This bulletin provides coding and abstracting direction to ensure the collection of accurate and quality data strictly related to opioid overdose:

  1. Confirmed opioid overdose
  2. Query (unconfirmed) opioid overdose
  3. Use of all available documentation

Opioid Overdose Coding Direction (PDF)

Webinar: In the Spotlight — Opioid Overdose

Transcript for In the Spotlight — Opioid overdose webinar

Margaret Penchoff: Let me introduce today’s speaker. Alana Lane is presenting today’s session. Alana is a program lead with Classifications and Terminologies. She has worked with CIHI for the past 10 years, during which time she has held various positions, namely classifications specialist, coder, data quality coordinator and an instructor for an HIM program. Alana is a location independent worker who works from her home in Prospect Bay, Nova Scotia. Now without further ado, Alana will begin the session. Over to you, Alana.

Slide 1: In the Spotlight — Opioid overdose

Alana Lane: Thanks Margaret, and good morning to those of you on the West coast and good afternoon to those of you on the East coast. We are really glad that you are able to join us today for the first Classifications and Terminologies “In the Spotlight” webinar. This is a new series for us this year, and these webinars are 20/30-minute presentations just to provide an opportunity to bring forward some specific topics of interest and provide an opportunity for you to ask some questions. We are targeting September for our next webinar. The target topic is Folio Views.

Slide 2: Agenda

So on the agenda today, we are going to speak to the bulletin Opioid Overdose Coding Direction. We are also going to briefly touch on the guidelines to support capturing cases of opioid overdose. We will visit a few eQuery coding questions related to opioids and their harm and answer any questions you may have.

Slide 3: The opioid crisis — A joint response

Opioid overdose has become a national crisis, as we all know. In 2016–17, an average of 16 Canadians were hospitalized every day with opioid poisoning. The national hospitalization rate for this type of poisoning has actually increased by 53% over the previous 10 years.

Pan-Canadian data on opioid-related harms is an urgent priority to inform the public health sector’s emergency response to the opioid crisis, which makes the accurate data collection of opioid harm a very important piece of providing support for this critical topic.

That leads to the reason of why we are having the discussion today. That is to ensure that the direction provided for coding cases of opioid overdose are understood to ensure quality data collection.
You can also visit our website at www.cihi.ca to learn more about our commitment to this national crisis.

Slide 4: v2018 updates

For version 2018, as you likely know by now, we have updated codes in ICD-10-CA. We have expanded T40.2 and T40.4 to the fifth digit. The reason we did this is to be able to identify specific types of opioids — such as hydromorphone, oxycodone — and also specific types of synthetic narcotics such as fentanyl. This was to make sure that we would be able to identify the types of drugs that are causing harm at a higher level for reporting.
To align with these changes, we have also expanded the external cause codes — the Y-codes — for adverse effects and therapeutic use due to opioids. These were expanded as well to denote the specific types of drugs

Slide 5: Opioid Overdose Coding Direction

The coder bulletin Opioid Overdose Coding Direction was released in November 2017, and we have since updated it as of April 1, 2018, to reflect the new ICD-10-CA codes. The direction did not change. We went in and updated the codes to reflect the new ones for version 2018.

This bulletin was released by CIHI and was developed in collaboration with the Ontario Ministry of Health and Long-Term Care. We worked with an advisory working group, which included emergency service physicians. The information in the bulletin was meant to bridge the opioid poisoning data collection gaps to allow for a more accurate and complete picture of the number of cases in order to gain support for initiatives to address the issue.

The bulletin specifically addresses confirmed opioid overdose, the use of Narcan (the antidote), query (unconfirmed) opioid overdose and also the use of all available documentation.
Now we are going to look at each of these briefly.

Slide 6: Opioid Overdose Coding Direction (confirmed/Narcan/toxicology)

The direction provided in the bulletin says that to code a confirmed physician-documented “opioid overdose” or drug overdose that identifies a specific opioid as an opioid poisoning. This is just business as usual. If the physician documents opioid overdose, then you code it as an opioid overdose.

The bulletin also provided new direction, however, for when there is use of an opioid antidote for a suspected overdose, and this is Narcan, otherwise known as… which is the opioid antidote used in Canada. It comes in multiple different forms which are injection, IV and also nasal. When there is documentation of a suspected overdose without further specification, and Narcan is administered with a positive effect, meaning the patient opens his or her eyes and regains consciousness and/or begins to breathe normally, then you can assume that the overdose is an opioid overdose.

Narcan is only used to treat an opioid overdose, meaning that Narcan is administered for a suspected overdose and, if the patient has not overdosed on an opioid, then Narcan won’t work. Then there will not be a positive effect. That’s why we can put out the direction to be able to rely on the fact that if you have a suspected overdose and then used Narcan with a positive effect, then you can assume it’s an opioid overdose.

So, often you will see documentation of suspected overdose or even just overdose. The use of Narcan with a positive effect is a confirmation that the overdose is opioid-related, even without the physician documenting “opioid overdose.” I also want to remind you today that coders must not interpret the toxicology reports and that they do require clinical correlation. Just because a substance appears in the toxicology report does not mean that that was the substance that caused the overdose. Again, there is the option to consider if Narcan was used with a positive effect if you suspect that the case may be a drug overdose related to an opioid.

Slide 7: Opioid Overdose Coding Direction (query and use of all available documentation)

The bulletin also provided some direction for query unconfirmed diagnosis. This is not new direction. The coding standard Unconfirmed Diagnosis applies, and the direction provided in the bulletin specific to query opioid overdose is based on the direction from that standard.

The direction simply says to classify an unconfirmed opioid overdose as an opioid poisoning and assign prefix Q.

However, if the case is a questionable opioid overdose, you can consider whether or not Narcan was used with a positive effect. If so, then you can classify it as a confirmed opioid overdose. If Narcan is not administered or it was without a positive effect, then the diagnosis remains as a query opioid diagnosis.

The bulletin also provides new direction to use all available documentation. That includes non-physician documentation — such as nurses’ notes, paramedic records, respiratory notes — and it also includes documentation for the inpatient visit when applicable.

We know it’s not typical for a coder to go into the inpatient record in order to capture an ER visit or accurately… however, in order to assist with identifying all cases of opioid overdose accurately, we have provided the direction that it is okay to do this for these cases, and it’s okay to go in and look at the other non-physician documentation as well.

Often, patients who have suffered an overdose are admitted quickly to an inpatient intensive care unit in order to stabilize their respiratory system. So the inpatient record is therefore a great source for confirming a suspected opioid overdose.

The intent is not for coders to do an exhaustive search, but to use the source documents available to you to identify opioid overdose cases accurately.

You also need to remember that the direction provided in the bulletin is strictly for opioid overdose cases and should not be applied to other diagnoses, meaning that you should not be using non-physician documentation or the inpatient record if the diagnosis you are trying to confirm is not an opioid overdose.

Slide 8: Guidelines to support capturing cases of opioid overdose (poisoning)

During the development of the opioid coding direction bulletin, it was also identified that there was a need to identify the documentation gaps.

The coder bulletin provides direction for coders in the absence of a documented diagnosis, but we also needed to consider what we could do to improve the documentation of these known cases of opioid overdose.
Therefore, in February 2018, we developed guidelines and in collaboration with the Ministry of Health and Long-Term Care and the advisory working group to support capturing cases of opioid overdose. These were released with a targeted audience for physicians. These guidelines identify the dos and don’ts of documentation to capture cases of opioid poisoning.

The guidelines encourage physicians to carefully document and refrain from using terms such as query or questionable for an opioid overdose diagnosis when it is reasonable to consider that the case is most likely due to opioid excess (at least in part), to ensure the case is classified as an opioid overdose.

Of course, the key of the guidelines is to document opioid overdose. So these guidelines were distributed and shared widely through email. And we distributed them to other health care organizations. We do encourage you to share these with your physicians as well.

The guidelines are available on our Codes and Classifications web page.

Slide 9: Intoxication vs. poisoning

I would like to take a moment to touch on intoxication versus poisoning, and we do encourage you to reference the coding standard Adverse Reactions in Therapeutic Use Versus Poisonings. And also in Appendix A in the opioid bulletin, we did provide the definitions for classifying a case as a poisoning.

The coder must remember that an overdose is synonymous with poisoning.

So I am bringing this forward to you today because we received questions and see documentation of “acute intoxication” or “drug abuse” or “substance abuse” where the case could really be an opioid overdose.
However, whether ingestion of a substance is considered intoxication or poisoning is a clinical decision.

But if you do suspect that the case is an overdose where the final diagnosis is documented as “intoxication” then, again, it is advised to check other available documentation, such as the inpatient record or even the ambulance record or nurses’ notes, to confirm if the case was actually an opioid overdose.

Further, in cases where there is documentation of “intoxication” and you suspect it may be an overdose, you can also consider whether or not the patient was unconscious and did they receive Narcan with a positive effect. This is another indication that the case was an opioid poisoning and not a case of acute intoxication.

A true case of acute intoxication is classified to F11.0, which is mutually exclusive from the codes in category T40 for poisoning by an opioid. So those 2 codes, meaning… mutually exclusive means that they must not appear together on an abstract.

So it is very important to be able to make that distinction or seek clarification from the physician.

Slide 10: eQuery coding questions

We thought it would be a good idea to bring some of those questions and highlight part of the responses that we felt were important to bring forward today. Most of them have the eQuery number associated with the question if you want to review them in their entirety in the eQuery database.

Slide 11: No need for physician documentation of “opioid overdose”

So we’ve received multiple coding questions where there is documentation of a suspected overdose and Narcan was administered with a positive effect.

The codes are concerned with assigning a code for opioid overdose in the absence of physician documentation of “opioid overdose.”

We wanted to reaffirm today that the point of identifying Narcan with a positive effect where this is a suspected overdose alleviates the need for the physician to specifically document “opioid overdose.” Again, this direction is provided in the bulletin. So if a patient is unconscious and you suspect that there is a suspected overdose, they receive Narcan with a positive effect, then there does not need to be physician documentation that there is quote unquote opioid overdose.

Slide 12: Incomplete documentation (use available doc and Narcan)

Coding question 66651 sent in the following question: “Does the physician notation of ‘probably ingested’ qualify as a poisoning? The physician did not document the diagnosis as an ‘overdose or poisoning.’”

This is another case of a suspected overdose. Again, there does not have to be documentation of overdose or poisoning by a physician if there is evidence that Narcan was given with a positive effect. Or if there was no Narcan provided, the coder can also check for other available documentation — again, those are your nurses’ notes, your ambulance records or the inpatient documentation, if available — to support whether or not the case qualifies as an overdose.

Even with a query diagnosis, you are still able to consider Narcan and all the other available documentation as per the direction in the bulletin. If none of these applies, a query opioid diagnosis is classified as an opioid overdose with the prefix of Q.

Slide 13: Left without being seen

Coding question 68647 provided the following scenario: “Emergency department physicians do not always document the diagnosis as ‘overdose.’ For example, when a patient uses opioids, has loss of consciousness and is given Narcan with positive results, but leaves the emergency department without being seen by the physician, the physician does not document ‘overdose’ as the final diagnosis. I do not feel comfortable classifying these cases as an overdose when the patient leaves without being seen.”

So very likely what you would see on these records is that a clerk or a nurse or even a registration person has written “left without being seen” (LWBS) on the record.

As we just discussed from the last coding question, there is no requirement for the physician to document opioid overdose if Narcan is given and has a positive effect for a suspected overdose.

In many instances, patients who suffer an opioid overdose… they are given Narcan by a friend, a paramedic, even a stranger on the street, as the Narcan antidote kits are available to anyone. The bulletin specifically says Narcan can be administered prior to arrival at the facility or during the episode of care.

Of those patients who suffer an overdose and are transported to hospital, there are instances where these patients will awaken immediately once they receive Narcan, and they may choose to leave the ER without being seen by the physician. However, it is still important to note that these are an opioid overdose case when the information is available to do so. So again, if you suspect that it is an overdose, going back to using all your available documentation and also considering whether Narcan was given with a positive effect.

Slide 14: Narcan intervention

Coding question 70332 was submitted to seek assistance with assigning a CCI code for Narcan administration during an episode of care. So, they are wondering, now that we are capturing, according to the information in the bulletin, to capture the opioid overdose, do we now have to capture the CCI code for Narcan? The answer is no, that currently it is optional to capture a CCI code for the administration of Narcan. We have not mandated it, and to ensure data consistency a decision to capture the code should be made at the facility or the jurisdictional level.

If your facility or jurisdiction decides to mandate the capture of Narcan, the code on this slide is used: 1.ZZ.35.HA-T2. However, you will notice that this is pharmacotherapy using various systemic agents, which also includes other types of antidotes and it’s not just specific to Narcan. So it is not a reliable code to use in order to identify all cases of Narcan.

Slide 15: Search eQuery using “opioid” keyword

Should you wish to review other eQueries or questions related to opioid harm, you can search the eQuery database with the keyword “opioid.”

Please remember to visit us at www.cihi.ca and visit the Codes and Classifications page for further information, and access to the bulletin and the clinical documentation guidelines for opioids.