Here’s the paper I’ve quoted before (or maybe it is a follow-on study since it uses controlled lab tests based on temperature measurements from the refugee camp):
There are a lot of relevant links to other studies in that paper. The paper also points out the importance of temperature fluctuations; during summer here in Oregon I see fluctuations comparable to those in the paper in hot years. My indoor temperature has sometimes hit 100F (37C) in the past; I had to develop a protocol for cooling the house overnight then closing all the windows at 7am 
In that paper the fluctuations maxed out at 37C too but they detected no significant degradation over 12 weeks. In comparison storing insulin at the average temperature (31C) resulted in degradation over a short period of time:
Since the median value of temperature during the cycling process (31°C) remains above the temperature specified by manufacturers, a series of vials was exposed continuously to this temperature, and insulin quantification was performed at 1, 4 and 8 weeks.
All insulin formulations started to degrade after 4 weeks (degradation proportion between 5 and 13.5%, compared to values obtained for the corresponding formulations at T = 0). There was further degradation after eight weeks (12 to 19%), with all formulations out of the range accepted by the pharmacopeia (100 +/- 10%)
Whereas, with the fluctuations around 31C:
After 12 weeks of temperature cycling, the determined values for the studied formulations were still between 96.4 and 101.9% of T = 0 measured concentrations
This result surprises me somewhat; as a chemist I know that these molecules denature (fall apart) almost instantly at some critical temperature, however for insulin this is around 70C:
https://pubs.acs.org/doi/abs/10.1021/bi0507940
I assume that this is sufficiently higher than the maxima under even semi-reasonable storage conditions that it doesn’t matter; just don’t leave the insulin on the parcel shelf of your car!
This isn’t just relevant for people living in tropical regions or people like me who don’t have cooling (although, truth to tell, I got a split unit a few years back; enormously expensive to operate).
I would like to see some studies where the temperature fluctuations are larger. It isn’t that much of an issue in the tropics, where diurnal temperature variations are lower, but it can be in temperate regions. My indoor max was probably around 100F but I know that my outdoor max has hit ~115F (45C) in the past. So far this year (which has been fairly mild) I’ve seen a maximum of 108F on August 14 and a minimum of 26F on October 29 (it’s a new weather station so the results only extent back to spring).
Someone camping with insulin in Oregon can easily face 100F+ for extended periods during the day, but the overnight temperature will typically be around 70F (37C to 21C); a larger variation than in Kenya as would be expected.