Recently, Coding Compliance Solutions (CCS) reviewed a laboratory that performed urine drug testing. The lab underwent a CMS Program Integrity Contractor (PIC) audit.
The PIC reported an 89.8% error rate citing the lack of medical necessity for performing definitive urine drug testing.
CCS reviewed the medical records obtained by the ordering providers. We read and re-read the regulatory guidance used by the PIC to make their determination of lack of medical necessity.
The Local Coverage Determinations (LCDs) all discuss risk stratification in terms of the risk associated with potential abuse and diversion.
The guidance outlines three patient groups:
- Group A: symptomatic patients, multiple drug ingestion, or patients with an unreliable history. This group focuses on patients who present with signs or symptoms of substance use toxicity.
- Group B: patients in an active drug treatment program for a substance abuse disorder or monitoring across different phases of recovery.
- Group C: patients are those who are on opiates for chronic pain. In this group, the LCD requires that documentation of medical necessity consist of:
- Patient history, physical exam, and previous lab findings;
- Current treatment plan;
- Prescribed medication(s); and
- Risk assessment plan.
Medical necessity is based on the ordering physician’s documentation that focuses on the risk of abuse or diversion of opiates.
As we performed the review, we noticed some glaring differences, namely the definition of risk.
The CMS guidance, LCD L35006, – Controlled Substance Monitoring and Drugs of Abuse Testing does not focus on obtaining drug testing to determine the therapeutic level of an opioid.
CMS cites the SSA indicating that it will not pay for “services which are not reasonable and necessary for the diagnosis or treatment of illness or injury.”
The LCD cites definitive UDT is reasonable and necessary to:
- Identify a specific substance or metabolite that is inadequately detected by presumptive UDT;
- Definitively identify specific drugs in a large family of drugs;
- Identify a specific substance or metabolite that is not detected by presumptive UDT such as fentanyl, meperidine, synthetic cannabinoids, and other synthetic/analog drugs;
- Identify drugs when a definitive concentration of a drug is needed to guide management (e.g., discontinuation of THC use according to a treatment plan);
- Identify a negative, or confirm a positive, presumptive UDT result that is inconsistent with a patient’s self-report, presentation, medical history, or current prescribed pain medication plan;
- Rule out an error as the cause of a presumptive UDT result;
- Identify non-prescribed medication or illicit use for ongoing safe prescribing of controlled substances; and
- Use in a differential assessment of medication efficacy, side effects, or drug-drug interactions.
None of the above relate to the diagnosis or treatment of an illness or injury.
A quick read of the diagnosis codes listed in the LCDs fails to mention chronic pain.
In our review, we found vastly different definitions of risk in the clinicians’ documentation.
Namely, clinicians classified patients as high risk of addiction or accidental overdose, not abuse or diversion. The patient records did not contain a “risk assessment tool” or “risk assessment plan”.
Rather, the records focused on the dosage of the medication, the adequacy of pain control, and the desire to reduce the dosage of the opioid.
In reviewing the PIC’s comments on their record review, we found their determination of lack of medical necessity was based on the fact that the definitive results showed exactly what the patient was taking for pain.
In the view of the PIC, there was no medical necessity for the frequency of the definitive testing itself. They were worried not about the risk of abuse or diversion of the drugs they were on, but rather, the risk of addiction or accidental overdose.
In reviewing the documentation in the record, we did not see variances in the drug screens nor indications in the medical records as to why the patients were classified as high-risk.
Even though many of the patients were on very high doses of opiates to control their pain, the physician’s progress notes indicated the current dosage of the medications and the attempts to reduce the dosage, yet still control the pain.
The notes all indicated that the patients were high risk, but not based on the risk of abuse or diversion.
Many of the patients were in their 70’s, 80’s, and 90’s. The patients’ ages ranged from 70 to 96.
When we interviewed the physicians asking why the patients were classified as high-risk, the overwhelming concern was that they needed to get the numeric value of the patient’s drug level. Ordering physicians were predominately primary care physicians (PCPs), not pain management specialists.
These patients had multiple chronic conditions, in addition to severe arthritis and degenerative joint disease, noted as the causes of much of their pain.
In 2015, CPT had a major overhaul of codes for presumptive, definitive, and therapeutic drug monitoring. However, these codes were never adopted by CMS.
Instead, requires the use of temporary G codes. The problem is that CMS maintains that the G codes are for testing for drugs of abuse. There is a major disconnect between CMS and practicing physicians who are treating patients with chronic pain.
Providers have no alternative set of codes to use to monitor the therapeutic level of opioids. Namely, quantitative (definitive) opioid drug levels can only be obtained by using the temporary G codes G0480-G0483 that test multiple classes of drugs.
The CMS guidance on using these G codes focuses as outlined in the title of the LCD “Controlled Substance Monitoring and Drugs of Abuse Testing”.
The literature is replete of studies that show pain in aging adults is untreated or under-treated. Both persistent pain and pain that is under-treated adversely affect independence and the quality of life in geriatric patients.
Pain management is particularly challenging in this age group because of the declining organ function, the presence of concurrent diseases, and the use of multiple drug regimens to treat these chronic conditions. Adding opiates to the mix requires that physicians monitor these patients closely.
How often is too often to monitor?
How should classes of drugs be tested to determine potential interactions?
All these factors seem to have eluded CMS and certainly the PICs conducting the reviews.
The medical necessity for ordering the testing lies with the treating physician yet, labs are being held accountable for the physician’s decisions and the physician’s documentation.
Labs rarely, if ever, see physician progress notes when tests are ordered. When a laboratory is under review, they are required to get the physician’s progress notes that supposedly document the CMS rationale for ordering the tests.
Why should a physician turn over their documentation of why a test was ordered even though they wrote the diagnosis and completed the requisition as required?
Urine drug testing (UDT) is one of many anomalies within the CMS system that holds the performing provider accountable
for the ordering provider’s actions. If one follows the CMS logic, labs should refuse to perform the UDT as they have no control over what physicians document, yet they are the ones stuck with the audits of the medical necessity.
Holding the lab accountable for the actions of the provider is not going to change behavior and stop UDT.
Providers will simply find other labs.
CMS has the ordering provider information. If they want the ordering providers’ documentation to support the justification, they should request it from the ordering provider.
In 2003, when CMS presented the admission certification, it stated “We continue to believe that the beneficiary’s treating physician—not any treating practitioner—is best situated to determine ‘‘in need’’ status, both because he or she is the primary caregiver and also is responsible for the beneficiary’s overall care.” (Federal Register, Vol. 68)
It seems that this got lost in defining the medical necessity of quantitative testing for “drugs of abuse”.
Cited:
Federal Register, Vol. 68, No. 216/Friday, November 7, 2003. Medicare Program: Review of National Coverage Determinations and Local Coverage Determinations
Recent Comments