In medicine, the term “compliance” brings up a host of images to those of us tasked with dealing with patients. I can think of any number of patients where compliance — or rather noncompliance — was an important factor in the success ( or failure) of their treatment. Of all the topics that we hear at conferences, compliance is that one topic that receives “honorable mention.” By this, I mean our national lecturers will gloss over noncompliance, but never discuss how they actually improve it. The comments I’ve heard are usually with the underlying tone that we treat our patients in spite of their noncompliance; recognize it’s there, but there’s nothing we can do about. I’m unsatisfied by this general tone of helplessness in the face of our patients’ noncompliance. So here’s my answer to the title question “is compliance possible?” Yes, but likely never 100%. If I’m right, and we can improve compliance, how do we do so?

What methods are likely to work?

My main method is "Scare them straight."

For my wound patients, I would tell them bluntly, "if this leg wound does not improve, you are facing infection, possible leg AMPUTATION, and death." I am legally required to tell them the worst case scenario anyways. If I repeat it enough, it tends to stick with them.  

Having said that, I am the worst, non-compliant patient in the world. A good friend of mine gave me some free supplements to take (Vitamin D and Co Q10), and I am supposed to take them every day for my "well being". Guess what, I think I forget half the time. What's missing here is the accountability. I won't lose my leg if I don't take my supplements!, so I do allow myself to be irresponsible, so to speak.

Just my 2 cents. KS. 

Kazu brings up an interesting point which I have observed in my practice. That is not to say that Kazu is a non-compliant patient since he does not take his vitamin supplements. I find, believe it or not, some of the most non-compliant patients I have treated have higher levels of education or degrees. I think that this stems from " I know better what is best for me" the patient legitimately believes. Usually, I will advise such a patient, when I seek an opinion of another professional (medical physician, accountant, lawyer, financial broker, etc.) ...... I'll listen and perhaps get another opinion or two and follow through with a consensus decision.
Despite our intentions as physicians to want the best for our patients, the treatment of such patients can place the practitioner and the patient at risk. Documentation of incidents of non-compliance with letters to the patient to adhere to recommendations, reinforcement of instructions in written form with the patient signing for these instructions, reinforcement of instructions with a family member present , insistence of the patient getting another opinion regarding your treatment or treatment recommendations and actually having a patient signing a consent or refusal for recommended treatment are important NOT only as medical-legal defenses but ALSO as educational reinforcements to the patient (believe it or not.) 
When treating the non-compliant patient, I always think of the phrase..."do no harm...". By that I mean, that if a patient has a history of non-compliance with previous treatment by yourself or other physicians, treat them conservatively, if you elect to continue to treat them at all. Non-compliance in medicine is probably no different than some individuals who don't exercise regularly, eat too much, drink to excess, etc. In my opinion, it is a behavioral etiology with some of us and our patients. After 20 years of practice, I have learned not to take it personally (most times...LOL) and have adapted my belief in some cases "you can lead a horse to water.....but you can't get them to drink" despite your best efforts/recommendations.  

A number of years ago I co-authored a chapter in a textbook on this topic as well as lectured on it at many seminars. One important "barrier" to compliance is financial. This is especially true in the case of the geriatric patient who might be on a fixed income. When dealing with life or limb threatening conditions and I suspect a lack of compliance, I might bring the patient's family into the picture. For example if a patient has an infection and I find they are not taking their antibiotics, I might call for help from the spouse and/or the patients children. In most cases the family is grateful for my having spoken with them and I have never been accused of breaking the laws of confidentiality. To not do everything possible to help a patient with a serious illness could potentially bring on a negiglence case especially if the patient loses part or all of a limb. The argument would be that the patient had dementia and could not be relied upon to make his or her own decisions and 'why didn't you consult with the family?"