A case for restraint of explosive growth of health information technology: first, do no harm.