It's probably worth mentioning that if the police are involved during a mental health crisis there is a strong likelihood of physical violence even if you are not a minority.
> The call resolution data further revealed that transports to hospitals were more frequent than arrests (44.9% versus 5.8% respectively)
Officers reported far more cases of arrest-worthy behavior/actions that they chose not to act on. The authors concluded:
> Our analysis of a sample of mental health call resolutions in the Chicago Police Department reveal that police are disinclined to use legal action, and considerably more inclined to transport individuals for hospital psychiatric assessments. Moreover, police routinely resort to on-scene call resolutions involving an array of informal tactics.
Appreciate your push for more concrete sources, I don't have any. There are four points that led to my comment:
[1]: Police violence in the news in the United States. How common it is is very debatable, the worst news bubbles to the top and catches the media's attention. For every police shooting there are probably 100+ stops with no shooting. Hopefully the increased media attention has led to concrete studies that will either prove or invalidate my point.
[2]: The very recent shooting of a 13 year old with autism whose mother called the police (https://www.theguardian.com/us-news/2020/sep/08/linden-camer...). This is one event, its my opinion that it is representative of many police encounters with the mentally disabled.
[3]: Living in two cities (Eugene OR, and San Francisco) with high rates of homelessness, vagrants, and transients, and observing the success of non-police based services versus police-based services interacting with the homeless.
[4]: Internet comments from people who I believe are EMTs, crisis counselors, and teachers, on stories like [2], sharing similar experiences, sharing their own de-escalation training, and calling out terrible process and active escalation by police in mental health situations.
EDIT:
[5]: For your own source is this violence/non-violence rate based on police reporting, or do they interview other non-police participants?
Thanks for such a thoughtful response. On #5, it appears to be based on police reports, interviews, and corresponding docs (arrests, hospital admissions). The arrests and hospital visit counts seem objective and the authors don't appear to have an agenda.
I don't have any particular expertise or strong opinions about this topic. My interest is that the threshold/evidence requirement for telling people to avoid a certain resource should be set pretty high.
Separately, the data confirms your #1: https://www.bjs.gov/content/pub/pdf/cpp15.pdf (https://news.ycombinator.com/item?id=23467379). The numerator is absolutely massive -- 53 million people per year. Roughly half of those are traffic stops, but even excluding those, it's still big. Citizens were asked the level of escalation of their most recent police-initiated contact or traffic accident and 0.3% said that a gun was pointed (page 17). That 0.3% is from citizens, not police, so it's probably an upper bound. For comparison, about 3% of people's most recent contact included any use of force, mostly handcuffs (1.8%) and push/grab/hit/kick (0.7%).
> [1]: Police violence in the news in the United States. How common it is is very debatable, the worst news bubbles to the top and catches the media's attention. For every police shooting there are probably 100+ stops with no shooting. Hopefully the increased media attention has led to concrete studies that will either prove or invalidate my point.
Without diving deeply into the data, I know that such collections efforts are occasionally prone to issues where different departments record similar events in drastically different ways. So there are known shortcomings in the data set, but there is data. I'm reasonably sure researchers have worked with this data.
> [3]: Living in two cities (Eugene OR, and San Francisco) with high rates of homelessness, vagrants, and transients, and observing the success of non-police based services versus police-based services interacting with the homeless.
I have no doubts whatsoever about the success of kind, compassionate, and humanitarian services in Eugene. However it seems to me to be contra-factual to describe SF's services to community members experiencing being unsheltered as at all a success except by the metric of quantity of money spent. Police-based or no seems minimally impactful.
I think the obvious point is that if the police are involved there is a small but non-zero chance of physical violence or arrest, and if they are not involved, there is a zero chance.
