I have no direct evidence because the numbers are not out there yet, but it's hard to argue anything different.
Mental Health and Coping During COVID-19 | CDC
COVID: Mental Health 'Epidemic' in Communities of Color (webmd.com)
COVID-19 and your mental health - Mayo Clinic
Teen Mental Health Has Suffered During the COVID-19 Pandemic | Time
CDC study sheds new light on 'mental health crisis' linked to coronavirus pandemic - CNN
Oh, I don't doubt that the pandemic has caused a lot of mental health problems. As has the racial tension over the past year, a more contentious than usual election, and the post-election craziness.
What I don't really know is whether there is a correlation of higher mental health issues in the "lockdown" states compared to "open" states. Maybe there is.
For example, I just looked at the first three google results on "suicide rates after covid by state":
https://www.thelancet.com/journals/lanpsy/article/PIIS2215-0366(20)30435-1/fulltextStudy was Queensland, AUS. Monthly suicide rate for the 5 years prior was 14.85/100K. Monthly suicide rate from Feb 1 [the day Queensland declared emergency] until Aug 31 was 14.07, so no change was seen. This study obviously doesn't really dig into lockdown.
https://www.thelancet.com/journals/lanpsy/article/PIIS2215-0366(20)30435-1/fulltextStudy was in Japan. Suicides in the first five months of their pandemic (Feb->June) were down 14% from previous years. However, suicides in the 5 months following (July->Oct) increased 16%. This study also doesn't really dig into the lockdown angle.
https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2775359This was in Massachussetts, and purported to only study during the stay-at-home advisory (March->May 2020). They found the previous rate was 0.8/100K "person-months", whereas the rate during this period was 0.67%/100K "person-months", so it decreased. Obviously this could be consistent with both the Queensland and Japanese studies, but as the Japanese study showed a decline followed by an increase, it's entirely possible that longer-term stay-at-home advisory could cause larger spikes as quarantine fatigue set in...
--------------
I think it's important that sometimes we *think* an outcome that we expect to happen is simply the logical result, but the numbers don't always bear that out.
Imagine a logical thought--the pandemic is scary for everyone. It's a new environment, and many people are scared to socially interact because they feel like they're trying to being safe. Others are unconcerned about the pandemic because they feel like they're in low-risk groups, and feel like they should be allowed to do what they want.
- In a state with a lockdown, the first group will feel like they've been told to do this and there is little social pressure to defect and cause cognitive dissonance by doing something they feel isn't right. Whereas in a state without a lockdown, the first group will feel ostracized/alienated as overly cautious or fearful because their personal thoughts are more risk-averse. They could end up either feeling more lonely as they see their social group cavorting around while they choose to stay at home, or they could feel more anxious as they fall victim to social pressure and go out with their social group while dealing with the fear.
- In a state without a lockdown, the second group will feel the same as they've usually felt. In a state with a lockdown, the second group would be more likely to feel anger IMHO than ostracized alienated, and more likely to simply react by doing what they want anyway despite what they're told to do. I think we certainly saw plenty of this in California.
The bad result for those who are fearful without a lockdown is additional loneliness or anxiety, whereas with a lockdown they feel validation and lack of social pressure to defect. Whereas those who are not fearful without a lockdown is status quo, and with a lockdown feel anger and potentially defiant, where are much more empowering emotions.
Now, it may be true that I'm oversimplifying a bunch of really complex emotions; I am. It might not be accurate. But hopefully it shows how certain responses that you might expect could be counterintuitive from someone else, compared to what you carry into this with your own emotional perspective.