Introduction
In 2050, it is estimated that more than 70% of the world’s population will live in an urban environment (Gruebner et al., 2017). By that time, health specialists predict more than a quarter of Americans will have some relationship with mental illness (Galea, Uddin, & Koenen, 2011; NAMI, 2018). While many social and technological developments have increased mental illness diagnoses, such as improved screening and counseling techniques, underlying causes are apparent (Weissman et al., 2017). Past and current research has questioned how environment may link to mental illness. This has brought attention to how one’s level of urbanization—from living in the city to a suburban or rural area—not only impacts one’s health, but also their mental state.
Since the twentieth century, most research focused on how urban environments correlated to mental disorders, failing to study other areas or individual factors such as race and gender. It was classic urban sociologists, such as George Simmel, who proposed living in highly populated areas results in an overall breakdown of social order, claiming that individuals in urban climates begin to quantify, discriminate, and detach themselves from society (Adams, 2005; Simmel, 1903). As a result, city life had been seen as an “economic hub”, suppressing emotional behaviors and individuality. When one’s individuality, creativity, and free-thought is limited, they are consequently more likely to exhibit ruts of depression and overall loss of interest in their lives.
This archaic theory sparked early studies which found concrete data linking urbanization to cognitive distress. For example, in the 1930s, sociologists, Robert E. Lee Faris and H. Warren Dunham, found high rates of schizophrenia among children born in inner-city Chicago (Schmidt, 2007). After similar studies were replicated in other countries, results showed that urban birth raises the baseline risk of schizophrenia by roughly 50% (Schmidt, 2007). Yet, these findings are problematic in that they only measured how one’s space contributed to mental illness, failing to recognize how socioeconomic or external factors, such as crime rate and poverty, may also play a role.
Since Lee and Duham’s study on schizophrenia, more thoughtful studies have been conducted in order to find a link between urbanization and mental illness. Researchers have theorized that the increase in mental illness in cities may be due to underlying socioeconomic disparities and higher crime rates. For example, the urban environment may not be as safe or familiar as the suburban or rural neighborhood, making relationships with neighbors or shop owners less valuable (Schmidt, 2007). As a result, some individuals lose critical social support. Crime and fear of crime, while not mutually exclusive, have been possible factors that link a neighborhood to an individual’s mental health (Schmidt, 2007).
Another argument for why one’s risk for mental illness may be higher in urban areas is because the physical aspects of cities generally contain higher rates of pollution, traffic, and tall buildings which may be perceived as oppressive (Gruebner et al., 2017). Science has shown through animal models that these environmental conditions can impact individual-level experiences, shaping their brain activity and consequent mental states (Galea et al., 2011). In one cognitive study, the psychologist would show the participant images from both rural and urban settings (Galea et al., 2011). Results found that each image activated different regions of the brain, with urban images stimulating the emotional areas of the brain, including the hippocampus and amygdala. From this, further research suggests that brain abnormalities in these regions can even be associated with mental disorders, such as post-traumatic stress disorder (PTSD) (Galea et al., 2011).
Nonetheless, when comparing these findings to studies linking rural climates to mental illness, the results contradict. Research on how rural environments influences one’s risk of mental illness has been underrepresented throughout the past century. In American folklore, the rural environment is seen as a place of tranquility, contrasting the stereotypical “rat-race” city environment. These biases skew mental illness research towards urban environments over rural ones. In actuality, many studies have found mental illness to be extremely prevalent in rural communities. For instance, researchers who studied the population of male farmers found that rates of suicide within farmers are on average higher than the rates for urban individuals (Eberhardt and Pamuk, 2004). Many factors may influence this, such as the physical stresses and hazards of agricultural work (Fraser et al., 2005). Also, many farmers live in close proximity to their workplace, merging their economic stresses with their home lives (Fraser et al., 2005).
A more concerning study linked the children of seasonal farm workers to above average proportions of mental illness. More so, less than half of those with a psychiatric diagnosis had seen a health professional about their mental state (Fraser et al., 2005). This may be because many rural environments lack psychiatric clinics and hospitals, leading to a decrease in mental health diagnosis. Furthermore, the infrastructure in rural climates may not be as developed as its urban and suburban counterparts, often times comprised of dirt roads, cornfields, and heavily forested trails which make transportation for treatment difficult. (Fraser et al., 2005). Moreover, rural populations often exhibit a stoic, hyper-masculine mentality, which creates a stigma behind receiving psychiatric treatment for mental illness (Fraser et al., 2005). Thus, due to cultural differences in many of these communities, resources may not be utilized to their greatest extent even if they are available (Fischer et al., 2005).
Yet, these studies are flawed in that they all lack a comprehensive and holistic view of mental illness in America. In solely studying which environment—rural or urban—poses greater risk of mental illness, past research has failed to consider how intrinsic factors, such as one’s race, ethnicity, and gender, may intersect with one’s level of urbanization and risk of mental illness. When accounting for these variables, the trends become murky. For instance, a study found that African American women who live in urban areas are more likely to develop mood disorders than those in rural areas (Weaver et al., 2015). Yet, in rural areas, non-Hispanic white women were at a higher probability of acquiring mood disorders (Weaver et al., 2015). Similar abnormalities have been presented in Mexican American populations. A team of researchers from California found that Mexican Americans had lower rates of lifetime psychiatric disorders than rates comparable to United States (Ethel et al., 2000). However, there was a higher prevalence of psychiatric disorders reported in Mexican Americans residing in urban environments when compared to small-town and rural areas. While these studies may seem to contradict each other, they introduce us to the idea that environment has a complex relationship with the individual.
Thus, it is crucial to understand America’s mental health dilemma in the context of more nuanced risk factors when examining mental illness in terms of environment. While substantial research has found that urban environments have greater risk for mental disorders, this is an outdated and myopic view of the issue. As stated previously, rural environments also have similar mental health issues, yet due to a lack of research and perpetual stigmas, they are less likely to be diagnosed or understand their illness. More so, few studies have taken into account how individual factors, such as race and gender, influence one’s mental state in both rural environments
In my study, I will take a holistic approach to examine if there is a link between one’s level of urbanization and mental health. While many sociologists focus solely on how urbanization or rural environments impact the mental health of certain subcultures, I will be researching mental illness nationwide, not just within certain populations or regions. This is imperative in that I will be taking a nationwide approach to the nationwide phenomenon of mental illness. In contrast to previous literature, my study will answer questions concerning how individual factors, including race, age, and gender, influence one’s risk of mental illness. With this, I hypothesize that urbanization has less of an effect on mental health than was previously thought. Instead, race, gender, and how both pertain to different social environments may be more plausible factors for mental illness.
Data and Methods
The data on mental illness in youth (12-17) and adult (18+) populations was part of the 2016 National Survey on Drug Use and Health (NSDUH), conducted by the Substance Abuse and Mental Health Services Administration (SAMHSA). The sample design was based on the state, where each state was proportioned based on its population size. In other words, states with higher populations were given a larger sample size than states with smaller populations. Individuals in each sample segment were then mailed screening letters to which they would answer preliminary questions. If accepted, these individuals would then be considered as respondents for the NSDUH interview. In-person interviews were the primary way the NSDUH collected their data, incorporating procedures to increase respondents’ willingness to report honestly about their mental health. For sensitive topics, audio computer-assisted self-interviewing (ACASI) was used. This way, the participant was able to read questions on the computer and then input their response in a private setting.
Participants in the survey were chosen at random. The only requirements were that the participants must be citizens of the United States, over 12 years of age, and residents of a household or homeless shelter. Individuals within military bases, hospitals, jails, as well as the homeless individuals not in shelters were not eligible for the study.
Operationalization of SMI and Urbanization
My dependent variable will be Serious Mental Illness (SMI) present in individuals. According to SAMHSA, a SMI is defined as “having a diagnosable mental, behavioral, or emotional disorder, other than a developmental or substance use disorder” (2017).
Operationalization for SMI varied based on the instruments used. When screening participants, the NSDUH used instruments from the Mental Health Surveillance Study (MHSS) and MHSS-SCID.1 It is important to note that the results from these screenings do not necessarily conclude diagnosable mental illnesses within individuals.
The independent variable for my study is level of urbanization. In order to operationalize urbanization, both “rural” and “urban” environments must first be defined by their population differences. An urbanized area is classified as any location with a population cluster of 50,000 or more (Irwin, 2016). Next, there are “Urban Clusters”, or metropolitan areas, with populations of at least 2,500 people but less than 50,000 people (Irwin, 2016). Thus, whatever was not urban was generally considered rural. Rural areas consisted of low populations, small communities, and high concentrations of farmland (Irwin, 2016).
With this, I will be using large metro, small metro, and non-metro counties as parameters for my study. Large metropolitan areas have a population of 1 million or more. Small metropolitan areas have a population of fewer than 1 million. Non-metropolitan areas include counties that are outside these parameters (SAMHSA, 2017). I will control for age, gender, and race, as these are all factors which may contribute to one’s mental state.
Results
Table 1 (see Table 1, Appendix A) represents descriptive statistics for the individuals in the study. These statistics were broken down into these subgroups: level of urbanization (small, large, non-metro), age (18-25, 26-34, 35+), race (White, Black, Latinx, Asian, Native American, Pacific Islander, Multiracial), gender (male, female), and presence of Serious Mental Illness (SMI).
Of the 41,643 individuals surveyed, 44% were from large metro areas, while 35% were from small metro areas and 24% were from non-metro areas. Thus, it is apparent the largest group of the individuals surveyed were from large metro areas, while non-metro individuals were the smallest group. A similar trend was shown with age groups, where the majority of the participants were in the age range of 35+ (31%) or 18-25 (29%), while only 20% of participants were in the 26-34 range.
The racial-breakdown of the data matched the demographic of the United States, with about 62% of individuals identifying as Caucasian or White, 16% Latinx, 12% Black, and 4% Asian. With regard to gender, there were more females surveyed (53%) than males (47%), yet the disparity between these numbers is not drastic.
More so, only 5% of all participants in the study exhibited a Serious Mental Illness (SMI) in the past year. Of this 5%, Table 2 (see Table 2, Appendix A) examines correlations between various subgroups and SMI frequency. Based on the data, the proportion of individuals exhibiting an SMI in large metro, small metro, and non-metro areas was relatively equal, around 5%. This is equivalent to the national proportion and gives reason to believe there is no correlation between one’s level of urbanization and presence of SMI.
To further this analysis, a linear regression test was conducted. In Model 1 (see Table 3, Model 1, Appendix A), the frequency of Serious Mental Illness (SMI) was compared with the level of urbanization, either large metro, small metro, and non-metro areas. The regression data showed small metro areas exhibited little to no deviation in slope from the constant. Put differently, this confirmed that one’s level of urbanization does not seem to impact their risk of developing a SMI. While non-metro areas also showed no deviation from the slope constant, given the high P value and low sample size, there is not enough data to conclude the latter.
Although there was no conclusive relationship between level of urbanization and SMIs, there was evidence to suggest other factors, such as race and gender, may contribute to risk of mental illness. In Model 2 (see Table 3, Model 2, Appendix A), when controlling for age and gender, interesting results occurred. For example, the data showed that 6% of females in the study exhibited a SMI, whereas only 3% of males did. From this, we understand that females may be twice as likely to develop a SMI when compared to males. This raises questions as to what influence gender may have on mental health. While there are certain external factors at stake that may contribute to this mental health disparity, the difference may also be due to females being more likely to report a SMI than males. Taking into account societal expectations, the man is generally expected to suppress emotion in order to maintain his masculinity. Thus, reporting mental illness within certain male populations may be highly stigmatized, contributing to this SMI gap.
Like gender, race also had an impact on SMI. As shown in Model 2, Black, Latinx, and Asian individuals overall had a lower percentage of SMIs in the past year than White individuals. For Latinx participants, 4% exhibited SMI, while 3% of Black and Asian participants exhibited SMI. Again, these differences may be cultural, which could make some racial/ethnic groups less likely to report SMI than others. A contradiction to this would be the results from the multiracial population, which exhibited the greatest proportion of SMI in study at around 7%. While little research is done surrounding multiracial individuals and mental health, it may be worth looking into given these findings.
Conclusion
Unlike studies showing dramatic mental health disparities between rural and urban environments, I did not find any striking differences between urbanization and risk of mental illness. My results went against historic literature that attributed “urban life” as a risk factor for psychological illness (Adams, 2005). My results also went against modern arguments which center on the rural and urban environment and its impact on mental illness. While some sociologists claim rural individuals are more prone to mental illness due to social isolation and lack of resources, my study found no differences in SMI between individuals from rural and urban areas (Adams, 2005; Simmel, 1903). While past literature found trends in city-living and mental disorders and attributed this to high rates of pollution and/or lack of greenery, my results did not support this relationship either (Weaver et al., 2015).
Instead, my study found that race, ethnicity, and gender are more plausible risk factors for mental illness. This relates to past literature which has taken a more nuanced approach to how environment and mental health influence each other. For instance, researchers in California found that individuals in Mexican American populations had a lower risk of lifetime illness than other individuals in the United States, despite living in urban or rural areas. In my research, I also found that Latinx, Black, Asian, and other minority populations were less likely to report an SMI compared to the white majority (Ethel et al., 2000).
My study also paralleled past literature which controlled for race and gender. In a study on mood disorders, it was found that African American women in urban environments were more likely to exhibit symptoms compared to white women in these environments (Weaver et al., 2015). Yet, the opposite trend was found for white women in rural environments in comparison to black women in rural environments (Weaver et al., 2015). My research supported these contradicting results since it showed that environmentally independent factors, such as race and gender, impact one’s risk of mental illness.
Like all research, my study had many limitations. For one, the population sampled for non-metro individuals was significantly smaller than for large and small metro individuals (see Table 2b). More data on non-metro individuals is necessary to give a comprehensive view of the “rural” population. Another limitation was the time of the study. While the study was completed in 2016, participants were surveyed in 2014. Since 2014, America’s political and economic climate has drastically changed, along with the increasing presence of technology and social media. These all serve as risk factors for SMIs, and since they are not reflected in the 2016 study, the results do not accurately represent present-day America. Also, confounding variables were rampant in the study. In contrast to a longitudinal study, which involves multiple measures over an extended period of time, my study was cross-sectional and only measured the population at one specific point in time. Because of this, certain subgroups studied may have been affected by cohort differences that arise from the particular experiences unique to that group. For instance, some older participants studied may have experienced a traumatic historical event in their upbringing, such as the Vietnam War, influencing their risk of developing a SMI. More broadly, some confounding variables, such as gender and cultural differences, as well as religious and other internal factors, may also influence one’s likelihood to report an SMI.
Although no relationship was shown between level of urbanization and serious mental illness, this research did raise questions as to how one’s race and gender affects their mental health. While my study only touched on some of the variables that affect one’s risk of serious mental illness, more research concerning these variables is crucial in order to present a valid relationship between race, gender, and mental health. Perhaps analyzing how different socio-economic aspects of one’s environment, such as how poverty and crime rate relate to race, gender, and mental illness, would be beneficial for future studies. Also, incorporating other intrinsic factors, such as sexuality, may also link to environment and mental illness.
Going forward, it is important for healthcare providers, school districts, and mental health organizations to recognize mental illness as a fluid matter caused by a multitude of factors. When screening for mental illness in children, it would be beneficial to take into account race, gender, and the environment in which they grow up. In taking a more holistic stance on the puzzle of mental illness, we can continue to eliminate the stigma. More so, we can seek new, individualized ways to treat mental illness in different social environments.
Appendix A
Descriptive Statistics
| Proportion | |
| Dependent Variable | |
| 0.05 | |
| Independent Variable | |
| Urbanicity | |
| Large Metro | 0.44 |
| Small Metro | 0.35 |
| Non-Metro | 0.21 |
| Controls | |
| Female | 0.53 |
| Age | |
| 18-25 years old | 0.31 |
| 26-34 years old | 0.20 |
| 35 years old and up | 0.29 |
| Race | |
| White | 0.62 |
| Black | 0.12 |
| Latinx | 0.16 |
| Asian | 0.04 |
| Native American | 0.02 |
| Pacific Islanders | 0.01 |
| Multiple Racial | 0.03 |
| Number of Obersvations | 41,643 |
Binary Comparisons
| Had a Serious Mental Illness | No Serious Mental Illness | |
| Independent Variable | ||
| Urbanicity | ||
| Large Metro | 0.05 | 0.95 |
| Small Metro | 0.05 | 0.95 |
| Non-Metro | 0.05 | 0.95 |
| Controls | ||
| Female | 0.06 | 0.94 |
| Male | 0.03 | 0.97 |
| Age | ||
| 18-25 years old | 0.05 | 0.95 |
| 26-35 years old | 0.05 | 0.95 |
| 35 years old and up | 0.05 | 0.95 |
| Race | ||
| White | 0.05 | 0.95 |
| Black | 0.03 | 0.97 |
| Latinx | 0.04 | 0.96 |
| Asian | 0.03 | 0.97 |
| Native American | 0.04 | 0.96 |
| Pacific Islanders | 0.06 | 0.94 |
| Multiple Racial | 0.07 | 0.93 |
| Number of Observations | 41,643 |
Table 3
Coefficients Predicting Serious Mental Illness in Past Year
| Model 1 | Model 2 | |
| Independent Variable | ||
| Urbanicity | ||
| Small Metro | 0.00 (0.00) | -0.00 (0.00) |
| Non-Metro | 0.01 (0.00) | 0.00 (0.00) |
| Controls | ||
| Female | 0.03 (0.00)* | |
| Age | ||
| 26-34 years old | 0.00 (0.00) | |
| 35 years old and up | -0.01 (0.00) | |
| Race | ||
| Black | -0.02 (0.00)* | |
| Latinx | -0.02 (0.00)* | |
| Asian | -0.03 (0.01)* | |
| Native American | -0.01 (0.01) | |
| Pacific Islanders | -0.01 (0.01) | |
| Multiple Racial | 0.02 (0.01)* | |
| Constant | 0.05 (0.00) | 0.01 (0.00) |
* p<0.05 = representative of overall population
References
Adams, Richard E. (1992). Is Happiness A Home in the Suburbs?: The Influence of Urban Versus Suburban Neighborhoods on Psychological Health”. Journal of Community Psychology, 20(4), 353-372.
Alderete, Ethel, William Armando Vega, Bohdan Kolody, and Sergio Aguilar-Gaxiola. (2000). Lifetime prevalence of and risk factors for psychiatric disorders among Mexican migrant farmworkers in California. American Journal of Public Health, 90(4): 608–614.
Eberhardt, Mark S. and Elsie R. Pamuk. 2004. “The Importance of Place of Residence: Examining Health in Rural and Nonrural Areas”. American Journal of Public Health 94(10): 1682-1686.
Fraser, C.E., K.B. Smith, F. Judd, J.S. Humphreys, & L.J. Fragar. (2005). Farming and Mental Health Problems and Mental Illness. Henderson International Journal of Social Psychiatry, 51(4): 340–349. Doi: 10.1177/0020764005060844
Galea, Sandro, Monica Uddin, and Karestan Koenen. (2011). The urban environment and mental disorders: Epigenetic links. Epigenetics, 6(4): 400–404. doi: 10.4161/epi.6.4.14944
Gruebner, Oliver, Michael A. Rapp, Mazda Adli, Ulrike Kluge, Sandro Galea, and Andreas Heinz. (2017). Cities and Mental Health. Deutsches Ärzteblatt International, 114(8): 121–127. doi: 10.3238/arztebl.2017.0121
Irwin, Ellen. (2016). The Urban-Suburban-Exurban-Rural Continuum: Definitions, Trends and Interdependencies. Ohio State University. Retrieved Oct. 5
National Alliance on Mental Illness (NAMI). (2012). Mental Health By The Numbers. Retrieved from https://www.nami.org/Learn-More/Mental-Health-By-the-Numbers.
Schmidt, Charles W. (2007). Environmental Connections: A Deeper Look into Mental Illness. Environmental Health Perspective, 115(8): A404–A410.
Simmel, Georg. (1903). The Metropolis and Mental Life. Malden, MA: Blackwell Publishing.
Substance Abuse and Mental Health Services Administration (SAMHSA). (2016). National Survey on Drug Use and Health 2016 (NSDUH-2016-DS0001). Retrieved from https://datafiles.samhsa.gov/study-dataset/national-survey-drug-use-and-... 6-ds0001-nid17185.
Substance Abuse and Mental Health Services Administration (SAMHSA). (2017). 2017 Methodological Summary and Definitions. Retrieved from https://www.samhsa.gov/data/report/2017-nsduh-detailed-tables.
Weaver, Addie, Joseph A. Himle, Robert Joseph Taylor, Niki N. Matusko, and Jamie M. Abelson. (2015). Urban vs Rural Residence and the Prevalence of Depression and Mood Disorder Among African American Women and Non-Hispanic White Women. JAMA Psychiatry, 72(6): 576–583. doi: 10.1001/jamapsychiatry.2015.10
Weissman, Judith, David Russell, Melanie Jay, Jeannette M. Beasley, Dolores Malaspina, and Cheryl Pegus. (2017). Disparities in Health Care Utilization and Functional Limitations Among Adults With Serious Psychological Distress, 2006–2014. Psychiatric Services, 68(7): 653-659
Footnotes
1 Structured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders–Fourth Edition–Research Version–Axis I Disorders (MHSS-SCID), which is based on the 4th edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV).