The reviewers included Karl Horvath, MD, of the internal medicine department at Austria's University of Graz.
The studies were published in medical journals between 2001 and 2006. They lasted for about six months to one year and took place in Europe, North America, South America, and Africa.
Each study was designed differently. Together, they compared an older insulin treatment called NPH to two newer, long-acting insulin treatments: Levemir (insulin detemir) and Lantus (insulin glargine).
The reviewers found that NPH insulin and the long-acting insulin treatments yielded similar blood sugar (glucose) control in the studies.
Horvath's team noticed one difference between the new and old insulin treatments: There were fewer cases of low blood sugar at night with the new insulin treatments. But those data were difficult to interpret, according to the reviewers.
The studies were of "low" quality and didn't cover long-term insulin use or quality of life, note the reviewers.
"Until long-term efficacy and safety data are available, we suggest a cautious approach to therapy with insulin glargine or detemir," write Horvath and colleagues.
Horvath and four other reviewers are part of a research group that studied short- and long-acting insulin with the companies Sanofi Aventis, Eli Lilly, and Novo Nordisk. Another reviewer has worked as a consultant for those three drug companies.
Levemir is made by Novo Nordisk. Lantus is made by Sanofi Aventis.
The review appears in The Cochrane Library.