A lot of this would have to with the problem the habituation is causing. If it is the economic, social, and similar impacts of Taking Care of Business to assure you do not run out of narcotics, maintenance or detoxification with substitutes may be helpful. If it is worry about dependence on oxycodone, then the status of your leg injury and what it does to your functioning is the paramount concern.
The pain is more significant I would think. People have a human right to freedom from pain, and treating pain often requires more than one modality -- in other words don't let anybody tell you that non-pharmacological modalities alone can take care of a fouled up leg, and certainly do not let them tell you that cannabis, as helpful as it is for many pain patients of many types, is an apples-to-apples replacement for narcotics.
Methodologically sound, comprehensive studies by non-politically-obligated organisations the last 220 years or so have found the true iatrogenic opioid addiction rate for compliant patients to be anywhere from 7 per 100 000 (an old US CDC study) to around 1.8 per cent and to have held more or less steady from the first studies of laudanum around 1800 to today. Chronic pain actually reduces the rate further, apparently. It does indeed happen, of course -- and that also means that the inverse can happen as well. A doctor, or in the case of chronic pain, often a team of medical professionals, can very well turn an unsupervised opioid user with a habit into a compliant pain patient just as much as they can, if they are very feckless, turn patients into addicts. In both cases, if the patient wants off the narcotics after resolution of the issue, there are multiple ways to do that.
Have you considered ketamine as part of both reducing narcotic use and pain management? Using that and other methods of narcotic potentiation by usually reduce the absolute quantity of narcotic required for given purpose and I suppose one could say do a fraction of the taper for you without extra pain from the leg. It is also possible to use ketamine to replace the narcotics outright to avoid many of the symptoms of withdrawal, as well. In my experience that level of ketamine regimen would anaesthetise the leg pain possibly altogether whilst you were on the ketamine.
Depending on the type of pain, clonidine not only reduces withdrawal symptoms but is also used in pain management. neuropathic pain in particular. Both clonidine and ketamine in my experience leave the symptom of diarrhoea largely untouched so that may need to be prevented with an additional medication, but that could just be me.
It may have changed a bit with more clinical experience with buprenorphine, but as I recall methadone is usually the suggested detoxification or maintenance agent for Opioid Substitution Therapy in cases where the patient is in severe pain from injuries and other matters. Narcotic withdrawal increases sensitivity to pain beyond baseline, so yes, it is going to be rather onerous in theory.
This is when there are the two choices, that is. This is one example of a number of reasons why, at a bare minimum, physicians and patients should have either extended-release dihydrocodeine or extended-release morphine as options as well.
Efforts in the past to find an analgesic which could be used with low risk in recovering and former narcotic addicts for emergency, acute, and chronic pain didn't really come to a conclusion and were abandoned as far as I know. There was interest in myrophine and azidomorphine for this purpose for a while as well as research on cyclazocine, but this failure may be evidence on the side of strong analgesia and narcotic euphoria being part of the same thing, above and beyond the obvious fact that people are going to feel better when they are in less pain. Most assuredly physical dependence at some level is related as the same opioid receptors mediate it; but there is also rebound even when one stops paracetamol. In all cases that is why medications are tapered, or should be.
Ideally the detoxification, maintenance, and reduction medicine cupboard ought to, and does in various places, also include levomethadone, immediate-release morphine, immediate and extended-release hydromorphone, piritramide, dextromoramide, levorphanol, dihydroetorphine, high-dose dextropropoxyphene, high-dose extended-release tramadol, one of the methadols, and dipipanone. More places are trying diamorphine (heroin) on prescription and an intermediate modality betwixt the straight-across heroin replacement and maintenance with long-acting and extended-release opioids would be something along the lines of a Smack Contin, which in truth would be very hard to distinguish from MS-Contin as heroin, like all 3,6 diesters of morphine, are converted into morphine rapidly, and hydrolysis and acid attack are two very efficient routes by which the body does this. If giving out H presents politically insurmountable problems in a locale, then I suggest an option of morphine and one of the other 3,6 diesters of morphine such as dibenzoylmorphine or nicomorphine. The heroin analogue of dihydromorphine, diacetyldihydromorphine (Paralaudin) also known as dihydroheroin, and the heroin analogue of hydromorphone (acetylmorphone) could be especially helpful for such cases. The come-up of smack and the effects of D. Paralaudin has legs to a greater extent than the others -- the usual duration of effects is 5-7 hours in my experience, at least for analgesia.