Because of the phrase "compare between other solutions and your solution," my guess is that the reviewer suggested some intervention that has a strong research component so that on top of improving well being, you also get scientific information out of it.
Generally, it would imply many things, among which: including a proper control group or other treatments for contrast, have appropriate power/sample size, prepare a clear protocol, randomization/group randomization, implement consent, plan all analysis before, etc. There are also some more meta-level components, for instance, incorporating cost-benefit analysis, RE-AIM protocol, effectiveness/efficacy research, etc.
The way I see the difference is that "evidence-based intervention" can be carried out with or without any intention to produce generalizable conclusion. We can perform a rigorous literature review, plan a program, and work with a community center to kick it off. No scientific design, no data collection, and no manuscript but yet evidence-based because all components were borrowed from carefully chosen published statistics or analyses.
From the tone, I don't think the reviewer seriously doubt your work. Void of context, I'd suggest you to incorporate that comment in your "next step" or "recommendation." After all, if it's a pilot study paper, it's unrealistic to expect a full-blown comparison. However, if what you submit is based on a full scale program that has research-design-related flaws (e.g. no proper sampling, low sample size, no control group), then perhaps you can address it by providing effectiveness/efficacy outcomes from other intervention studies that involved similar sample participants to back up your program. You may also address those lacks in the response. This is hard to tell without knowing what you did.