50 Besides, the study subjects having 100% coverage of medical insurance were more than two times more likely to suffer from depression than those having no medical insurance In addition, people having average household income more than 10 million VND were at a 50% reduced likelihood of

The initial effectiveness of the community-based intervention with

Among 427 people having a PHQ-9 score of 10 or higher, 382 eligible people were invited to the group intervention at the CHS In which, 359 individuals agreed to join the group intervention at CHS and were interviewed for data collection at the baseline At 3 months, 6 months, and 12 months after the intervention, researchers and local collaborators contacted to invite research subjects to go to the CHS for re-evaluating and data gathering If the research subjects refused to participate, their information was stored for further contact by health workers in the locality As a result, 3 subjects did not participate in data collection after the intervention for personal reasons.

(Figure 3.5) No adverse events were recorded during the intervention and follow-ups.

Most intervention subjects were female (93.54%), and the average age was 55.1 years old Most of them completed secondary school or higher (92.98%) The number of married individuals accounted for more than 70% of the intervention population.

Figure 3.5 Distribution of depression severity over follow-up time.

At baseline, most subjects predominantly exhibited moderate to severe depressive symptoms, with moderate levels of approximately 75% After 3 months, a significant shift was observed with the vast majority of study subjects showing either an absence of symptoms or only mild symptoms remained Notably, moderate-to-moderately-severe-symptom group persisted in nearly 10% of subjects After 12 months, over 80% of study subjects were seen without depressive symptoms It is noteworthy that moderate-to-severe- symptom groups were confined to a mere 4 individuals, constituting 1.2% of the intervention population. months, 6 months, and 12 months after intervention (n56)

Overall, the mean score of PHQ-9 decreased significantly between the time before and after the intervention, and there was a gradual decline of this figure over 12 months of follow-up after the intervention In particular, after three months of follow-up, PHQ-9 scores reduced dramatically from an average of 13.29 points before the intervention to an average of 4.96 points.

At the 12- month follow-up, the mean PHQ-9 score dramatically decreased to2.83, equivalent to a nearly five times reduction.

PHQ-9 scores Baseline After intervention

Pre-post effect size (Cohen's d) p- value

After three months, the average PHQ-9 score of the study subjects decreased significantly from 13.29 to 4.96, then continued to decline to 3.55 points at the six-month follow-up with an effect size of 3.09 After 12 months, the intervention helped the mean depressive score reduce by 10.45 points.

Table 3.6 The change of depression severity based on the PHQ-9 score at baseline compared to the time of 3-, 6-, 12-months follow-up (n56)

Depression severity based on PHQ-9 score

Severe (≥ 20) 15 (4.2) 0 (0.0) 1 (0.3) 2 (0.6) this rate decreased to less than 10% at the point of 3 months after the end of the intervention, and decreased to less than 4% after 12 months At 12 months post- intervention, the majority of patients had returned to the depression-free level In the cases which remained depressive symptoms, most were mild The differences in depression levels at 3-, 6-, and 12 months after the intervention compared to the baseline were statistically significant (p < 0.001).

Table 3.7 Patient's response at 3-, 6-, 12 months after intervention (n56)

After 3 months After 6 months After 12 months n (%) n (%) n (%)

Table 3.7 illustrated the significant response of depressive individuals after the intervention up to one year follow-up At the point of three months after the intervention completion, more than 90% of patients recovered (at least a 50% reduction in baseline PHQ-9 scores), and 75% of study subjects had at least 6 point decline on the PHQ-9 (defined as reliable improvement).

These figures even improved over time of 1-year follow-up with ninety-six percent of study subjects recovered and 90% had reliable amelioration in depressive symptoms.

Pairwise comparison of regression coefficients Coef p-value 95% CI

Table 3.8 demonstrated the results of a univariate linear regression model analyzing the efficacy of each follow-up time point after the intervention The 12-month follow-up period demonstrated the most remarkable intervention efficacy, with an average decrease in PHQ-9 scores of 10.461 points This result is statistically significant with p < 0.001.

Table 3.9 The generalized estimating equation regression model estimating the intervention effectiveness according to the PHQ-9 score

After adjusting for scores of GAD-7, Q-LES-Q-SF, and BRCS, the 12- month follow-up period was associated with the most significant decline inPHQ-9 scores, with an average decrease of 5.87 points. score (Adjusted for demographics)

The generalized estimating equation model, after adjusting for several factors, showed that the mean PHQ-9 score dropped by an average of 4.864,5.313, and 5.811 points at the point of 3-,6-,12-month follow-up, respectively.

Table 3.11 The change of GAD-7 scores at baseline compared to the time at

GAD-7 scores Baseline After intervention

Pre-post effect size (Cohen's d) p- value

Table 3.11 demonstrates the effectiveness of the intervention in reducing GAD-7 scores over a 12-month period At baseline, the mean PHQ-9 score was 9.40, which significantly decreased post-intervention across all time points measured At 3 months, the mean score was reduced by 5.81 points, with a large effect size (Cohen's d = -1.42), indicating a substantial clinical improvement This trend continued at 6 and 12 months, with mean reductions of 7.26 and 8.0 points, respectively, and correspondingly larger effect sizes of -1.90 and -2.21 The p-values for all time points were less than0.001, underscoring the statistical significance of the findings.

Q-LES-Q- SF scores Baseline After intervention

Pre-post effect size (Cohen's d) p- value

Table 3.12 provides results of the intervention's effect on Quality of Life Scale (Q-LES-Q-SF) scores at 3, 6, and 12 months post-intervention It is observed that the mean pre-intervention score was static at 31.99 (SD=5.27).

Post-intervention measurements indicate a progressive improvement in quality of life scores, with mean differences of 10.04, 10.38, and 20.97 at 3, 6,and 12 months respectively The effect size, denoted by Cohen's d, suggests a substantial increase from 1.42 at 3 months to 2.10 at 12 months, with all intervals showing statistical significance (p