If nothing else, CRM has been a paradigm shift. It has forced the industry to recognise that we need more than simply stick and rudder skills to fly safely. The lack of measurable success is no reason to discard the concept: maybe it evidence that we still have some way to go. Maybe 'accident rates' is the wrong measure?
Spooning 'CRM' into 'SMS' is not an answer. The SMS must inform the CRM training but the fact remains both the SMS 'action' and CRM represent interventions. If poorly designed in the first place, neither is likely to work. Is an SMS any better at fixing problems? Does an SMS guarantee a 100% success rate in dealing with issues arising?