• MHA Data Methodology

MHA Data Methodology

MHA Data Methodology

All of the data presented in the MHA State and County Dashboard is collected through the MHA Online Screening Program, a collection of 11 free, anonymous, confidential, and clinically validated screens that are among the most commonly used mental health screening tools in clinical settings.

The Dashboard contains data collected from January 1, 2020, through September 30, 2024.

MHA did not ask for any identifiable personal information as part of MHA Screening. All identifiable information provided by screeners in question responses, including email addresses, phone numbers, home addresses, and names, were immediately removed from the dataset.

While most individuals access MHA Screening organically, MHA has 200 affiliate organizations and multiple partner organizations that often refer users to the MHA Screening Program. To reduce oversampling in areas where these organizations are located, data referred from affiliates and partners were removed from the dataset. Data were only included in the final set if it was referred from search engines (including Google, Bing, and Yahoo, among others), from the MHA National main website, or from national social media platforms (including Instagram, Twitter, Reddit, and YouTube).

When individuals take a screen, they are asked optional demographic questions. We conducted analyses using only results from individuals who had reported living in the U.S. on the state demographic question. In response to that question, users either select the state they live in, "I live outside the U.S.," or "I live in a U.S. territory." All individuals who responded, “I live outside the U.S.,” “I live in a U.S. territory,” or who did not respond to the question were excluded from the dataset. U.S. Census 2022 population estimates[i]were used for state population totals for 2022-2023 data analyses. U.S. Census 2020-2021 population estimates were used for state population totals for 2020-2021 data analyses. For 2020-2021 analyses using race/ethnicity, U.S. Census 2021 annual state resident population estimates for 6 race groups by Age, Sex, and Hispanic Origin[ii] were used for state population totals of each race/ethnicity category. For 2022-2023 analyses using race/ethnicity, U.S. Census 2022 annual state resident population estimates for 6 race groups by Age, Sex, and Hispanic Origin[iii] were used for state population totals of each race/ethnicity category. For 2020-2021 analyses using age, U.S. Census 2021 annual state resident population estimates by Age, Sex, and Hispanic Origin[iv] were used for state population totals by age group. For 2022-2023 analyses using age, U.S. Census 2022 annual state resident population estimates by Age, Sex, and Hispanic Origin[v] were used for state population totals by age group. Age was recoded as “Youth (Under 18)” for ages 0-17 and “Adults (Over 18)” for ages 18-85+. We conducted county-level analyses using results from the ZIP Code demographic question, in which users can type in their ZIP Code. ZIP Codes were then consolidated into counties on Tableau Prep, using an online U.S. ZIP Code database.[vi] For county-level analyses, additional data cleaning was performed to ensure accurate counts. In some cases, users will enter their ZIP Code but will not report their state or will report a state that does not match the ZIP Code they entered. Where a user’s response for state did not match the ZIP Code they provided in the demographic questions, or they did not answer the state demographic question, they were removed from the dataset. In some cases, zip codes may include areas across multiple counties. In these cases, zip codes were assigned to the county that included the majority of the zip code area. County population totals were then calculated by summing the populations for each of the zip codes assigned to the county, using the online U.S. ZIP Code database.

Age and Race/Ethnicity information are gathered from the optional demographic questions asked once someone has completed a screen. For Age, individuals can select from a list of age ranges, including: 8-10, 11-13, 14-15, 16-17, 18-24, 25-34, 35-44, 45-54, 55-64, and 65+. For Race/Ethnicity, individuals can select from a list including: American Indian or Alaska Native, Asian, Black or African American (non-Hispanic), Hispanic or Latino, Middle Eastern or North African, Native Hawaiian or Other Pacific Islander, White (non-Hispanic), More than one of the above, and Other. The categories Middle Eastern or North African and Other are not included in the Race/Ethnicity filter on the dashboard because population information on those two categories were not available at the state level from the U.S. Census. However, screeners who identified as Middle Eastern or North African, Other, or who did not answer the race/ethnicity demographic question are included in the category "All Races/Ethnicities."

In February 2021, MHA created a login feature for individuals who would like to save the results of their screens over time. About 0.2% of all users on MHA Screening save their results under a login. Any individual with a login was immediately removed from the dataset to minimize error in counting repeat users. In 2020-2021, MHA saw an increase in the number of organizations, such as schools, libraries, and workplaces, directing people to use MHA Screening. If multiple individuals take a screen from one of these locations, they are recorded with the same Remote IP address. To minimize counting duplicate records while allowing for records from these organizations to be counted, MHA engaged in a multi-step cleaning process. If there were multiple records for one Remote IP Address in a single day with the same reported demographic information (zip code, state, gender, age, and race/ethnicity), only the first record was counted, and the others were removed from the dataset. If there were multiple records for one Remote IP address and the reported demographic information was different, all records were kept in the dataset.

All rates per 100,000 are calculated by dividing the number of individuals who scored at risk on the mental health screening tool selected by the total state or county population and multiplying by 100,000. When the data is not filtered by Year, that rate is then divided by 4 (because it is calculated using data across 4 years, from 2020-2024). This gives an average annual rate of individuals scoring at risk for the condition per 100,000.

Data Suppression

MHA works to ensure that no one individual is identifiable from information within this dataset. These analyses removed all demographic or other potentially identifiable information. For all conditions, counties were excluded if they had fewer than 5 individuals with a “positive” result within the county.

 

Dashboard Glossary

Depression

The Depression dashboard contains data from individuals who took the Patient Health Questionnaire 9-item tool (PHQ-9) to screen for depression.[vii] The PHQ-9 depression screening tool consists of nine scored items to assess risk for depression. For each item, respondents are asked, “Over the last two weeks, how often have you been bothered by any of the following problems?” The nine items include: little interest or pleasure in doing things; feeling down, depressed, or hopeless; feeling tired or having little energy; feeling bad about yourself - or that you are a failure or have let yourself or your family down; trouble concentrating on things; and thoughts that you would be better off dead, or of hurting yourself; among others. Respondents can select one of four options: not at all, several days, more than half the days, or nearly every day. The 10th question of the screening tool is not included in scoring but asks, “If you checked off any problems, how difficult have these problems made it for you at work, home, or with other people?” For that question, individuals can select: not difficult at all, somewhat difficult, very difficult, or extremely difficult.

Severe depression is defined as any result where an individual reports experiencing symptoms of depression more than half the days to nearly every day for a period of two weeks and thus scored between 20-27 points on the PHQ-9. People who score moderately severe are still significantly impacted, but this dashboard focuses only on users with severe depression and the highest need for imminent support.

Dashboard Terminology:

  • Total Depression (PHQ-9) Responses: The total number of individuals who took the PHQ-9 Depression screen.
  • Total Scoring Severe Depression: The total number of individuals who scored between 20-27 points on the PHQ-9, indicating risk for severe depression.
  • Number of People Scoring with Severe Depression per 100K of State Population: The number of individuals who scored with severe depression on the PHQ-9 within that state divided by the total state population, multiplied by 100,000.
  • Number of People Scoring with Severe Depression per 100K of County Population: The number of individuals who scored with severe depression on the PHQ-9 within that county divided by the total county population, multiplied by 100,000.
  • Percent at Risk for Severe Depression Among Screeners (State Level): The number of individuals who scored with severe depression on the PHQ-9 divided by the total number of individuals who took a PHQ-9 depression screen within that state, multiplied by 100.

Suicide

The Suicide dashboard contains data from individuals who took the Patient Health Questionnaire 9-item tool (PHQ-9) to screen for depression.[viii] The PHQ-9 depression screening tool consists of nine scored items to assess risk for depression. For each item, respondents are asked, “Over the last two weeks, how often have you been bothered by any of the following problems?” The nine items include: little interest or pleasure in doing things; feeling down, depressed, or hopeless; feeling tired or having little energy; feeling bad about yourself - or that you are a failure or have let yourself or your family down; and trouble concentrating on things; among others.

Question nine of the PHQ-9 assesses suicide risk by asking how often in the previous two weeks individuals have had "thoughts that you would be better off dead, or of hurting yourself." Respondents can select one of four options: Not at all, Several days, More than half the days, or Nearly every day. For these analyses, we considered individuals who answered this question with either "More than half the days" or "Nearly every day" to be experiencing frequent suicidal ideation.

Dashboard Terminology:

  • Total PHQ-9 Responses: The total number of individuals who took the PHQ-9 Depression screen.
  • Total Reporting Frequent Suicidal Ideation: The total number of individuals who reported experiencing “thoughts that you would be better off dead, or of hurting yourself” more than half the days or nearly every day of the previous two weeks on the PHQ-9.
  • Number of People Reporting Frequent Suicidal Ideation per 100K of State Population: The number of individuals who reported frequent suicidal ideation on the PHQ-9 within that state divided by the total state population, multiplied by 100,000.
  • Number of People Reporting Frequent Suicidal Ideation per 100K of County Population: The number of individuals who reported frequent suicidal ideation on the PHQ-9 within that county divided by the total county population, multiplied by 100,000.
  • Percent Reporting Frequent Suicidal Ideation Among Screeners (State Level): The number of individuals who reported frequent suicidal ideation on the PHQ-9 divided by the total number of individuals who took a PHQ-9 depression screen within that state, multiplied by 100.

PTSD

The PTSD dashboard contains data from individuals who took the Primary Care Post-Traumatic Stress Disorder screen for DSM-5 (PC-PTSD-5) to screen for PTSD.[ix] The PC-PTSD screening tool consists of five scored items to assess risk for PTSD. For each item, respondents are asked, “In the past month, have you…?” The five items include:

  • had nightmares about the event(s) or thought about the event(s) when you did not want to;
  • tried hard not to think about the event(s) or went out of your way to avoid situations that reminded you of the event(s);
  • been constantly on guard, watchful, or easily startled;
  • felt numb or detached from people, activities, or your surroundings; and
  • felt guilty or unable to stop blaming yourself or others for the event(s) or any problems the event(s) may have caused.

Respondents can select either “Yes” or “No” in response to each of these questions. The results of the PC-PTSD screen are considered positive when an individual answers “Yes” to any three items.

Dashboard Terminology:

  • Total PTSD Responses: The total number of individuals who took the PC-PTSD screen.
  • Total Scoring Positive for PTSD: The total number of individuals who scored three or higher on the PC-PTSD, indicating a positive result.
  • Number of People Scoring Positive for PTSD per 100K of State Population: The number of individuals who scored positive for PTSD on the PC-PTSD screen within that state divided by the total state population, multiplied by 100,000.
  • Number of People Scoring Positive for PTSD per 100K of County Population: The number of individuals who scored positive for PTSD on the PC-PTSD screen within that county divided by the total county population, multiplied by 100,000.
  • Percent at Risk for PTSD Among Screeners (State Level): The number of individuals who scored positive for PTSD on the PC-PTSD screen divided by the total number of individuals who took a PC-PTSD screen within that state, multiplied by 100.

Trauma Survivors

On each of the 11 mental health screening tools in the Online Screening Program, users are asked a series of optional demographic questions following the completion of the screening tool. Users are not required to answer these questions to receive the results of their screen. One of these questions asks, “Which of the following populations describes you? Select all that apply.” The options respondents can select from are “Student,” “LGBTQ+,” “Trauma Survivor,” “New or Expecting Mother,” “Caregiver of Someone Living with Emotional or Physical Illness,” “Veteran or Active Duty Military,” and “Health Care Worker.”

The Trauma Survivors dashboard contains data from individuals who took any of the 11 mental health screens on MHA Screening and answerd the special populations question in the optional demographic questions.

Dashboard Terminology:

  • Number of People Self-Identifying as Trauma Survivors: The total number of individuals who self-identified as a trauma survivor on the special populations question when taking any of the 11 mental health screens.
  • Number of People Identifying as Trauma Survivors per 100K of State Population: The number of individuals who self-identified as a trauma survivor on the special populations question within that state divided by the total state population, multiplied by 100,000.
  • Number of People Identifying as Trauma Survivors per 100K of County Population: The number of individuals who self-identified as a trauma survivor on the special populations question within that county divided by the total county population, multiplied by 100,000.
  • Percent of Trauma Survivors Among Screeners (State Level): The number of individuals who self-identified as a trauma survivor on the special populations question divided by the total number of individuals who answered the special populations question within that state, multiplied by 100.

Psychosis

The Psychosis dashboard contains data from individuals who took the Prodromal Questionnaire – Brief Version screen (PQ-B) to screen for clinical high risk for psychosis.[x] The PQ-B screening tool consists of 21 scored items to assess risk for clinical high risk for psychosis. For each item, respondents are asked, “In the past month, have you had the following thoughts, feelings, or experiences?” Respondents can select either “Yes” or “No” in response to each of these questions. If an individual answers “Yes” to an item, they are asked to respond to the statement, “When this happens, I feel frightened, concerned, or it causes problems for me,” on a Likert scale with five options ranging from “Strongly Disagree” to “Strongly Agree.” The Likert scale responses are scored from one (Strongly Disagree) to five (Strongly Agree). Item scores are summed, with a possible range of scores from 0-105. An individual is considered at heightened risk of developing psychosis on the PQ-B screen if they score 24 or higher on the distress questions. The PQ-B is designed to test for clinical high risk for psychosis and is considered the first step in a two-stage screening process. A positive score on the PQ-B suggests the need for further evaluation by a qualified health or mental health professional who is trained in recognizing the early signs of psychosis.[xi]

Dashboard Terminology:

  • Total Psychosis Responses: The total number of individuals who took the PQ-B Psychosis screen.
  • Total Scoring At Risk for Psychotic-Like Experiences: The total number of individuals who scored 24 or higher on the PQ-B distress questions, indicating a heightened risk of developing psychosis.
  • Number of People Scoring At Risk for Psychotic-Like Experiences per 100K of State Population: The number of individuals who scored at risk for psychotic-like experiences on the PQ-B within that state divided by the total state population, multiplied by 100,000.
  • Number of People Scoring At Risk for Psychotic-Like Experiences per 100K of County Population: The number of individuals who scored at risk for psychotic-like experiences on the PQ-B within that county divided by the total county population, multiplied by 100,000.
  • Percent at Risk for Psychotic-Like Experiences Among Screeners (State Level): The number of individuals who scored at risk for psychotic-like experiences on the PQ-B screen divided by the total number of individuals who took a PQ-B psychosis screen within that state, multiplied by 100.

---------------

[i] U.S. Census Bureau (2023). State Population by Characteristics: 2020-2023. U.S. Census Bureau. Retrieved from https://www.census.gov/data/tables/time-series/demo/popest/2020s-state-total.html#v2022

[ii] U.S. Census Bureau (2023). State Population by Characteristics: 2020-2023. U.S. Census Bureau. Retrieved from https://www.census.gov/data/datasets/time-series/demo/popest/2020s-state-detail.html

[iii] U.S. Census Bureau (2023). State Population by Characteristics: 2020-2023. U.S. Census Bureau. Retrieved from https://www.census.gov/data/datasets/time-series/demo/popest/2020s-state-detail.html

[iv]Ibid.

[v]Ibid.

[vi] SimpleMaps(2023). U.S. zip codes database. Retrieved from https://simplemaps.com/data/us-zips

[vii] Kroenke, K., Spitzer, R. L., & Williams, J. B. (2001). The PHQ-9. Journal of general internal medicine, 16(9), 606-613. http://onlinelibrary.wiley.com/doi/10.1046/j.1525-1497.2001.016009606.x/pdf

[viii] Kroenke, K., Spitzer, R. L., & Williams, J. B. (2001). The PHQ-9. Journal of general internal medicine, 16(9), 606-613. http://onlinelibrary.wiley.com/doi/10.1046/j.1525-1497.2001.016009606.x/pdf

[ix]Prins, A., Bovin, M. J., Kimerling, R., Kaloupek, D. G, Marx, B. P., Pless Kaiser, A., & Schnurr, P. P. (2015). Primary Care PTSD Screen for DSM-5 (PC-PTSD-5) [Measurement instrument]. https://www.ptsd.va.gov/professional/assessment/documents/pc-ptsd5-screen.pdf

[x]Loewy, R. L., Pearson, R., Vinogradov, S., Bearden, C. E., & Cannon, T. D. (2011). Psychosis risk screening with the Prodromal Questionnaire—brief version (PQ-B). Schizophrenia research, 129(1), 42-46. http://www.sciencedirect.com/science/article/pii/S0920996411001770

[xi]Loewy, R. L., Pearson, R., Vinogradov, S., Bearden, C. E., & Cannon, T. D. (2011). Psychosis risk screening with the Prodromal Questionnaire—brief version (PQ-B). Schizophrenia research, 129(1), 42-46. http://www.sciencedirect.com/science/article/pii/S0920996411001770