Introduction and Background
When it comes to taking care of ourselves and our health, we need to take more than just our physical health into consideration. While eating healthy, remaining active, and going to the doctor regularly are all important and expected aspects of keeping our health in check, it’s also important not to overlook our mental health. According to the CDC, mental health is described as including our “emotional, psychological, and social well-being”. Not only can poor mental health lead to mental illnesses, but they can also manifest in physical ways if left untreated, such as heart disease and type 2 diabetes.
Taking care of our minds is just as crucial as our bodies, but it’s not as easy to seek and receive help compared to a physical health problem. A key issue surrounding this is the stigma associated with mental health and mental illness. It can be hard for those suffering to feel understood or like their struggles are valid. This can lead to reluctance to reach out for support and get the treatment they need to improve their quality of life. Alongside the stigma and judgement many feel surrounding their mental health comes the issue of access to care. If one does feel ready to seek treatment, do they have the resources in their area to do so? Without the proper care, mental illnesses can severely impact one’s life, ranging from disruptions in their day to day life to contemplating or even going through with committing suicide. With over 200 forms of mental illnesses, this public health crisis is not to be taken lightly. This then leads into this report and its aim to delve deeper into the topic of mental health, suicide, and access to care among the U.S. and any differnces that may exist between states.
Goals of this Report
This report aims to delve into a few different issues surrounding mental health and access to care in the United States. Several datasets will be looked at in order to make comparisons between a few different pieces of information regarding mental health. The data I will look at concerns the following:
- The percentage of adults per state who reported having 14 more poor mental health days over a span of 30 days, also referred to as “frequent mental distress”.
- The percentage of adults per state with a mental illness.
- Suicide rates by state.
- The number of mental health providers in each state per 100k residents.
- The Google Trends search frequency of the phrase “therapy near me” over the last 5 years to gauge overall interest in seeking treatment for mental health concerns.
Most of this data will span 5 years, depending on how much is available and when the most recent pieces of information were gathered and made available. Each area of data will be broken down into sections in this report and comparisons will be made where appropriate. Details on each dataset and methods will be provided in each individual section. The order of this analysis is set up in a way to go from reporting poor mental health, to those suffering from a mental illness, to those who suffer to such a great deal that they take their own lives. Ultimately, I want to compare this data to how available providers and treatment are in each state, and whether that impacts the different rates discussed in earlier sections. Does this number of providers reflect interest in mental health treatment? Is there higher interest in states with poor mental health? Or maybe not enough interest in states that could probably use the support, indicating a potential cultural impact on openness regarding mental illness and stigma associated with seeking help. These are a few of the questions guiding this report. The key takeaway I aim to get at is how are all of these factors related, if at all?
Setting up the Data:
The following analysis will all be done in R and Tableau. To begin in R, I will load the necessary packages and the workspace where I am pulling the data from. From here, I will delve into the first section which will look at the prevalence of frequent mental ditress in adults.
Prevalence of Mental Distress
The data for this section was obtained through America’s Health Rankings who cite their data coming from the CDC’s Behavioral Risk Factor Surveillance System. More than 400,000 adults are contacted via telephone to complete the largest continuously conducted health related survey in the world. The data spans all 50 states and The District of Columbia and includes questions related to risk behaviors, chronic health conditions, and use of preventive services. With mental health being a relatively new area of concern, data regarding frequent mental distress did not start to be collected in this survey until 2016. The most recent available report is 2019.
The definition of frequent mental distress is having 14 or more days over the span of 30 days of reported poor mental health days, or days where one’s mental state makes it difficult to go about their typical day-to-day life. According to studies, this 14 day period is used because there is a strong correlation between this duration of time and those who have or go on to have clinically diagnosable mental disorders.
The inclusion of this measure in this report is to provide a baseline for those in the U.S. who suffer from frequent poor mental health. This does not necessarily mean that these individuals have a diagnosable mental illness, but it is often a good indicator of one or developing one in the future. This will also lend itself to be useful information later on in this report to determine if there is a relationship between this rate and the rate at which people in each state are searching online for treatment or what kind of support is available in their area. It is possible those reporting 14 days of mental distress already have a mental illness, but do not have the resources to receive an official diagnosis. The visualizations in this section will explore this measure further.
mean_md_2016<-mean(mental_distress$Percentage[mental_distress$Year==2016])
mean_md_2017<-mean(mental_distress$Percentage[mental_distress$Year==2017])
mean_md_2018<-mean(mental_distress$Percentage[mental_distress$Year==2018])
mean_md_2019<-mean(mental_distress$Percentage[mental_distress$Year==2019])
mean_md_overall<-data.frame("Year"= c("2016","2017","2018","2019"),
"Rate"=c(mean_md_2016,mean_md_2017,mean_md_2018,mean_md_2019))
mean_md_overall %>%
ggplot(aes(Year, Rate))+geom_col(fill='#C67D70')+ theme_pomological()+labs(title="Average Rate of Adults Experiencing Poor Mental Health Days",subtitle="Percentage of adults reporting 14 days of mental distress over a 1 month period")
Based on what we see above, there has been a slow yet steady increase in the average rate of adults experiencing poor mental health across the U.S. as a whole over the last four years. In 2016, this rate was at 11.46 and increased to 12.95 by 2019. While not a largely significant difference, it is still worth noting that the mental health of U.S. adults has not been getting better. This would be particularly interesting data to look at for 2020 and consider the impact COVID-19 may have had on the mental well being of many individuals. While we don’t have that data yet, we can’t say for sure, but over the last few years, it would appear that poor mental health may be on the rise. We do have data, though, on each individual state. The visualization below shows the percent of adults on a state-by-state basis that reported frequent mental distress. This can be looked at from the years 2016 to 2019.
Over the years, some states improve, some get worse, while some remain the same. In 2016, the states in the South Eastern portion of the country appear to have the highest rates of mental distress with most rates in the double digits. This looks to be the only clear pattern we can discern from 2016. In the following years, nearly all states report higher rates than they did the previous year. One state that breaks this trend is Nevada, which had a rate of 14.2 in 2017 that dropped to 11.7 and back up to 13.1 in 2018 and 2019, respectively. While the South East remains steady in holding some of the highest rates in the country for mental distress, many other states are not far behind as the years progress. One state does stand out the most, though, as being significantly lighter on the map than other states. South Dakota consistently reports the lowest rates of mental distress, save for 2018 where Minnesota took its spot. The following table displays the states with the highest and lowest rates of mental distress per year.
worst_2016<-(mental_distress %>%
filter(Year == 2016) %>%
arrange(-Percentage) %>%
head(1))$State
worst_2017<-(mental_distress %>%
filter(Year == 2017) %>%
arrange(-Percentage) %>%
head(1))$State
worst_2018<-(mental_distress %>%
filter(Year == 2018) %>%
arrange(-Percentage) %>%
head(1))$State
worst_2019<-(mental_distress %>%
filter(Year == 2019) %>%
arrange(-Percentage) %>%
head(1))$State
top_2016<-(mental_distress %>%
filter(Year == 2016) %>%
arrange(Percentage) %>%
head(1))$State
top_2017<-(mental_distress %>%
filter(Year == 2017) %>%
arrange(Percentage) %>%
head(1))$State
top_2018<-(mental_distress %>%
filter(Year == 2018) %>%
arrange(Percentage) %>%
head(1))$State
top_2019<-(mental_distress %>%
filter(Year == 2019) %>%
arrange(Percentage) %>%
head(1))$State
best_and_worst<-data.frame("Year" = c("2016","2017","2018","2019"),
"Highest"=c(worst_2016,worst_2017,worst_2018,worst_2019),
"Lowest"=c(top_2016,top_2017,top_2018,top_2019))
best_and_worst %>%
kbl() %>%
kable_material(c("striped", "hover"))| Year | Highest | Lowest |
|---|---|---|
| 2016 | West Virginia | South Dakota |
| 2017 | West Virginia | South Dakota |
| 2018 | Arkansas | Minnesota |
| 2019 | West Virginia | South Dakota |
Apart from 2018, West Virginia and South Dakota remain steady in their ranking as states with either the highest percentage of adults experiencing frequent mental distress or the lowest. The following sections will look at how these rates could potentially relate to other mental health measures and indicators and whether we see a pattern emerge for certain states, especially West Virginia and South Dakota.
Prevalence of Mental Illness
Not all individuals who suffer from frequent mental distress have also been diagnosed with or suffer from a mental illness. Though, experiencing poor mental health for 14 out of 30 days is a good indicator that there could be something more serious going on. Again, why this may not always be the case, as a person could just be experiencing a particularly difficult stretch of time in their life, for example, but it is a good starting point and segue into discussing mental illness. Most sufferers from these illnesses could be lumped into the category of those experiencing frequent mental distress, or that is where signs and symptoms first began to show, so this next section will delve into the rates of mental illnesses in adults across the U.S.
The data for this section was gathered from the annual reports created by Mental Health America on the State of Mental Health in America. Data was gathered from these reports for the years 2017 to 2020. According to MHA, they cite The Substance Abuse and Mental Health Services Administration (SAMHSA) as to how they record an instance for an adult with any mental illness. SAMHSA reports that an adult who suffers from any mental illness (AMI) is one is “aged 18 or older who currently or at any time in the past year have had a diagnosable mental, behavioral, or emotional disorder” that meets the criteria of the DSM-IV, “regardless of the level of impairment in carrying out major life activities”. They also include that AMI includes mild, moderate, and severe mental illnesses. MHA used this definition in their reports to determine the percentage of adults in each state who meet this criteria.
The first visualization below looks at the average rate of mental illness for the U.S. as a whole from 2017-2020.
mean_ami_2017<-mean(adult_ami$Rate[adult_ami$Year==2017])
mean_ami_2018<-mean(adult_ami$Rate[adult_ami$Year==2018])
mean_ami_2019<-mean(adult_ami$Rate[adult_ami$Year==2019])
mean_ami_2020<-mean(adult_ami$Rate[adult_ami$Year==2020])
mean_ami_overall<-data.frame("Year"= c("2017","2018","2019","2020"),
"Rate"=c(mean_ami_2017,mean_ami_2018,mean_ami_2019,mean_ami_2020))
mean_ami_overall %>%
ggplot(aes(Year, Rate))+geom_col(fill='#C67D70')+ theme_pomological()+
labs(title="Average Rate of Adults Experiencing Mental Illness",
subtitle="Percentage of adults in the U.S. suffering from a mental illness")
Looking at the overall trend of the U.S. over the last 4 years, there has not been much change in the number of adults dealing with a mental illness. From 2017 to 2020, there has been about a 1% increase from 18.9% to 19.9%. While rates for the country as a whole remain relatively stable, what I am more interested in is looking at each state individually to compare these rates to the others looked at in this report.
The following map displays the percentage of adults in each state with AMI for the years 2017-2020.
There are not any discernible patterns we can tell by looking at the map, but it is interesting to note that from 2017-2018, there was an overall decrease in the rate as can be seen with the map as a whole getting a lighter shade of blue. In 2019 and 2020, though, we see a more distinct shift where most states report higher percentages of adults with AMI. Certain states also become more prominent compared to others as the years progress. For example, in 2017 Oregon is the noticeable standout with a rate of 22.66. In 2018, states like New Hampshire and Kentucky start seeing an increase. In 2019 and 2020, some of these states follow this trend as others join as well, such as Idaho, Utah, and Alabama. To get an idea of the states with the highest and lowest rates of mental illness, I created the table below.
worst_ami_2017<-(adult_ami %>%
filter(Year == 2017) %>%
arrange(-Rate) %>%
head(1))$State
worst_ami_2018<-(adult_ami %>%
filter(Year == 2018) %>%
arrange(-Rate) %>%
head(1))$State
worst_ami_2019<-(adult_ami %>%
filter(Year == 2019) %>%
arrange(-Rate) %>%
head(1))$State
worst_ami_2020<-(adult_ami %>%
filter(Year == 2020) %>%
arrange(-Rate) %>%
head(1))$State
top_ami_2017<-(adult_ami %>%
filter(Year == 2017) %>%
arrange(Rate) %>%
head(1))$State
top_ami_2018<-(adult_ami %>%
filter(Year == 2018) %>%
arrange(Rate) %>%
head(1))$State
top_ami_2019<-(adult_ami %>%
filter(Year == 2019) %>%
arrange(Rate) %>%
head(1))$State
top_ami_2020<-(adult_ami%>%
filter(Year == 2020) %>%
arrange(Rate) %>%
head(1))$State
best_and_worst_ami<-data.frame("Year" = c("2017","2018","2019","2020"),
"Highest"=c(worst_ami_2017,worst_ami_2018,worst_ami_2019,worst_ami_2020),
"Lowest"=c(top_ami_2017,top_ami_2018,top_ami_2019,top_ami_2020))
best_and_worst_ami %>%
kbl() %>%
kable_material(c("striped", "hover"))| Year | Highest | Lowest |
|---|---|---|
| 2017 | Oregon | Florida |
| 2018 | New Hampshire | Hawaii |
| 2019 | Oregon | New Jersey |
| 2020 | Utah | New Jersey |
Oregon and New Jersey are the two states that appear twice in this table, indicating that Oregon typically has had the highest rates of mental illness in the last 4 years with New Jersey on the lower end. How do these states compare to the ones that ranked the highest and lowest for mental distress rates? The next two tables show the states who have the highest rates of mental illness and mental distress followed by the states with the lowest rates of these two measures, respectively. Only the years 2017-2019 are shown, as those are the three years where the data overlaps for these particular measures.
data.frame("Year"=c("2017","2018","2019"),
"Highest Rate AMI" = c(worst_ami_2017,worst_ami_2018,worst_ami_2019),
"Value High AMI"=c(22.7,21.7,22.6),
"Highest Rate MD"= c(worst_2017,worst_2018,worst_2019),
"Value High MD"=c(16.5,17.3,18.9),check.names = FALSE) %>%
kbl() %>%
kable_material(c("striped", "hover"))| Year | Highest Rate AMI | Value High AMI | Highest Rate MD | Value High MD |
|---|---|---|---|---|
| 2017 | Oregon | 22.7 | West Virginia | 16.5 |
| 2018 | New Hampshire | 21.7 | Arkansas | 17.3 |
| 2019 | Oregon | 22.6 | West Virginia | 18.9 |
data.frame("Year"=c("2017","2018","2019"),
"Lowest Rate AMI"=c(top_ami_2017,top_ami_2018,top_ami_2019),
"Value Low AMI"=c(16.0,15.9,15.5),
"Lowest Rate MD"=c(top_2017,top_2018,top_2019),
"Value Low MD"=c(8.3,9.2,9.3),check.names = FALSE)%>%
kbl() %>%
kable_material(c("striped", "hover")) | Year | Lowest Rate AMI | Value Low AMI | Lowest Rate MD | Value Low MD |
|---|---|---|---|---|
| 2017 | Florida | 16.0 | South Dakota | 8.3 |
| 2018 | Hawaii | 15.9 | Minnesota | 9.2 |
| 2019 | New Jersey | 15.5 | South Dakota | 9.3 |
Looking at the states with the highest rates in either category, it’s interesting to note that there is no overlap. The states with the highest rates of mental illness do not have the highest rates of frequent mental distress. The same can be said for the states with the lowest rates. Furthermore, the rates of mental illness are higher overall, both on the low and high end, when compared to that of mental distress. Though, as mentioned earlier, those suffering from frequent mental distress may also fall into the category of having a diagnosed mental illness, therefore contributing to both measures. Additional visualizations are provided below, which give more of a look into the comparison between these two categories.
The map above shows a side by side view of the maps for each measure. Just by looking at it at first glance, the visualization on the right, which displays rates of mental illness, is a darker shade of blue, indicating a higher overall rate for AMI than those for frequent mental distress. Also, as the years progress, there is less of a change in rates for AMI than there is for experiencing mental distress. This lesser rate of change could be attributed to the fact that mental illnesses are long term, whereas mental distress could or could not affect an individual for an extended period of time. Therefore, if an individual is diagnosed with a mental illness, chances are they will have that same diagnosis a year or two later. What we don’t see is much of a relationship between the two measures. A state with a high rate for mental distress does not necessarily mean a high rate for mental illness and vice versa. This holds true in some states, but not all, so it’s difficult to make a general statement on this comparison. This leads into the last comparison I want to make in this section in regards to the best and worst ranked states from each category.
This chart displays the nine states from the table above who ranked either the top or lowest for rates of either mental illness or mental distress. Red indicates a state with high rates in either category, while green represents low rates. For example, New Jersey (green) had a low rate of mental illness and West Virginia (red) had a high rate for mental distress. The pink dot on each bar is the rate of frequent mental distress and the height of the bar is percent of adults with a mental illness. By looking at this, it’s difficult to draw any substantial conclusions seeing as some states still have higher rates of mental illness despite lower rates of mental distress. Minnesota in 2017, for example, had similar AMI rates to New Hampshire, but a lower rate of frequent mental distress. While most of the states who fall into the red, low ranking category also experience slightly higher mental distress rates, this difference is not drastic nor sizable enough to say with any certainty that one guarantees the other.
Suicide Rates
Those fighting a mental illness may struggle with thoughts of harming themselves or ending their lives. Unfortunately, suicide is a leading cause of death in the U.S. In 2018, suicide was the tenth leading cause of death overall, taking 48,000 lives. From 1999 to 2018, there has been a 35% increase in the total suicide rate, further emphasizing the importance of this public health issue. Most people who attempt to take their own lives suffer from a mental illness, with 30%-70% of victims suffering from depression or bipolar disorder. While suicide itself is not a mental disorder, it is closely linked to and the most tragic outcome of several debilitating mental illnesses. For this reason, the following section will take a look at the trend of suicide rates throughout the country and compare those to rates of mental illness. The aim of this portion is to see the connection, if any, between states with high or low rates of suicide and their corresponding rates of mental illness.
As with previous measures, the first thing I want to take a look at is overall suicide rates for the U.S. Data for this section was obtained from the CDC, where data is available for 2005 and then for the years 2014-2015.Rates represent deaths per 100,000 population and are age-adjusted. While this report is primarily focused on the last five years, I will include 2005 and 2014 in the overall average graph just to get a sense of how rates are trending. For the rest of the analysis in this section, I will focus on the years 2015-2018 for cohesiveness of the report and ease of comparison to other measures.
mean_s_2005<-mean(suicide$Rate[suicide$Year==2005])
mean_s_2014<-mean(suicide$Rate[suicide$Year==2014])
mean_s_2015<-mean(suicide$Rate[suicide$Year==2015])
mean_s_2016<-mean(suicide$Rate[suicide$Year==2016])
mean_s_2017<-mean(suicide$Rate[suicide$Year==2017])
mean_s_2018<-mean(suicide$Rate[suicide$Year==2018])
mean_s_overall<-data.frame("Year"= c("2005","2014","2015","2016","2017","2018"),
"Rate"=c(mean_s_2005,mean_s_2014,mean_s_2015,mean_s_2018,mean_s_2017,mean_s_2018))
mean_s_overall %>%
ggplot(aes(Year, Rate))+geom_col(fill='#C67D70')+ theme_pomological()+
labs(title="Average Overall Suicide Rates",
subtitle="Rate of deaths by suicide in the U.S. from 2005 and 2014-2018")
While we don’t have data for 2019 or 2020, it is safe to say that there has been an increase in suicide rates over the last 5 to 6 years when compared to that of 2005. There appears to be a steady upward trend of rates, with a slight decrease in 2018. This paints a picture for the U.S. as whole, but what would be better suited for comparison is to look at each state.
The trend appears to be that more rural, less populated states in the western part of the country experience the highest rates of suicide. States such as Alaska, Idaho, Montana, and Wyoming remain steady in these high rates for all 4 years. Other states see an increase ranging from a small to large magnitude over this time span. Some states on the eastern coast, such as West Virginia and New Hampshire, also experience high rates, who stand out the most among their neighboring states by 2018.
The following table shows the states with the highest and lowest rates of suicide for the years 2015-2018.
worst_s_2015<-(suicide %>%
filter(Year == 2015) %>%
arrange(-Rate) %>%
head(1))$State
worst_s_2016<-(suicide %>%
filter(Year == 2016) %>%
arrange(-Rate) %>%
head(1))$State
worst_s_2017<-(suicide %>%
filter(Year == 2017) %>%
arrange(-Rate) %>%
head(1))$State
worst_s_2018<-(suicide %>%
filter(Year == 2018) %>%
arrange(-Rate) %>%
head(1))$State
top_s_2015<-(suicide %>%
filter(Year == 2015) %>%
arrange(Rate) %>%
head(1))$State
top_s_2016<-(suicide %>%
filter(Year == 2016) %>%
arrange(Rate) %>%
head(1))$State
top_s_2017<-(suicide %>%
filter(Year == 2017) %>%
arrange(Rate) %>%
head(1))$State
top_s_2018<-(suicide%>%
filter(Year == 2018) %>%
arrange(Rate) %>%
head(1))$State
best_and_worst_s<-data.frame("Year" = c("2015","2016","2017","2018"),
"Highest"=c(worst_s_2015,worst_s_2016,worst_s_2017,worst_s_2018),
"Lowest"=c(top_s_2016,top_s_2016,top_s_2017,top_s_2018))
best_and_worst_s %>%
kbl() %>%
kable_material(c("striped", "hover"))| Year | Highest | Lowest |
|---|---|---|
| 2015 | WY | NJ |
| 2016 | MT | NJ |
| 2017 | MT | NY |
| 2018 | WY | NJ |
The same four states appear in these rankings. Montana and Wyoming have the highest suicide rates while New York and New Jersey have the lowest. Looking at these rates on their own is interesting given that New York and New Jersey are highly populated states when compared to Wyoming and Montana. So these differences in rates could be attributed to demographic factors that differ among rural vs. non-rural areas. This will be explored further in the conclusion of this report.
Next, I want to compare suicide rates to rates of mental illness, since the two are so closely tied.
The two maps above show the comparison for each state’s mental illness rate and suicide rate for the two years the data overlaps, 2017 and 2018. Less states stand out in the left than on the right, but looking closer, states with comparatively high suicide rates also have hgher mental illness rates. The visualizations below look into this further.
This graph shows the states with the highest suicide rates along with their respective rates of mental illness. Typically, suicide rates far exceed those of mental illness, which is interesting but could be explained in the data collection process. Suicide is a rate a lot easier to measure than whether or not someone has a mental illness. Nevertheless, states with high suicide rates have AMI rates that fall into the high teens-low twenties range. The next visualziation shows the states with the lowest suicide rates.
AMI rates for these states fall in the same general vicinity as those prior, but with far lower rates of suicide. Though taking a look at the states in each graph shows a clear differnce- many of the states with lower suicide rates are less rural and more populated than those with higher rates. So while mental illness rates are higher, it’s possible that these states have more resources and support available for those suffering, allowing individuals to seek support before choosing to end their life. Again, demographic differences could play into this disparity, which will be discussed later as a potential key factor. The next section will explore this idea of mental health support and how it may play a role in mental illness.
Mental Health Providers
Treatment for those suffering with a mental illness varies depending on the nature and severity of the disorder, but the importance of treatment is the same regardless. Therapy is most often the form of support people seek when trying to heal from a mental illness, but several other methods exist as well. Whether it’s an outpatient provider, inpatient treatment, or hospitalization, proper care is crucial when dealing with a mental illness. Many suicide victims never saw a mental health professional and lack of access to this treatment is one of the main risk factors for suicide. Not only can untreated mental illness lead to suicide, but can take a toll on ones physical health as well. Disorders such as depression and anxiety put the body under a lot of stress and trauma, which can have long lasting effects. Other illnesses can be more direct, such as eating disorders, that can lead to malnourishment, osteoporosis, and heart failure. Therefore, seeking out a mental health professional (or team) is imperative to overcoming these illnesses. Unfortunately, though, this is not always accessible for many Americans.
The data for this section was obtained from America’s Health Rankings, who compiled their data based on reports from the U.S. HSS, the Census Bureau, and other sources. The resulting dataset consists of the number of psychiatrists, psychologists, licensed clinical social workers, counselors, marriage and family therapists, providers that treat alcohol and other drug abuse as well as advanced practice nurses specializing in mental health care per 100k residents in each state. Data is available from 2017 to 2019. To begin this analysis, I want to first look at how much the U.S. has seen an overall increase (or decrease) in the number of mental health providers available. To do so, I created the same graph as for the other measures, showing the average number of professionals per 100k for the years 2017-2019.
mean_p_2017 <- mean(providers_per100k$Per100k[providers_per100k$Year == 2017])
mean_p_2018 <- mean(providers_per100k$Per100k[providers_per100k$Year == 2018])
mean_p_2019 <- mean(providers_per100k$Per100k[providers_per100k$Year == 2019])
mean_p_overall <- data.frame(Year = c("2017", "2018", "2019"), Providers = c(mean_p_2017,
mean_p_2018, mean_p_2019))
mean_p_overall %>% ggplot(aes(Year, Providers)) + geom_col(fill = "#C67D70") + theme_pomological() +
labs(title = "Average Number of Mental Health Professionals per 100k People",
subtitle = "Years 2017-2019")
Since 2017, the U.S has seen an overall increase in the number of mental health professionals available. From an average of 236 in 2017 to 265 in 2019, there’s a good chance this trend will continue in future years, especially as mental health and seeking treatment become less stigmatized topics. While this may be the case for the U.S. as a whole, each state could be telling a different story. The map below shows the average number of providers per 100k per state for 2017-2019.
Over the course of the three years, there is not much of a shift in the availability in any individual state. The south and the midwest appear to be the states with the most lacking resources as the northeast, Oregon, and even Oklahoma have the most. The table below shows the states with the highest versus lowest number of providers.
worst_p_2017 <- (providers_per100k %>% filter(Year == 2017) %>% arrange(-Per100k) %>%
head(1))$State
worst_p_2018 <- (providers_per100k %>% filter(Year == 2018) %>% arrange(-Per100k) %>%
head(1))$State
worst_p_2019 <- (providers_per100k %>% filter(Year == 2019) %>% arrange(-Per100k) %>%
head(1))$State
top_p_2017 <- (providers_per100k %>% filter(Year == 2017) %>% arrange(Per100k) %>%
head(1))$State
top_p_2018 <- (providers_per100k %>% filter(Year == 2018) %>% arrange(Per100k) %>%
head(1))$State
top_p_2019 <- (providers_per100k %>% filter(Year == 2019) %>% arrange(Per100k) %>%
head(1))$State
best_and_worst_p <- data.frame(Year = c("2017", "2018", "2019"), Highest = c(worst_p_2017,
worst_p_2018, worst_p_2019), Lowest = c(top_p_2017, top_p_2018, top_p_2019))
best_and_worst_p %>% kbl() %>% kable_material(c("striped", "hover"))| Year | Highest | Lowest |
|---|---|---|
| 2017 | Massachusetts | Alabama |
| 2018 | Massachusetts | Alabama |
| 2019 | Massachusetts | Alabama |
Based on the map above, the results of this table are not surprising. Massachusetts maintains over 500 mental health professionals per 100k residents from 2017-2019 while Alabama only has 100 or less, depending on the year. As mentioned with other measures, this could be attributed to demographic differences, but could also relate to demand. If a state generally has residents that suffer more from mental illness, then it’s possible there are more resources available. Though it is also possible that individuals are suffering more because of the lack of care. The next visualizations will explore this idea further.
These are two maps we’ve already looked at, just placed together for easier comparison. The first is the map showing providers per 100k, and the second is the rate of mental illness.
A few similar patterns emerge between these two maps, particularly for the northeast and Oregon. These areas have high values for both measures, indicating a level of support and treatment available that reflects the rate of mental illness. Most states, though, do not follow this pattern. This holds true mostly for rural states and the south. The graph below shows the average values from 2017 to 2019 for both providers and mental illness rates. I compiled the states that have the highest (green bars) and lowest (red bars) average number of providers per 100k residents and plotted that relative to the average mental illness rate for each state.
The resulting visualization shows that while all of these states have a similar range of mental illness rates, there is a disparity in the treatment available. Many of the states that fall into the lower end of providers available are southern or more rural states, while this is not the case for states highlighted in green. This is important to note, as this is most likely attributed to demographic differences as well as attitudes surrounding mental health. The last section of this report will look at the latter by considering how interested people are in seeking treatment for their mental health issues.
Interest in Seeking Treatment
The first step to receiving help is seeking it out in the first place. With a wealth of information right at our fingertips, I took a look at SEO phrases and keywords that therapists and counselors frequently use and see to be effective to determine how people are using Google to search for support. Among these phrases, “counseling” and “therapy” are the two most popular, but I wanted to make this more narrow to people actively seeking treatment. Someone could search the term therapy for any number of reasons, but those searching “therapy near me” are most likely doing so with the goal being to start attending therapy sessions. For that reason, I used the phrase “therapy near me” to gather Google Trends data on the frequency of people searching for or showing an interest in mental health support.
For the cohesiveness of this report, I gathered data on the last 5 years, since most of the other data used in this analysis ranged from the years 2015-2020. This gives a timeframe from November of 2015 to November of 2020 for the search interest of this phrase. The first visualization in this section shows the overall trend of the search frequency for “therapy near me” for the U.S.
trend_5YR %>%
ggplot(aes(Week, Frequency))+geom_line(color='#C67D70')+ theme_pomological()+labs(title="5 year Search Trend of 'Therapy Near me'",subtitle="November 2015-November 2020",x="Time")
There has been a clear, steady increase in the search frequency for this phrase. It reaches a peak around early 2020 before it drops off drastically a few months later-presumably due to COVID-19 and not having access to in-person support during quarantine. This drop-off doesn’t last long, though, as we see it peak even higher in the months to follow. People are generally becoming more interested in seeking out therapy in their area. Now let’s take a look at this data on a state-by-state level. The following map shows the search frequency of “therapy near me” over the last 5 years.
There is not much of a regional trend except for a few states in the northeast, but there does seem to be lower search frequencies in the rural midwest. The next table shows the top 5 states with the highest vs. lowest search frequencies.
best_sf <- (search_state5YR %>% arrange(-Frequency) %>% head(5))$State
worst_sf <- (search_state5YR %>% arrange(Frequency) %>% head(5))$State
best_and_worst_sf <- data.frame(Highest = best_sf, Lowest = worst_sf)
best_and_worst_sf %>% kbl() %>% kable_material(c("striped", "hover"))| Highest | Lowest |
|---|---|
| Delaware | South Dakota |
| Pennsylvania | Hawaii |
| New Jersey | Alaska |
| Michigan | Montana |
| New Hampshire | Wyoming |
A few of these states may look familiar, seeing that there is some overlap between this table and others previously presented in this report. New Hampshire was a state with a high rate of mental illness, while New Jersey had a low rate. New Jersey also had low suicide rates, whereas Montana and Wyoming had some of the highest. Hawaii also showed up as having low rates of mental illness and South Dakota having low rates of mental distress. So with nearly all of these states standing out among the other datasets, I next will visualize some comparisons.
This first graph compares the search frequency over the last 5 years to the average number of mental health providers (per 100k residents) from 2017-2019. The states with the highest and lowest search frequencies are shown. Green bars are states with the highest search frequencies while red states have the lowest. The pink dots are the average number of mental health providers.
Generally, there is not much of a trend of relationship to be seen between the two measures. Some states with a large number of providers have a high search frequency for therapy whereas some have a low number, and vice versa. Delaware, for example, has the highest search frequency with around 250 providers per 100k residents. Nebraska has nearly the same amount, but a much smaller search frequency. In fact, the states with less interest in seeking out therapy have more mental health providers than their more interested counterparts, save for Massachusetts. This could indicate a few things. One could be that the resources and support is there, but people are either not seeking it out or don’t need it. Another could be that there are not enough resources available to those states who search for it the most often, and therefore may need it most. To look into these ideas further, I also want to compare this measure to others discussed in this report. The first will be comparing search frequency to mental illness rates.
Below is a similar graph but instead of comparing to providers, it is comparing the average rate of mental illness for states with the highest vs. lowest search frequencies. The colors represent the same thing as before, but pink dots are now average AMI rates for the years 2017-2020.
Here we see that despite similar rates of mental illness, there is a disparity in the interest in seeking out therapy. Some states, such as Pennsylvania, Kansas, and Nebraska, all have similar mental illness rates, but different search frequencies. Pennsylvania is the second most interested in therapy while the latter two are some of the least. This could say a lot about demographic differences and attitudes towards mental health in these states. Rural, less populated states seem to have lower search frequencies than those with high ones. This could be an indicator of something deeper and more systemic that is setting these two groups apart.
The last comparison I want to make is between search frequencies and average suicide rates. Again, the same graph is presented below, but plotted against average suicide rates for the years 2016-2019.
The disparity becomes even clearer when comparing these two measures. States with the highest suicide rates have the lowest search frequencies, while states on the lower end for suicide rates sek out therapy more. This could say a lot about differences in attitude regarding mental health among these two groups of states. Some of the states with the highest suicide rates have the lowest interst in seeking therpy in their area, which could definitely be a contributing factor into said high rates of suicide.
Conclusion and Further Analysis
There was a lot of information and comparisons in this report, so to sum up my findings, I will discuss each piece more briefly in this section. Analysis into this topic could take many different routes and go further and deeper than what is presented here. This only scratches the surface of what encompasses the mental health crisis in America. In the first section of this report, I discussed mental distress, which may or may not mean an individual suffers from a co-occuring mental illness. These rates have been on the rise in the U.S, with West Virginia and Oregon having the highest rates and South Dakota and Minnesota having the lowest. Mental illness rates are also increasing throughout the country, though different states were present in these rankings than for mental distress. Oregon and New Hampshire reported high mental illness rates, with New Jersey, Florida, and Hawaii at the lower end. The fact that there is no overlap between the top and bottom states in these two measures is interesting, and could be attributed to having been diagnosed. It is possible that an individual could be suffering from a mental illness, but not have the access to care in order to receive a diagnosis. This leads into my next area of analysis: suicide rates.
Wyoming and Montana have seen the highest rates of suicide, while New York and New Jersey have seen the lowest. A few visualizations showed that despite having similar mental illness rates, there was a disparity in suicide rates. Some of the states with low suicide percentages had very, if not more, mental illness rates than states with high suicide rates. Looking deeper into the states that fall into each category shows a trend that begins to emerge. Rural, less populated states seem to make up a majority of those experiencing higher suicide rates when compared to those experiencing the lowest. With this trend kept in mind, I proceeded to look into other areas of interest to see if this pattern continues to arise.
Mental health providers are most available to those in Massachusetts while Alabama has the least. Again, a comparison was made between mental illness rates and the number of mental health providers on a per state basis. A similar pattern emerged here. The states with the highest number of providers had similar rates of mental illness to those with the least number of providers, and those with the least fell into the category of belonging to either rural or southern America. At this point, the demographic differences are becoming more apparent in this exploration and I wanted to take it one step further to explore interest in seeking out mental health support.
Using the search phrase “therapy near me”, Google Trends revealed that again, states who search this phrase the most are more populated, urban states when compared to those who search for it the least. There was some overlap in these states and those that stood out in other measures, such as New Jersey and New Hampshire as well as Wyoming and Montana. A few comparisons were made with this measure to others, the first being comparing this search frequency to the number of mental health providers in each state. This revealed that despite having a lot of providers available to them, there is a low search frequency in some states, namely rural areas. Furthermore, a comparison was made between search frequency and mental illness rates as well as suicide rates. States with vastly different search frequencies have very similar mental illness rates, which means those that are struggling aren’t seeking help in some areas. Where are those areas mostly located? Rural America. This trend comes into view possibly the most when looking at suicide rates vs. search frequency. States with the highest rates of suicide see lower search frequencies for therapy than states with lower uiscide rates. Again, these states follow the trend laid out by previous pieces of data and measures.
At the end of all of these comparisons, there appears to be one consistent pattern that exists for nearly every measure explored in this report. It looks to be like a comparison between rural and non-rural America. While there are numerous factors that contribute to mental health and seeking care, demographics and attitudes definitely play a role. Studies have already been done on the disparities between these two segments of America in relation to mental health, and the results indicate that there are in fact differences in how these populations approach this topic. Stigma is a very common reason for these differences. One study found that due to lack of anonymity in rural areas, people may be less likely to seek help due to fear of judgement. This study in particular looked at stigma associated with depression and depressive episodes and found that people from rural areas labeled individuals who sought mental health treatment more negatively than those from urban areas. Another study found that older adults in rural America, of which 10%-25% suffer from poor mental health, believe that they “should not need help” or would not talk to a stranger about their personal matter. Younger people, on the other hand, are more likely to be open and share their struggles. Given that rural areas are typically older, this sentiment could apply to the area in general. An article written for the New York Times takes a bit of a different route and dives into the soaring death rates due to suicide, drug overdose, and alcoholism among working class Americans without a college education. Given that individuals living in suburban or urban areas have a higher likelihood of having received a college degree, it’s reasonable to assume that these higher rates of “deaths of despair”, as the authors of the article call them, are happening more frequently in rural areas. Rural America is also home to a lot of the working class, largely for this same reason. This explains the findings from my exploration into this data quite well, considering that the states who fared more poorly in terms of mental health measures were often rural areas. The same article explains that poor mental health among this demographic is often due to “lack of structure, status, or meaning” when compared to their college educated counterparts.
Discrepancies among different demographic groups in the U.S. regarding mental health is a key issue that came to light when completing this report, but urban versus rural America is only a small piece of the story. Other factors, such as age, race, income, and more, could also contribute to differences not only in mental health, but access to care and proper diagnoses. As mentioned earlier, these findings only scratch the surface of this topic. Further analysis and exploration of these ideas could lend to a more complete story, and possibly even proposals and recommendations for change for what is clearly a system that needs it.
Additional Resources
If you or someone you know is suffering from a mental health issue, please seek treatment or use any of the available hotlines below. You are not alone. You deserve support.
National Suicide Hotline : 1-800-273-8255
National Eating Disorder Hotline: 1-800-931-2237
Depression and Bipolar Support: 800-273-TALK (8255)
National Alliance on Mental Illness (NAMI): 1-800-950-NAMI (6264)
The Trevor Project (LGBTQ Support): 1-866-488-7386