The Prevalence of Symptomatology and Risk Factors in Mental Health in Mexico: The 2016–17 ENCODAT Cohort

There is little recent information about the prevalence of symptomatology of mental health disorders in representative population samples in Mexico. To determine the prevalence of mental health symptoms in Mexico and its comorbidity with tobacco, alcohol, and drug use disorder (SUD), we used the 2016–17 National Survey of Drug, Alcohol, and Tobacco Use (Encuesta Nacional de Consumo de Drogas, Alcohol y Tabaco, ENCODAT 2016–2017). The data were collected from households using a cross-sectional, stratified, multistage design, with a confidence level of 90% and a response rate of 73.6%. The final sample included 56,877 completed interviews of individuals aged 12–65, with a subsample of 13,130 who answered the section on mental health. Symptoms of mania and hypomania (7.9%), depression (6.4%), and post-traumatic stress (5.7%) were the three main problems reported. Of this subsample, 56.7% reported using a legal or illegal drug without SUD, 5.4% reported SUD at one time on alcohol, 0.8% on tobacco, and 1.3% on medical or illegal drugs, 15.9% reported symptoms related to mental health, and 2.9% comorbidity. The prevalence found is consistent with those reported in previous studies, except for an increase in post-traumatic stress, which is consistent with the country’s increase in trauma.


Introduction
Mental health disorders are among the health problems associated with the greatest levels of disability [1]. Mood disorders, anxiety disorders (including specific phobias), and alcohol use disorders are the most prevalent worldwide [1]. Among the risk factors that increase the probability of experiencing these disorders are low educational level, violence, low socioeconomic level, polygenic load, problems at work, and interpersonal problems. There may also be increased risk from having presented one or more mental health difficulties or disorders in early childhood or adolescence [2,3]. The National Health Survey in Mexico (Encuesta Nacional de Salud, ENSANUT, 2018) includes questions about depression symptomatology [4], and the national addiction surveys include questions about suicide attempts and emotional distress [5].
Surveys from various countries, including Mexico, show a lifetime prevalence of mental disorders according to DSM-IV criteria, with range of 47.4% and 12.0% in the U.S. and Nigeria, respectively. The extremes for the prevalence of anxiety and depression disorders are in the U.S. (31% and 21%, respectively) and China (4.8% and 3.6%, respectively). Disorders of impulse control are highest in the U.S. (25%) and the lowest in Nigeria (0.3%) [1]. The prevalence of substance use disorders is highest in Ukraine (15%) and lowest in Italy (1.3%) [1]. Glantz et al. (2020) [6] found a range in the prevalence of alcohol use disorder (AUD) from 0.2% in Iraq to 6% in the U.S. and from 0.5% in Iraq to 18.7 % in Australia for alcohol abuse. Auerbach et al. (2018) evaluated the prevalence of mental illness in 19 universities in eight countries, including Mexico, and estimated that 35% of those surveyed suffered from at least one disorder such as major depression, mania/hypomania, generalized anxiety disorder, panic disorder, alcohol use disorder, or substance use disorder [7]. In Mexico, the most recent national survey of psychiatric epidemiology was in 2002, when the prevalence of psychiatric disorders was estimated as 30.4% in men and 27.1% in women. That survey reported that 6.7% of men and 11.2% of women had presented an affective disorder (depression, mania, hypomania, or dysthymia) at least once in their lives, 9.5% of men and 18.5% of women an anxiety disorder (panic, agoraphobia, social phobia, specific phobia, generalized anxiety, or post-traumatic stress), 17.6% of men and 2.0% of women a substance use disorder, 11.5% of men and 1.0% of women AUD, 0.7% of men and 0.2% of women drug use disorder (DUD), and 2.9% of men and 0.9% of women nicotine use disorder (NUD) [2]. Benjet et al. (2009) found that the prevalence is lower in adolescents, but that the more common disorders were specific phobias and social phobia. In women, the more common disorders were phobias, major depression, oppositional defiant disorder, agoraphobia without panic disorder, and separation anxiety; in men, they were oppositional defiant disorder, alcohol abuse, and conduct disorder. In general, women presented a greater number of disorders and a greater prevalence of each [3]. It is important to note that, to date, neither the frequency of obsessive-compulsive disorder nor that of post-traumatic stress disorder has been reported on the populational level.
Despite various efforts, there has been no more recent study in Mexico of the state of the mental health of the population, but updated data are important for the improvement of care and treatment. The objective of this study is, thus, to determine the prevalence of mental illness symptomatology in Mexico and its comorbidity with the use of tobacco, alcohol, and drugs, obtained through the 2016-17 National Survey of Drugs, Alcohol, and Tobacco Use (Encuesta Nacional de Consumo de Drogas, Alcohol y Tabaco 2016-17, ENCODAT 2016-17).

Materials and Methods
The study was carried out on a subsample of the National Survey of Drug, Alcohol, and Tobacco Use, administered in 2016 to participants aged 12-65 in urban and rural Mexican households [5]. The objective of the survey was to evaluate the patterns of use of different psychoactive substances and mental health problems in Mexico. The sample had a cross-sectional, stratified, probabilistic, and multistage design, with a confidence level of 90% and a response rate of 73.6%. The sample universe was made up of primary sampling units (PSUs) that were the sum of basic statistical geographic areas (BSGAs; Áreas Geográficas Estadísticas Básicas), stratified by state and urban-rural community. An adult aged 18-65 was chosen at random in each household, and in those with at least one household member under 18, an adolescent aged 12-17 was chosen at random. The persons chosen were administered individual questionnaires. The final sample included 56,877 completed interviews: 27,463 men and 29,414 women; 9563 adolescents and 47,314 adults.
The second administration of the survey included a subsample of 13,130 respondents who agreed to give a saliva sample for DNA analysis, to be used in analyzing the relationship of mental health, addiction, and genetics. This sample was weighted according to selection probability to be representative on the national level. Mental health symptomatology for these individuals was evaluated using the Diagnostic Interview for Psychosis and Affective Disorders (DI-PAD) using the operational criteria for psychotic disorders (OPCRIT v. 4.0), administered by interviewers trained in its use. This scale has shown a high level of consensus among interviewers (kappa = 0.8), with the best consensus for estimations of the presence of disorders during the lifetime of the interviewee [8]. It includes the evaluation of symptomatology of psychosis, depression, anxiety, obsession/compulsion, mania/hypomania, and post-traumatic stress [9]. The analysis also employed data taken from the first stage of the survey related to suicide attempts and pathological gambling [10,11]. As not all of the indicators of every aspect of mental health were included, the data were considered to represent symptoms of different illnesses, not definitive diagnoses.

Religious Affiliation
Respondents were asked if they had a religious affiliation, and if so, which the variable used for the analysis was dichotomous, with 1 = yes (including Catholic, Protestant, Evangelical, Jewish, Christian, or other) and 2 = no.

Socioeconomic Index
An indicator was constructed using socioeconomic data, following the method of Gutiérrez et al. (2015), which considered ownership of goods (home, automobile, computer, DVD player, microwave oven) and access to services (internet, cable, and telephone). The index included five levels based on percentile distribution, classified into three groups: 1 = low and low-medium, 2 = medium and high-medium, and 3 = high [12].

Symptoms of Mania/Hypomania
A positive response was defined as having been diagnosed with bipolar disorder or meeting the following two criteria: (1) having felt unusually happy, irritable, energetic, or hyperactive for three days or more, and (2) not having needed much sleep, without feeling tired, or having more energy than normal.

Psychotic Symptoms
A positive response was defined as having been diagnosed with schizophrenia or meeting the following two criteria: (1) having experienced a period of hearing voices when no one was present or having visions or seeing things that others could not see, and (2) having had ideas or beliefs that others did not share or that turned out not to be true [8,9].

Anxiety Symptoms
A positive response was defined as having had the following three symptoms: (1) having experienced a sudden feeling of anxiety or fear; (2) having, as part of that experience, an accelerated heart rate, chest pains, shortness of breath, a choking sensation, nausea, sweating, weakness, or the fear of going crazy or dying; and (3) having these feelings worsen or intensify in the first ten minutes [8,9].

Depression Symptoms
A positive response was defined as "having ever felt depressed, sad, or discouraged almost every day for two weeks or more" and also meeting one of the following crite-ria: (1) having ever lost most or all interest in normal activities for two weeks or more, (2) having had feelings of uselessness or guilt during that period, or spent a lot of time with thoughts of death, suicide, or self-harm, or (3) having noted significant changes in appetite during that period, or unexpected gain or loss in weight, changes in normal sleep patterns, or difficulty concentrating [8,9].

Obsessive/Compulsive Symptoms
A positive response was defined as having had the following two symptoms: (1) having ever had repetitive thoughts or images, much more exaggerated than ordinary worries, that could not be stopped, and (2) having repeated certain behaviors over and over for an hour or more a day [8,9].

Post-Traumatic Stress Symptoms
A positive response was defined as meeting the following two criteria: (1) having experienced a traumatic event that felt life-threatening, and (2) having experienced strong images or memories of traumatic events that return suddenly, such as uncontrollable thoughts or recurrent nightmares [8,9].

Suicide Attempt
This variable was evaluated with the question: "In the last 12 months, have you tried to take your own life?" The variable was dichotomous, considering that there was a suicide attempt where respondents answered positively.

Pathological Gambling
This variable was defined as recurrent and persistent dysfunctional gambling behavior that causes a clinically significant deterioration or illness, where an individual presents four or more symptoms in the previous 12 months, according to the criteria of the DSM-5 [11,13].
In addition, a variable was created with different groups of interest to this study, based on the presence of certain symptoms or comorbidity. These were:

Comorbidity
The persons in this group met two criteria. The first was that they had at least one of the symptoms of mental illness: mania/hypomania, psychosis, anxiety, depression, obsession/compulsion, post-traumatic stress, suicide attempt, or pathological gambling. The second was that they had an alcohol, drug, or tobacco use disorder (SUD).

Mental Illness Symptomatology
Participants in this group presented positive symptomatology for at least one of the previously mentioned mental health symptoms, but not SUD.

Substance Use Disorder (SUD)
Participants in this group had none of the mental health symptoms described but did indicate having three or more symptoms of SUD during a single year at some point in their lives, based on the DSM-IV-TR [14], either with illegal or prescription drugs or with alcohol. Tobacco use disorder was evaluated using the scale and criteria of Fagerström [15].

Drug Use without SUD
Participants in this group included those who had ever in their lives used a legal drug (alcohol or tobacco), an illegal drug, or an unprescribed medical drug, but without meeting the criteria for SUD and without presenting any mental illness symptomatology.

Group without Mental Illness Symptomatology or Substance Use
This category included participants who said they had never used any legal drug (alcohol or tobacco), illegal drug (marijuana, cocaine, crack, inhalants, hallucinogens, or heroin), or unprescribed medical drug, and who had never presented any of the mental illness symptomatology described in this section.

Ethical Considerations
All the protocols for this study were approved by the Ethics Committees of the Instituto Nacional de Psiquiatría Ramón de la Fuente Muñiz (Approval No. CEI/C/083/2015) and the Instituto Nacional de Medicina Genómica (Approval No. 01/2017/I).

Statistical Analysis
Statistical analyses were carried out using the program STATA, version 13. Estimations were obtained of the prevalence of each symptomatology with an analysis of prevalence ratios (PR) based on generalized linear models (GLM) with log-link and binomial distribution. A multinomial logistic regression model was also carried out, where the dependent variable was that defining the substance use and comorbidity groups and the reference group that without symptomatology or use. These procedures allowed for better predictors and control of an important number of variables.

Sample Characteristics
The majority of the participant population was aged 18-29 (30.8%) or 40-65 (34.9%), and 91% reported a religious affiliation. Almost half had a low or medium-low socioeconomic index (48.5%), and only 12.7% had an undergraduate education or more.

Factors Associated with Comorbidity
A multinomial logistic regression model was carried out to analyze the association of various factors with drug use and comorbidity, using the population that reported no drug use or mental illness symptomatology as the base category (Table 4). With respect to the base category, men have a risk 1.65 times greater than women for using drugs without symptoms of SUD, 18.9 times greater for presenting SUD, a 59% greater risk of presenting any symptom of mental illness, and a risk 10.8 times greater of having a comorbidity. Participants over 18 have a greater risk of presenting the four conditions analyzed than those who are younger, with the highest risk for SUD: 34.27 times greater in those aged [18][19][20][21][22][23][24][25][26][27][28][29]34.75 times greater in those aged 30-39, and 54.36 times greater in those aged 40-65. Those reporting no religion have a 59% greater risk of presenting a symptom of mental illness. Those with junior high school and undergraduate education or more have a risk that is 39% and 72% greater, respectively, of presenting SUD, while for those with high school, the risk is 2.06 times greater than for those with elementary or no education, and their risk for SUD and symptomatology of mental illness is 1.44 and 1.43 times higher, respectively. Those with an undergraduate education or more have a greater risk of comorbidity than those with elementary or no education. Finally, those with a medium-high or high socioeconomic level have a 50% and 49% greater risk, respectively, of presenting without SUD and mental illness symptomatology.

Discussion
Psychiatric or mental health disorders have increased in recent years. Studies have been carried out in Mexico that indicate their prevalence in different populations, such as the general population, adolescents, and university students [2,3,16], but there is little current data. This study offers a view of the problem in urban and rural areas of the country in the population aged 12-65 prior to the COVID-19 pandemic.
Before proceeding to the discussion of the findings, the limitations of this study should be recognized. The results come from screening scales and not from diagnostic instruments; however, the scale been used has shown its consistency in estimating the frequency of mental illness symptoms in the general population and its utility in identifying those symptoms during the lifetime of those who respond [8,9,17]. The Diagnostic Interview for Psychosis and Affective Disorders (DI-PAD) aids mental health symptomatology, and it has been shown to have a good correlation coefficient with psychiatric diagnoses in large samples of psychiatric genetics (9). It is also important to note that the data were obtained from a self-assessment by the persons interviewed, and it is to be expected that a proportion of these persons present symptomatology but have not been diagnosed.
The Survey of Psychiatric Epidemiology (Encuesta de Epidemiología Psiquiátrica, ENEP) showed that 83% of persons with mental health problems sought help once or twice in their lives. However, only 22% received minimally appropriate care for persons with a diagnosis, which includes at least four visits per year to any type of provider, or at least two visits and any type of medication, or to be in treatment at the moment of the interview. This low figure is due mainly to the lack of access to mental health services. The ENEP found an average of 13 years between diagnosis and treatment of major depression, and 30 years for anxiety disorders [18], making clear the limitations of treatment center records and the need for data from national surveys.
The information compiled in this study is highly useful for guiding public policy. The results find that 15.9% of respondents report symptoms of a mental health disorder at some time in their lives. The three most common symptoms are mania/hypomania (7.9%), followed by depression (6.4%), and post-traumatic stress (5.7%). The ENEP found a similar prevalence for depression (7.2%), but a lower prevalence for post-traumatic stress, which is to be expected, given that it was estimated before the increase in homicides and other forms of violence that are associated with this disorder [19]. Persons with SUD show greater comorbidity with mental illness, which is also consistent with what has been reported [6]. Compared to what has been reported in Latin America using the CIDI, our estimates for depression are lower as compared to Sao Paulo Brazil (10%) and higher than those reported in Argentina (3.7%), Colombia (5.3% and Peru (2.7%). Though similar to the rates reported for Mexico in 2003 also using the CIDI (3.7%) [2]. Which is, as mentioned before, expected due the changes in risk factors related to violence and increased poverty in Mexico, since that study [20].
The analysis of risk factors found that women with symptomatology of mental illness more commonly reported having a religion, a lower educational level, and a lower socioeconomic level, as has been reported in other studies [21,22]. Other studies have likewise found an association between religious belief and both unipolar depressive disorders and generalized anxiety [23].
Also consistent with prior studies carried out in Mexico is the finding that men show greater prevalence of substance use and a greater comorbidity between mental disorders and both alcohol and other substance use and pathological gambling. Women report a greater frequency of suicide attempt, a greater prevalence of mental illness symptomatology, and more depression and anxiety than men [2,5,10]. This is the first time the epidemiological frequency of obsessive-compulsive disorder has been analyzed on the national level, and this frequency coincides, as expected, with that described in other populations [17].
Telephone and online surveys carried out during the COVID-19 pandemic lockdown found a growth in the symptomatology of mental illness [24,25]. Though measured with different scales, comparisons with information available after COVID suggest increased anxiety and depression rates. The National Survey on Health and Nutrition 2021 [26] revealed that 10% of the population interviewed reported anxiety and 7.5% depression due to COVID, rates higher than our estimations (3.9% and 6.4%, respectively) before COVID. This increase is expected due to the known impact of COVID on Mental Health. Telephone Interviews reported higher prevalences of these two disorders, reaching a third of the population [25,27]. Data from this study, combined with information obtained during the pandemic, point to the importance of attending to an important problem affecting an important part of the population [28].

Conclusions
The results find that 15.9% of respondents reported symptoms of a mental health disorder at some time in their live. The three most common symptoms were of mania/hypomania (7.9%), followed by depression (6.4%), and post-traumatic stress (5.7%). In addition, people with a high school education had higher risk factors for DUD, SUD, and mental health symptoms comorbid with psychiatric symptoms. Results that despite coming from screening scales, clearly inform us of the mental health of Mexicans.

Limitations
A limitation of the study that we must point out is that the evaluation of psychiatric symptoms was obtained with the DI-PAD screening instrument, which only allows us to know the symptoms and not the diagnosis. So the definition of the comorbidity group is only an approximation and not the exact meaning of it, which is indicated in the Section 2.

Institutional Review Board Statement:
The study was conducted in accordance with the Declaration of Helsinki, and approved by the Ethics Committee of the Instituto Nacional de Psiquiatría Ramón de la Fuente Muñiz and the Instituto Nacional de Medicina Genómica.
Informed Consent Statement: Written informed consent was provided by the participants or their parents or legal guardians.

Conflicts of Interest:
The authors declare no conflict of interest.