Best Toxicology LIS and Lab Software

Every row in the plate map is one step of the same journey. A general-purpose LIS handles the first and last steps well (what an LIS does). The middle steps are where toxicology labs lose time, money and, at inspection, findings.

1. Order. CLIA asks the requisition to solicit “any additional information relevant and necessary for a specific test to ensure accurate and timely testing and reporting of results, including interpretation” § 493.1241(c)(8). For a drug panel, that is the medication list. Diagnosis codes ride on the same order for the claim. Requisitions are kept at least 2 years § 493.1105(a)(1).
2. Receipt and validity. The lab logs the date and time of receipt § 493.1242(b), runs its validity checks against written acceptability and rejection criteria § 493.1242(a)(7), and records the condition and disposition of any specimen that fails § 493.1283(a)(3).
3. Presumptive screen. An immunoassay reports a drug class as positive when it reads above a cutoff.
4. Reflex. A rule sends some screens on to definitive testing: positives to confirm them, and negatives for a drug the patient is supposed to be taking.
5. Definitive test. GC-MS or LC-MS/MS names the drug and its metabolites, in house or at a reference lab.
6. Interpretation. Each drug is read against the medication list as consistent or inconsistent, which is the part the ordering clinician acts on.
7. Report and claim. The report carries the content in § 493.1291(c); the claim carries the codes, the diagnosis and, where needed, an ABN.

A vendor that documents steps 1, 4 and 6 is selling a toxicology LIS. One that only documents steps 3 and 5 is selling an instrument interface, which every LIS on our small-lab LIS comparison already does.

Your reflex rules are a billing decision as much as a lab one. Palmetto GBA's Urine Drug Testing local coverage determination (L35724) is a useful model because it spells reflex out. It says that, since reference laboratories do not have access to patient-specific data, reflex testing is reasonable and necessary in two cases:

• to verify a presumptive positive with a definitive method before the presumptive finding is reported, without an additional order from the clinician;
• to confirm the absence of a prescribed medication when the presumptive screen is negative “for a prescribed medication listed by the ordering clinician”.

The second case only works if the medication list reached the LIS as data, not as a scanned page. The same LCD notes that oxycodone and oxymorphone “will not typically be detected even at a 300 ng/mL cutoff” on a presumptive opiate screen, so a negative screen for a patient on oxycodone is exactly the specimen the rule exists for.

The LCD also lists what it will not pay for. Three of them are LIS settings:

• reflex definitive testing when the presumptive test was done at the point of care;
• “routine standing orders for all patients in a physician's practice”;
• more than one presumptive result per patient per date of service, “regardless of the number of billing providers”.

It sets frequency limits for presumptive and definitive testing over rolling 7, 30, 90 and 365-day windows, by patient group, and calls specimen validity testing “quality assurance, not a Medicare benefit, and thus not separately payable by Medicare”.

If your LC-MS/MS confirmations were developed in house, or you modified a cleared test system, CLIA makes the lab establish the method's performance itself before reporting patient results: accuracy, precision, analytical sensitivity, analytical specificity “to include interfering substances”, reportable range and reference intervals § 493.1253(b)(2). Your cutoffs come out of that work. The procedure manual then carries the reportable range and reference intervals § 493.1251(b)(6), (10), and the director approves, signs and dates any change before use § 493.1251(d).

That gives the LIS cutoff table three jobs. It must match the validation file and the signed procedure, it must change only when they do, and it must remember which cutoff was in force for each past report. CLIA has no section titled “audit trail”, but a dated history of cutoff changes is how you show that a report from last spring was read against last spring's numbers § 493.1283, § 493.1291(k). Performance specifications themselves are kept for as long as the test system is in use and no less than 2 years § 493.1105(a)(3)(i).

Proficiency testing for urine drugs

The toxicology analytes with CLIA proficiency testing rules are serum, plasma or blood tests such as acetaminophen, blood alcohol, blood lead, digoxin and lithium § 493.937. Urine drugs of abuse are not on that list. For a test outside subpart I, the lab must verify its accuracy “at least twice annually” § 493.1236(c), and document it § 493.1236(d). Our comparison of PT programs covers the providers.