Understanding the Difference Between Medical Necessity and Medical Care

Medical necessity guidelines must also be carefully reviewed for their applicability to the individual patient.

The term “medical necessity” is used far and wide in medical care. But medical necessity is not a singular concept, like gravity, nor is it completely undefinable, like love. As a broad definition, it means that the plan of care is going to benefit the patient, and that harm may result if it is not provided. For example, hospital care is deemed medically necessary when the patient requires care that can only be safely provided in a hospital. Removing skin tags is not medically necessary, as not removing them will not result in any adverse outcome. Looking deeper, when we hospitalize patients, physicians often provide routine admitting orders. For example, the nursing staff will check vital signs on a regular basis, usually every eight hours, but occasionally more frequently. When I was in training in the late 1980s and in practice in the early 1990s, it was also a routine procedure to write “standing orders” for all hospitalized patients. This included an as-needed order for medication for constipation (usually ducosate 100 mg), a medication for insomnia (who can sleep with all the nurses waking you to ask how well you are sleeping?), and a medication if the patient’s blood pressure went too high (usually sublingual nifedipine 10 mg if blood pressure exceeded 140/90).

Subsequent studies have shown that not only is ducosate no more effective than a placebo in treating constipation, but treating elevated blood pressure reflexively in the absence of clinical signs of hypertensive emergency leads to heart attacks and strokes. Treatments that we thought were medically appropriate turned out to be worthless, and potentially harmful. It makes sense that payers would not want to pay for that type of care, because it was not medically necessary.

While these are two relatively low-cost interventions, for many treatments, the cost and risk are much more substantial. Few insurance companies would want to pay for a patient to have surgery or a device implanted if that treatment would not benefit the patient, nor would one expect that the patient would agree to undergo a procedure if they were told there is no proven benefit. The gold standard for proving medical necessity is the double-blind, randomized, controlled clinical trial. In these trials, some participants receive the intervention and others do not, and the clinical outcomes are compared without the patient or the investigator knowing which patient gets the intervention.

An interesting example is the treatment of vertebral compression fractures with a procedure called vertebral augmentation. Initial trials of this demonstrated significant improvement in pain for those who had the intervention. The problem was that such patients were not blinded; they knew they had a needle stuck in their back, and those who did not get the intervention obviously knew that as well. Then, in 2009, another trial was performed, whereby those patients who received the placebo actually underwent a sham procedure, where a needle was introduced into their spine and all the steps of the procedure were performed – except the acrylic cement was not injected. In this trial, there was no difference in pain or disability. 

Benefit and medical necessity, though, is often argued as a subjective measure. It is not uncommon for a physician to say, “I have treated many patients with this treatment, and the patients all did well,” nor is it uncommon for a patient to ask for a treatment after hearing it benefitted a friend, or seeing a commercial touting its benefits. While this may sound compelling, it is not a scientifically sound determination. For example, as surgeons became more proficient with laparoscopic and robotic techniques for performing surgery, the use of this technology proliferated. It came as a shock, therefore, when a large study revealed that patients who underwent laparoscopic surgery for cervical cancer had poorer outcomes than those who had traditional open surgery.

Because medical necessity is often determined “in the eye of the provider,” payers, including Medicare, develop medical necessity guidelines. For Medicare, the genesis of these rules is in the Social Security Act, which states, “notwithstanding any other provision of this subchapter, no payment may be made under part A or part B for any expenses incurred for items or services which, except for items and services described in a succeeding subparagraph or additional preventive services … are not reasonable and necessary for the diagnosis or treatment of illness or injury or to improve the functioning of a malformed body member … which are not reasonable and necessary for the prevention of illness, and in the case of hospice care, which are not reasonable and necessary for the palliation or management of terminal illness.”

Medicare has a set process for development of National Coverage Determinations (NCDs) and Local Coverage Determinations (LCDs) that includes comprehensive literature reviews, an expert panel, review of published guidelines, professional society recommendations, and ample opportunity for public comment.

Where there is no formal coverage guideline, Medicare requires that the service be furnished in accordance with accepted standards of medical practice, as determined by the reviewing physician.

However, there is a stark difference between medical necessity determined from the peer-reviewed medical literature based on clinical trials and the rather fuzzy concept of accepted standards of medical practice. An example of this is the issue of joint replacements. While there is no NCD for this surgery, there are LCDs, and the provider community is in agreement that in most cases, joint replacement should only take place after a trial of conservative therapy. But the medical necessity of the length of stay in the hospital is another issue. In many communities, patients having uncomplicated joint replacement, even those 65 years old and older, are discharged the same day, or the day after surgery. Recently, a local orthopedist posted on LinkedIn about his 85-year-old patient who had a hip replacement in an ambulatory surgery center and was discharged the same afternoon. But at the same time, some hospitals and physicians around the country will routinely keep all joint replacement patients hospitalized two or more days after surgery. While this may seem inconsequential, for the Medicare patient, the expected and actual length of stay determines the patient’s status, and there is a significant payment difference between the joint replacement performed as outpatient and inpatient. This means the less efficient hospital would actually be financially rewarded under the Two-Midnight Rule for keeping the patient hospitalized longer than others. So, is a two-day post-operative hospital stay an “accepted standard of medical practice” for the routine patient? Does that hospital get to benefit financially from their relative lagging behind others in providing efficient care? 

Medical necessity guidelines must also be carefully reviewed for their applicability to the individual patient. Clearly, care provided to traditional Medicare beneficiaries must follow the Medicare medical necessity guidelines, but so too must care provided to Medicare Advantage (MA) enrollees, with one caveat. The MA plans cannot make their requirements more stringent, but can make them less stringent, by offering the service as a supplemental benefit. For example, Medicare covers acupuncture only for chronic lower back pain, but some MA plans offer coverage for other conditions. It should be noted, though, that while Medicare requires that MA plans provide the same or more generous benefits to their enrollees, this does not mean that the MA plans must adhere to the Two-Midnight Rule. The Two-Midnight Rule addresses the status of the patient while they are receiving the services and how the providers will be paid – not the actual provision of the care. Medicare considers status and payment to be a contractual issue, and stays out of the discussion. In other words, the Medicare Advantage patient’s status is not an issue of medical necessity.

At the same time, an insurer that provides both commercial and Medicare Advantage plans may offer coverage for a service to their MA patients, since it is covered by Medicare, but not to their commercially insured patients. Until recently, one large insurer covered MitraClip for MA patients, but not for commercial patients. The Centers for Medicare & Medicaid Services (CMS) has an NCD for MitraClip, so they must abide by that for MA patients, but they noted at the time that their internal technology assessment did not find that the procedure offered proven benefit to patients. When reviewing medical necessity guidelines, it is crucial that the policy is applicable to the patient’s payment source, as, once again, medical necessity can rely on the determination made “in the eye of the payer.” The question lingers of why Medicare would find MitraClip medically necessary in patients who meet the NCD, but that this specific payer determined that the procedure is never medically necessary is beyond the scope of this article.

These guidelines must also be carefully scrutinized to determine if they specify patients for whom the procedure is covered, patients for whom the procedure is non-covered, or both. In some cases, the covered diagnoses are listed with indication that coverage for other conditions will be at the discretion of the reviewer – or coverage for other diagnoses will only be for patients enrolled in clinical trials. If the patient falls into the discretionary category, this would warrant ensuring that the documentation from the physician clearly describes why the procedure is necessary. Providing those few extra words may prevent a denial, and the need for appeal. If the physician can provide a reference to a journal article describing the use in similar patients, that would add to the likelihood of success.

Some policies indicate both covered and non-covered diagnoses, or simply state that “all other diagnoses are non-covered.” In the case of a Medicare NCD, “non-covered” means just that. If the patient does not meet the requirements of the NCD, there is no argument that can be made to overturn the denial. On the other hand, LCDs are not binding on reviewers. While the Medicare Administrative Contractor (MAC) that established the LCD is unlikely to approve a “non-covered” indication, an appeal may be successful at higher levels if the argument is compelling and supported by the medical literature. For other insurers, there may be opportunity for appeal, but this is payer- and contract-specific.

A newer realm of medical necessity that is more frequently being addressed by payers is the site of service. Many services can be provided in several settings. Physicians perform skin biopsies, remove lesions, perform endoscopies, and handle many other procedures in the office. This is not only safe, in most cases, but also cost-effective. As the site of service “increases” in complexity, from office to surgery center to outpatient hospital to inpatient hospital, so does the cost to the payer for the use of the facility. And there are an ever-growing number of surgeries that fall in between a simple skin biopsy and open-heart surgery, that can safely be performed in a non-hospital setting. Does the joint replacement need to be done in the hospital, or could it be performed in an ambulatory surgery center (ASC)? This is a determination that must be made based on medical necessity. The surgeon may have an open slot in their OR schedule that day at the hospital and offer it to a healthy patient who could be treated at an ASC, but this would represent convenience to the physician, and not medical necessity. The patient may ask for their surgery to be performed at the hospital, because the hospital is much closer, but this does not constitute medical necessity, instead representing convenience to the patient.

On the other hand, there would be unacceptable risks to performing most surgeries at an ASC – consider a patient with chronic systolic heart failure and sleep apnea, who has a much higher likelihood of requiring intensive perioperative and postoperative monitoring, necessitating care past midnight. While Medicare leaves the decision of the site of service up to the physician, some payers have established preferential prior authorization guidelines, allowing physicians to perform select surgeries in an ASC without prior authorization – but requiring prior authorization if the surgery is to be done at the hospital. 

As alluded to earlier, the medical necessity of the status of the patient is also a common area of confusion. In this case, the determination is not a patient safety issue, as patients who are treated as outpatients receive the same care with the same staff and same equipment as inpatients, but it certainly is a payment issue with broad implications.

Every medical provider wants to do what is right for every patient. But there are often competing forces at play. Ensuring that every patient receives medically necessary care at every encounter is a goal we should all strive to attain.