In the latest round of ACA and Mental Health Parity FAQs (part 34, if you’re counting at home), the triumvirate agencies addressed tobacco cessation, medication assisted treatment for heroin (like methadone maintenance), and other mental health parity issues.

Big Tobacco. The US Preventive Services Task Force (USPSTF) updated its recommendation regarding tobacco cessation on September 22, 2015. Under the Affordable Care Act preventive care rules, group health plans have to cover items and services under the recommendation without cost sharing for plan years that begin September 22, 2016. For calendar year plans, that’s the plan year starting January 1, 2017.

The new recommendation requires detailed behavioral interventions. It also describes the seven FDA-approved medications now available for treating tobacco use. The question that the agencies are grappling with is how to apply the updated recommendation.

Much like a college sophomore pulling an all-nighter on a term paper before the deadline, the agencies are just now asking for comments on this issue. Plan sponsors who currently cover tobacco cessation should review Q&A 1 closely and consider providing comments to the email address marketform@cms.hhs.gov. Comments are due by January 3, 2017. The guidance does not say this, but the implication is that until a revised set of rules is issued, the existing guidance on tobacco cessation seems to control.

Nonquantitative Treatment Limitations. Under applicable mental health parity rules, group health plans generally cannot impose “nonquantitative treatment limitations” (NQTLs) that are more stringent for mental health and substance use disorder (MH/SUD) benefits than they are for medical/surgical benefits. “Nonquantitative” includes items like medical necessity criteria, step-therapy/fail-first policies, formulary design, etc. By their very nature, these items are (to use a technical legal term) squishy.

Importantly for plan sponsors, the agencies gave examples of impermissible NQTLs in Q&As 4 and 5. In Q&A 4, they describe a plan that requires an in-person examination as part of getting pre-authorized for inpatient mental health treatment, but does preauthorization over the phone for medical benefits. The agencies say this does not work.

Additionally, Q&A 5 addresses a situation where a plan implements a step therapy protocol that requires intensive outpatient therapy before inpatient treatment is approved for substance use disorder treatment. The plan requires similar step therapy for comparable medical/surgical benefits. So far, so good. However, in the Q&A, intensive outpatient therapy centers are not geographically convenient to the participant, while similar first step treatments are convenient for medical surgical benefits. Under these facts, applying the step therapy protocol to the participant is not permitted. The upshot, from the Q&A, is that plan sponsors might have to waive such protocols in similar situations. This particular interpretation will not be enforced before March 1, 2017.

Substantially All Analysis. To be able to apply a financial requirement (e.g. copayment) or quantitative treatment limitation (e.g. maximum number of visits) to a MH/SUD benefit, a plan must look at the amount spent under the plan for similar medical surgical benefits (e.g. in-patient, in-network or prescription drugs, as just two examples). Among other requirements, the financial requirement or treatment limitation must apply to “substantially all” (defined as at least two-thirds) of similar medical/surgical benefits.

The details of that calculation are beyond the scope of this post, but Q&A 3 sets out some ground rules. First, if actual plan-level data is available and is credible, that data should be used. Second, if an appropriately experienced actuary determines that plan-level data will not work, then other “reasonable” data may be used. This includes data from similarly-structured plans with similar demographics. To the extent possible, claims data should be customized to the particular group health plan.

This means that, when conducting this analysis, plan sponsors should question the data their providers are using. If it is not plan-specific data, other more general data sets (such as data for an insurer’s similar products that it sells) may not be sufficient. Additionally, general claims data that may be available from other sources is probably insufficient on its own to conduct these analyses.

Medication Assisted Treatment. The agencies previously clarified the MHPAEA applies to medication assisted treatment of opioid use disorder (e.g., methadone). Q&As 6 and 7 provide examples of more stringent NQTLs and are fairly straightforward. Q&A 8 addresses a situation where a plan says that it follows nationally-recognized treatment guidelines for prescription drugs, but then deviates from those guidelines. The agencies say a mere deviation by a plan’s pharmacy and therapeutics (P&T) committee, for example, from national guidelines can be permissible. However, the P&T committee’s work will be evaluated under the mental health parity rules, such as by taking into account whether the committee has sufficient MH/SUD expertise. Like we said, it’s squishy.

Court-Ordered Treatment. Q&A 9 specifically addresses whether plans or issuers may exclude court-ordered treatment for mental health or substance use disorders. You guessed it — a plan or issuer may not exclude court-ordered treatment for MH/SUD if it does not have a similar limitation for medical/surgical benefits. However, a plan can apply medically necessary criteria to court-ordered treatment.

The Bottom Line. The bottom line of all this guidance is that plan sponsors may need to take a harder look at how their third party administrators apply their plan rules. Given the lack of real concrete guidance, there’s a fair amount of room for second guessing by the government. Therefore, plan sponsors should document any decisions carefully and retain that documentation.