Number of people is actually not as big a factor as it is made out to be. For major medical and certain supplemental health like dental, there is a load for small groups, but it isn't very much. The risk is certainly higher, but one large claim can make a group's loss ratio beyond what can be recovered, so the additional premium has to be spread across the small group category. The biggest factor which is often overlooked is that the insured is "actively at work" which lowers the potential risk, because working people are healthier and the insurance company can add underwriting factors for certain industries. Another thing to note is even if you are allowed to select a voluntary product like dental or health indemnity product, there is little or no choice on the specific riders or benefit amounts which lowers the "anti-selection" for riders. Products that are partially employer-paid like major med and/or a group's participation percentage being high also limits the potential anti-selection.
Disclaimer: work in Actuarial
That makes sense to an extent. It goes out the door with family plans (to me). Is a spouse more likely to be healthier because their partner works? Child because parent?