The more recent outcome studies, for whatever reason, also do not seem to explore the effects of institutionalization on outcome, as far as I'm aware.
The following is from Kanner (1971), a follow-up study on 11 autistic children originally reported by Leo Kanner in 1943.
Richard M., Barbara K., Virginia S., and Charles N. (Cases 3, 5, 6, and 9), who spent most of their lives in institutional care, have all lost their luster early after their admission. Originally fighting for their aloneness and basking in the contentment that it gave them, originally alert to unwelcome changes and, in their own way, struggling for the status quo, originally astounding the observer with their phenomenal feats of memory, they yielded readily to the uninterrupted self-isolation and soon settled down in a life not too remote from a nirvana-like existence. If at all responsive to psychological testing, their IQ's dropped down to figures usually referred to as low-grade moron or imbecile.
The four autistics mentioned in the preceeding paragraph were placed in institutions from an early age.
Kanner lost track of two cases, Paul A. (Case 4) and Alfred N. (Case 8). However, it's notable that Alfred, a child with an IQ of 140, was placed in many different schools and hospitals. At one point he was given Thorazine.
Two other cases, Donald T. (Case 1) and Frederick W. (Case 2), were considered success stories by Kanner. They were never placed in institutions.
John F. (Case 10) died suddenly at the age of 29.
The remaining 2 cases were not considered success stories by Kanner. But they did not do poorly. Herbert B. (Case 7) was placed in a home only for a short period of time. He was still mute in adulthood, but he ended up in a farm where he was able to help out with various chores. Elaine C. (Case 11) was apparently first institutionalized in 1950 at the age of 18. As of 1970, she was still at Hudson River State Hospital and was reported to be independent, neat and clean.
Kanner comments on the institutionalization factor as follows.
One cannot help but gain the impression that State Hospital admission was tantamount to a life sentence, with evanescence of the astounding facts of rote memory, abandonment of the earlier pathological yet active struggle for the maintenance of sameness, and loss of the interest in objects added to the basically poor relation to people; in other words, a total retreat to near-nothingness...
The question arises whether these children might have fared better in a different setting or whether Donald and Frederick, the able bank teller and the duplicating machine operator, would have shared the dismal fate of Richard and Charles in a State Hospital environment. Even though an affirmative answer would most likely be correct, one cannot get away from wondering whether another element, not as yet determinable, may have an influence on the future of autistic children...
Maybe a life sentence is too dismal. It's not impossible for autistics to manage to come out of institutions and resume their lives. But this is probably extremely rare.
Not institutionalizing autistic children when they are young doesn't guarantee independence and employment, of course. But it appears to be practically a requirement if that sort of outcome is to be attained. Note that Kanner reached this conclusion after analyzing 96 cases, not just the 11 mentioned.
Now, why is it that strong warnings against institutionalization are not issued by doctors and other professionals? Why is it that instead some autism associations give the impression that institutionalization is indicated unless there's treatment?