Evaluation of perceptions and knowledge of mental illness in the United States through crowdsourcing

Crowdsourcing is a method of data collection with possible benefits in assessing perceptions of mental illness in a large US population.

The objective was to describe perceptions and trends of stigma surrounding mental illness in the United States using crowdsourcing. An online survey was conducted evaluating adults in the United States recruited via the online resource Amazon Mechanical Turk. Questions evaluated demographics and perceptions of mental illness. Survey data were adjusted for demographic variables and compared via logistic regression.

Respondents (n = 1422) were predominately 18 to 30 years of age (n = 743; 52.3%) and white (n = 1101; 77.4%). Over half reported an individual close to them had mental illness (n = 932; 65.5%), and more than one quarter (n = 397; 27.9%) reported having a current or previous mental illness. Non-whites were less likely to agree that: medications are effective (odds ratio [OR] 0.63); they would be comfortable around a coworker with mental illness (OR 0.66); and mental illness is inheritable (OR 0.74). They are also more likely to agree that mental illness is preventable (OR 1.49). Individuals reporting mental illness were more likely to agree that medications (OR 1.34; 95% confidence interval 1.03 to 1.74) and talk therapy (OR 1.46; 95% confidence interval 1.12 to 1.90) are effective. Those reporting some or no college were more likely to agree that the United States has good access to mental health treatment.

Crowdsourcing may be an effective way to obtain information regarding demographics, stigma, and mental illness. Personal experiences with mental illness, ethnicity, and educational level appear to continue to impact perceptions of mental illness.