In the years since my daughter's dx, I've met many, many wonderful parents in varying stages of living with T1. Several have had children dx'd very early, either through participating in TrialNet or just luck. The same basic approach to initiating insulin was followed with all of them. It was not based so much on A1C as on random BG values, particularly post meal values. Insulin was begun when more than 1 reading over 200-250 was reported. The reason given was safety. In the early stages of T1, reactive hypoglycemia is common. In all of those cases, treatment began with small doses of basal insulin only (Lantus, Levemir or in a few cases, NPH).
I have seen studies & expert opinions state that beginning early treatment with insulin can not only help preserve the remaining beta cells, but also result in better "control" over the long-term. All of the ped endos I've spoken with agree with that. The truth is that many ideas related to T1 are unproven, or at least not proven beyond beyond a doubt. The answer to too many questions is simply "we think" or "we don't know" I may have some links still saved on my pc, but I'll have to search through them. This link is the only one saved to my phone.
From the link above:
" The partial remission phase may be defined as an insulin requirement of<0.5 units/kg of body weight per day and HbA1c<7% (51). Recently, insulin dose adjusted HbA1c, defined as HbA1c (%)+4 × [insulin dose in units/kg/24 h] has been proposed as a more specific measure of remission (53, 54). The phase commences within days or weeks of the start of insulin therapy and may last for weeks to years. During this period, blood glucose levels are frequently stable within the normal range, despite fluctuations in diet and exercise. Ketoacidosis at presentation (55), and younger age at diabetes onset reduce the likelihood of a remission phase (53, 56). Intensive therapy leads to better metabolic control and a reduction in insulin requirements (57). While there may be a transient effect of intensive therapy on β-cell function, the effect is not sustained (58). Nevertheless, preserving β-cell function decreases the risk of developing vascular complications and severe hypoglycemia (57, 59). Most people with recent onset type 1 diabetes retain some β-cell function that may persist for decades following diagnosis (60, 61)."
FWIW, my daughter was dxd at age 11 with a BG (discovered by me) of 593. In the ER, her labs showed a BG of 623 & A1C of 13%, positive for 3 of 4 antibodies. (It's been so long I honestly don't remember which 3.) We were very fortunate. She was close to but not yet in DKA. For about 14 months we enjoyed the honeymoon. I say "about" because it didn't just suddenly end. It was more a gradual sputtering. For a few months, her TDD (total daily dose, basal & bolus combined) was only 4 units. The honeymoon was never strong enough for her be without injected insulin. Adjusting insulin during that time could be challenging, but nothing like the unforgiving beast of T1 we've lived with since.