How Much Insulin Do Other People Take?

I was trying to write a post on partial diabetes remissions (honeymoons) when I came across the following problem:
Type 1 diabetics need insulin, right. In full blown type 1 diabetes, our bodies make very little insulin, so we need to inject enough for all of our needs. If a person needs a very small amount of insulin, or no insulin, and still gets decent blood sugars, we can assume that a person is making insulin. Also, if a person needs an unusually large amount of insulin, we can assume that the person has insulin resistance.
So the question is, what is a normal amount of insulin for a type 1 diabetic who isn't making any to need? Generally in studies on people with type 1 diabetes that are looking for people still making some insulin, a lower cut off is used based on insulin use plus bodyweight. The most common lower cut off is 0.5 u/kg/day, although sometimes cut offs of 0.4 u/kg/day or even 0.3 u/kg/day. Below 0.3 u/kg/day, it is very clear: you're making insulin.

I looked further at two large scale studies. One of them was a Polish study of over 22,000 children and adolescents with type 1 diabetes who were at least two years post-diagnosis (which probably means that a very few were still making insulin). Normal insulin use varied by age, sex, andwhether the person used shots or a pump. Puberty clearly caused an increase in insulin need. The difference between insulin needs in injectors and pumpers most probably had to do with different ways of calculating insulin use- in fact I think the pumpers' insulin usages should probably be used for computing everybody's normal. If you are under 25, you may want to click over to take a look at the graphs for yourself. In people my age, they find that the the 95th percentile of insulin usage was roughly 1.2 u/kg day; the 75th percentile was roughly 0.9 u/kg/day, the 50th percentile was roughly 0.75 u/kg/day, the 25th percentile is roughly 0.55 u/kg/day, and the 5th percentile is roughly 0.4 u/kg/day.
http://eje-online.org/content/158/4/543.long

The other study I looked at was a Japanese study including all adults diagnosed with type 1 diabetes under Japanese criteria, which almost certainly includes a large number of people with what we'd call LADA and other who are making a large amount of their own insulin. In the group of 400 people dx'd under the age of 20, the average insulin use was 0.9 u/kg/day; in the group as a whole, average insulin use was 0.7 u/kg/day.
http://www.jstage.jst.go.jp/article/endocrj/55/6/1025/_pdf

By the way, to compute your insulin use in this form, you first have to figure out how many units of insulin you use on the average day, which may or may not be easy depending on if you use a pump (which will figure this out for you) or, if on MDI, how consistant you are about how much insulin you use. Anyway, once you have that figure, you can divide by your weight in kilograms. Alternately, you could multiply the number of units you use per day by 2.2, then divide by your weight in pounds.
For instance, I use roughly 35 units of insulin per day. To the nearest kilogram, I weigh 45 kilo. That makes my insulin use 0.78 u/kg/day, which makes my insulin needs well within the normal range for a person my age.

What does this mean for you? If you take less than the 5th percentile of insulin usage without taking oral medications, it means you might want to see if you really have a form of diabetes in which treatment with oral medications is possible.