What’s in that Opioid Plan from CMS?

Several physicians have asked me to justify concerns regarding a plan announced by the Centers for Medicare and Medicaid Services (CMS), a plan that Dr. Adam Gordon and I have described as enforcing opioid prescribing recommendations laid out by the 2016 CDC Guideline for Prescribing Opioids for Chronic Pain in a non-patient-centered way. They don’t doubt our clinical concern. However, they want details on why we think the CMS Plan will have the impact that we project it will have, given what’s in it.

Answering that question requires some technical explanation, some field observation, and a reminder of how these (and prior year) CMS policies relate to the CDC Guideline for Prescribing Opioids for Chronic Pain.

I will try to weave these elements together as best as I can. I’m sorry the explanation is long. My understanding emerged from reviewing roughly 4 separate documents from CMS, 5 to 7 email exchanges with CMS staff, plus “real world” reports from a commercial pharmacist, 50% of whose practice is tied to pain and addiction patients. I have added some key excerpts as items A through E in this post.

Let me restate the core concern first, however. CMS is proposing to expand ongoing policies that it believes will implement the objectives central to the 2016 CDC Guideline on Prescribing Opioids for Chronic Pain. They posted an Advance Notice on February 1. CMS will accept public comment to AdvanceNotice2018@cms.hhs.gov until March 3, 2017, a short 4-week comment period.

In my judgment, some of the CMS-proposed practices will accelerate a well-remarked national pattern of involuntary, non-patient-centered dose reduction or discontinuation, with attendant harms to patients. Dr. Gordon and I lay these out here, but you can also read anecdotes in the medical literature here, and from patients here. Involuntary, non-patient centered dose reductions were NOT endorsed by the CDC Guideline.

As of March 1, I had heard from two members of the CDC Guideline’s Core Expert Group who have expressed alarm at what CMS proposes to do. I have heard similar concerns from 2 other experts who assisted the CDC in its Guideline, although they have not committed to write to CMS. When four CDC Guideline experts express a degree of concern at what CMS proposes to do, it suggests that CMS is not adhering to the intent of the CDC Guideline, but violating it.

Problematically, payers tend to be one to two steps more aggressive in pushing what amounts to an involuntary prescription clamp-down. We are seeing harm from such practices, but it’s only reported anecdotally thus far.

Why you could elect not to worry:

That said, and before I dive into technicalities, let me acknowledge that there is one way to read the new CMS plan and not be worried at all. In essence, if one sets aside reports from the field, and if one has 100% confidence that physicians will regularly anticipate and fluidly act to overcome pharmacy plan payment barriers, and if one has similar confidence that pharmacies and insurers will consistently process physician requests for support quickly and in good faith, then there may be nothing to worry about. Patients won’t be turned away from pharmacies empty-handed and be forced into active withdrawal, for example. If one lacks this confidence, then we should consider the particulars, in context of the CMS opioid strategy (January, 2017), and community observations.

The particulars:

The CMS Advance Notice proposal for open comment until March 3, 2017 appears on pages 145 to 150. They are further explained by CMS under Section C below, and illustrated by pharmacist James DiMicco in items A1 and A2 below.

In broad strokes CMS will push for the following two key metrics to be used to guide practice, with the general objective that both numbers be reduced.

“number of patients at >90 MME” (that’s Morphine Milligram Equivalents) and

“number of patients >200 MME”

Both figures appear in the CMS Advance notice, but the figure of 90 MME also appears as key to a CMS strategy declaration, on page 12.

The 2017 CMS Advance Notice mostly revises and makes more stringent some policies that it advanced a year ago. There are three key provisions.

First, CMS will require a “hard formulary safety edit” (bottom of page 148) in which prescriptions are stopped at point of sale if calculated MME exceeds 200. This can be overcome only through utilization review interactions between pharmacy, physician and payer. This means the patient brings their prescription, is likely to leave the pharmacy empty-handed until or unless the physician can overcome utilization review. If they enter acute withdrawal over the weekend, so be it. Patient hospitalization for withdrawal or death from suicide is not a quality metric in the CMS Opioid Strategy.

Second, CMS calls for “case management” and “retrospective case review” based on any patient where MME exceeds 90 MME (top of page 148). This sounds more benign, but reality is that CMS has recommended Pharmacy Plans act aggressively, as Mr. Ahern’s email below (Section C) explains. As Mr. Ahern explains, CMS encourages a “soft edit”, which means any prescription >90 MME is refused payment at point of sale, unless the pharmacist decides to override, based on considerations that are essentially up to the pharmacist. Obviously, those considerations could include race, sex, ethnicity, gender, clothing, etc, since there is no guidance. Even Mr. Ahern’s email is not a full description of what is happening in practice. As pharmacist James DiMicco writes below in sections A1 and A2 below. “It has been my experience that Medicare plans and some commercial plans have already been imposing hard edits when it is not currently required.” He sent me a PDF plan rejection that he managed to override, but he said a “soft edit” (where he can override) is increasingly the exception, even for doses of >90 MME.

Third, CMS seeks to refine and broaden a category called “Opioid overutilization” (bottom of page 145 through 147), which is based on a combination of number of pharmacies, number of prescribers, and a maximum dose of 90 MME or greater. In a separate document, this is targeted as “egregious overutilization” in which the goal is to “implement appropriate plan-level claim controls at point-of-sale (POS) for opioids, use improved retrospective drug utilization review to identify beneficiaries at high risk for an adverse event due to opioids, and perform case management with the identified beneficiaries’ prescribers followed by beneficiary-specific POS edits to prevent Part D coverage of opioid overutilization, if necessary.”

In all cases, there is a universal presumption that if any patient is at a higher dose, they are always and in nearly every instance going to be more safe if they are taken to a lower dose.

How does this relate to the CDC Guideline? How about evidence?

For physicians wishing to get out of the firing line, the implicitly encouraged step is involuntary dose reduction to bring doses down, even if the patient is functionally “stable” on their current dose, and without their consent. But that course of action has absolutely no trial data to support it.

What the CDC Guideline review did summarize was that higher doses are, on average, associated with higher risk of overdose (and other outcomes, e.g. depression, falls, etc). We should assume that there is a component of cause and effect here, at least for overdose. I don’t escalate doses, much, on that basis. And good scientific papers have been published showing that expert teams, working with voluntary patients, can often bring very high doses down safely. For patients where I have determined that harm frankly outweighs benefit, I have executed involuntary tapers, to good effect.

But the impact of the CMS plan will not involve voluntary dose tapering by expert physicians, but involuntary tapering by non-expert physicians.

The CDC Guideline explicitly concluded that there was insufficient evidence to support this, and that is because we have no clinical trials and no prospective follow-up studies on this practice.

Thus many colleagues and I expect that the care of patients will be subject to a widening distance between scientific evidence and the practices CMS intends to mandate, with a narrowing space for physicians to exercise judgment based on individualized assessment of risk and benefit, which is something the CDC Guideline DID recommend, quite strongly.

The cumulative impact of CMS policies, including those initiated last year, alongside the more aggressive Pharmacy Benefit Plans, has already been to force non-patient-centered decision-making, i.e. what the CDC Guideline counsels against. I have cared for a few patients in the immediate aftermath of involuntary tapers introduced by others (i.e. a self-inflicted gunshot). We have a rising tide of concerning reports. I am aware of hospital legal teams that are concerned about the liability implications of such practices. So we have some reason here to be concerned.

For that reason I hope you will look at the material below, the links as much as you wish, and consider writing at least a short note to CMS by March 3 at: AdvanceNotice2018@cms.hhs.gov

Stefan G. Kertesz, MD, MSc

Associate Professor, University of Alabama at Birmingham School of Medicine

Views expressed are those of the author and do not represent positions of the Department of Veterans Affairs or the State of Alabama


Two emails from Pain Specialty Pharmacist James DeMicco (who gave his permission for me to post his name and qualifications) (2/23/2017). Mr. DeMicco works in a commercial pharmacy where 50% of the business is patients with pain or addiction.

A1: email of 2/22/2017

Dear Stefan,

Hope all is well. I follow you on Twitter, although I have not signed up myself yet. I saw a post regarding hard and soft edits related to the processing of prescriptions and wanted to share some information to shed some light. At the point of sale (dispensing or billing the prescription to the insurance carrier) a few things might happen, some of which are obvious. 1) it gets approved 2) it gets rejected. It can be rejected for several reasons by the insurance, too soon, duplicate therapy, non-formulary, prior authorization, etc… Insurance companies refer to rejections or edits as “soft” or “hard.” “Hard” edits refer to rejections that can only be possibly resolved through direct correspondence between the prescriber and the insurance company, such as a prior authorization. “Soft” edits are rejections from the insurance that sends a message or flag to the pharmacy via computer software in which some instances the pharmacy can put in some override in the computer, reprocess and the prescription will go thru.

I have attached a printout of a screen shot of an example I had today regarding MED greater than 90mg which I was able to override without having the prescriber have to do the prior authorization because this particular plan considers it a “soft” edit (which is rare) and allows drug utilization review (DUR) pharmacist professional service (pps) code overrides. With the new Medicare D opioid utilization policy, the majority of prescriptions reject as “hard” edits requiring traditional prior authorization nonsense and this is the only one I have come across where I was able to help. Unfortunately, I have seen many patients similar to the ones you have posted about forced to taper (because the insurance will deny the PA and appeal and only approve x dose of opioid). The patient goes from working and having a quality of life to being bed ridden and suffering in pain. It is horrible what is happening to pain patients as the pendulum is swinging.

Appreciate your posts and insights.



James DeMicco, PharmD

J&J State Street Pharmacy

A2: email of 2/24/2017


You are correct. It has been my experience that Medicare plans and some commercial plans have already been imposing hard edits when it is not currently required. It seems anything above 90mg of MED becomes problematic. I work closely with several pain management specialist that treat some of the more severe pain patients, so I will double check that their doses have not exceeded into the hard edit threshold but I do not believe that to be the case and believe these hard edits are coming when it is not required. I can even run test claims to ensure accuracy.

I have observed what happens in an event like this and would gladly share a few stories. You can definitely report and name me without hesitation.



Original Question to CMS of 2/9/2017

From: “Kertesz, Stefan G”

To: “Conway, Patrick H. (CMS/OA)”

Subject: seeking clarification on CMS proposed “hard edit”

Dear Mr. Conway

I am writing to seek clarification/verification of my understanding of pages 145–149, focused on opioids, in the following Advance Notice document. I wish to comment by the March 3, 2017 deadline but would like to verify my understanding of the word “hard edit” on page 148. I reached out to my friend and colleague Dr. Karen DeSalvo on who she thought might be able to answer this question, and her recommendation was to start with you.

The document of interest is this one:

Advance Notice of Methodological Changes for Calendar Year (CY) 2018 for Medicare Advantage (MA) Capitation Rates, Part C and Part D Payment Policies and 2018 Call Letter

(link: https://www.cms.gov/Newsroom/MediaReleaseDatabase/Fact-sheets/2017-Fact-Sheet-items/2017-02-01.html)

I’m a physician in internal medicine and researcher on addictions with a professional focus on advancing opioid safety in my local VA hospital, Birmingham VA Medical Center. My challenge is that I’m less familiar with the language used for pharmacy policy.

Before commenting, I would like to make sure I have a clear understanding of the following subsection, where the words “edit” and “hard edit” (pg 148) are used by CMS.

Basically I am trying to figure out if the new plan would require pharmacists (or perhaps pharmacy benefits plans) to institute a formal denial for opioid prescriptions above a specific dose of 200 MME (perhaps that’s what a “hard edit” is). Separately I wish to understand if there is something like a “soft edit” in which pharmacists are urged to consider not filling Rx’s for customers at a dose of 90 MME?

The language that drew my attention is on page 148 of the document (attached) and reproduced below my signature.

Can you clarify what the meaning of the term “hard edit” is in this document when applied at a 200 MME threshold, and can you indicate further if there is a pharmacist-initiated “soft edit” anticipated at a lower threshold?

Thanks kindly

Stefan G. Kertesz, MD, MSc

Associate Professor, University of Alabama at Birmingham School of Medicine

Opioid Safety Initiative, Birmingham VA Medical Center

Diplomate, American Boards of Addiction Medicine and Internal Medicine


Summary explanation from Robert Ahern of CMS received 2/9/2017

Mr. Kertesz — Please find our response to your inquiry below. thx

Correct, as discussed in the 2018 draft Call Letter, to prospectively prevent opioid overutilization, in 2018, we expect sponsors to implement a hard formulary-level cumulative morphine equivalent dose (MED) opioid edit at point of sale (POS), while excluding beneficiaries with known exceptions from the edit. In the 2017 Call Letter, we discussed that sponsors are expected to implement a soft and/or hard cumulative MED opioid edit beginning in 2017, working towards at least a hard edit in 2018.

A claim is rejected at the POS if the beneficiary’s active or overlapping opioid prescriptions reach or exceed a certain daily cumulative MED threshold. We recommend that Part D sponsors set the soft edit threshold at 90 mg to 120 mg MED. The soft edit rejection can be overridden by the pharmacist. We recommend that the hard edit threshold be set at 200mg MED or above and include additional criteria to minimize false positives by accounting for known exceptions such as hospice care, certain cancer diagnoses, reasonable overlapping dispensing dates for prescription refills or new prescription orders for continuing fills, and high-dose opioid usage previously determined to be medically necessary such as through coverage determinations, prior authorization, case management, or appeals processes. Hard edit claim rejections would not be approved for Part D coverage in the absence of a plan decision to override the edit.


From: Kertesz, Stefan G 
 Sent: Friday, February 24, 2017 1:55 PM
 To: Ahern, Robert J. (CMS/CM) Cc: James DeMicco ; Gordon, Adam J 
 Subject: RE: seeking verification on CMS proposed “hard and soft edit”

I would like to verify one aspect of the proposal for the 2018 letter. Please confirm or clarify if my reading of CMS proposed policy is incorrect.

Page 147 describe a formal “over utilization” criterion based on meeting both of 2 conditions:

  • 3 or more prescribers
  • exceeding certain dose thresholds

Page 148 details certain required responses based on dose thresholds alone.

Please confirm this interpretation: I have assessed that this is a proposed modification of the “Opioid Specific Safety Edits” described on page 25 of Chapter 6 of the Medicare Prescription Drug Benefit Manual Rev 18, 01–15–16, attached to this email)

My reading of page 148 of the 2018 letter is that the requirement for pharmacy benefit plans to impose hard and soft edits sits independent of meeting the 2 part definition of “overutilizaton” from page 147.

Put another way, a patient does not need to meet the “3 or more prescribers” criterion in order to be subject to a CMS-mandated hard or soft edit (according to dose of 200 or 90 milligrams equivalent).

do you agree?

A separate ambiguity is whether CMS is moving from an “encouraged” soft/hard edit based on dose alone to a “mandated’ soft/hard edit based on dose alone. My intepretation of page 148 is that it will be mandated now, whereas it was not in the past.

Please confirm or clarify that too.

Thanks very much

Stefan Kertesz, MD, MSc

Associate Professor,

University of Alabama at Birmingham School of Medicine


Clarification from Robert Ahern of CMS received 2/27/2017

Dr. Kertesz — Please find our response below. If you have additional questions please use the link below to submit them and the team will get back to you with a response. thx


On pages 144–147, we propose revisions to the retrospective drug utilization review criteria used by CMS to identify potential opioid over utilizers through the OMS to monitor sponsors’ compliance with our opioid overutilization policy, and proposed guidance on how sponsors should revise their internal opioid criteria for retrospective identification of opioid overutilization that may trigger case management.

Separately, after the section on CARA, on pages 147–148, we describe our proposed expectations for sponsors to, at a minimum, implement hard formulary-level safety edits based on a cumulative morphine equivalent dose (MED) approach to prospectively prevent opioid overuse at point of sale at the pharmacy in 2018. (Per the 2017 Call Letter, sponsors are expected to implement a soft and/or hard cumulative MED edit beginning in 2017, working towards at least a hard edit in 2018). We propose recommendations for the hard edit criteria (such as no lower than 200 mg MED) which are set by the sponsors’ P&T committees. Sponsors may choose to include a prescriber count criterion in the edit specifications to enhance the safety aspect of the edit. We solicit feedback from sponsors on their experience with these edits to date, including pharmacist overrides / responses, if available, and setting the threshold at or above 200 mg MED.

Question to CM from Stefan Kertesz (2/24/2017) prompting clarification from CMS:


Excerpts from CMS Opioid Misuse Strategy of January, 2018:

CMS is in the exploratory phase of identifying metrics to quantify and track progress in each priority area. For priority area 1, metrics are currently under consideration in the following areas:

For prescribers enrolled in Medicare who prescribe Part D drugs:

Percentage of opioid prescriptions:
 o Exceeding CDC guideline of 90 morphine milligram equivalents (MME) per day
 o Exceeding 7 days of treatment
 o Written for extended release/long-acting opioids

Percentage with beneficiaries receiving an opioid prescription without other supportive therapies/treatments

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.