Polysubstance abuse ICD-10: F19 and coding each substance

In this article▾

1. Polysubstance abuse ICD-10 codes at a glance
2. What is the polysubstance abuse ICD-10 code?
3. When should you use F19 for polysubstance use?
4. What the research found about the word "polysubstance"
5. When can F19 create a coding or billing problem?
6. How do you code someone using four substances?
7. What should you write instead?
8. What does it cost to code all substances together?
9. How to prevent diagnosis-documentation mismatches
10. FAQ
11. Sources

Polysubstance abuse ICD-10 coding depends on what is documented. F19 codes cover other psychoactive substance-related disorders, including situations where the substance is genuinely unknown or falls outside the named categories. When specific substances are identified, substance-specific F codes should be considered instead.

For example, a client with documented alcohol dependence, opioid dependence, cocaine abuse, and nicotine dependence would generally have a diagnosis for each substance rather than having the entire history collapsed into F19.

Polysubstance abuse ICD-10 codes at a glance

Descriptions are quoted from the ICD-10-CM code set in effect for FY2027, which starts October 1, 2026. Always check the current Index and Tabular List against your own documentation before you code.

Two things the table cannot show you. F17 has no abuse code, because ICD-10-CM gives nicotine dependence subcategories only, so the abuse-or-dependence choice you make for every other substance does not exist here. And F19.90, like every other F1x.9- code, is narrower than it looks: the Official Guidelines allow the unspecified use codes only when the substance use is associated with a substance-related disorder or medical condition that the provider has documented. An unspecified code is not a shrug.

Sources: ICD-10-CM Official Guidelines FY2027, § I.C.5.b · NCHS ICD-10-CM FY2027 files

What is the polysubstance abuse ICD-10 code?

The nearest thing is F19, which ICD-10-CM titles other psychoactive substance related disorders. It follows the same structure as the rest of the block, so F19.10 is abuse uncomplicated and F19.20 is dependence uncomplicated.

The name is the important part. F19 covers substances that are other or unknown, and it exists to catch what the nine specific categories cannot: alcohol (F10), opioids (F11), cannabis (F12), sedatives and hypnotics (F13), cocaine (F14), other stimulants (F15), hallucinogens (F16), nicotine (F17) and inhalants (F18). Our guide to substance use disorder ICD-10 codes covers how the block is structured.

The practical distinction is important: F19 is not automatically the right choice simply because a client uses several substances.

When the substances are known and documented, the specific substance categories can preserve information that a general F19 code does not.

What F19 is actually for: substances outside the named categories, and cases where the record genuinely cannot establish what someone took. It should not automatically be treated as a container for a list of known substances that happens to be long.

When should you use F19 for polysubstance use?

The question is not simply whether a client uses multiple substances.

It is whether the documentation supports an F19 diagnosis rather than a more specific substance-related diagnosis.

If intake collected a substance history, the screen came back, and you documented the specific substances, the relevant substance-specific codes should be considered.

Coding F19.20 for that person can tell the record less than you actually know.

There's a clinical cost as well as a billing one. An opioid-dependent client in withdrawal needs different management from a cannabis-dependent one, and a nonspecific code can make that distinction less visible to whoever picks up the case next.

The goal is not to avoid F19. The goal is to use it when the documentation supports it.

What the research found about the word "polysubstance"

Researchers at one health system counted 48 separate diagnostic phrases containing the word, none of which named the substances involved. In a single year, clinicians selected one of them in over 87,000 patient encounters, 3.4% of all encounters system-wide (Journal of General Internal Medicine, 2022).

The authors were direct about what gets lost. The imprecise term omits diagnostic and substance use details that could have guided treatment decisions, because different substance combinations need different clinical interventions.

DSM-5 removed polysubstance in 2013. It has been retired for over a decade and survives because it is genuinely convenient and because it is still sitting in most EHR dropdowns.

ICD-10-CM did not follow. No code title contains the word, and it does not appear once in the Official Guidelines. What it does have is one line in the Alphabetic Index: Dependence, polysubstance, F19.20. There is no matching Index entry for polysubstance abuse or polysubstance use.

That single line is the whole mechanism. Write "polysubstance dependence" in the assessment and your coder looks it up, finds F19.20, and assigns it. They are not being careless. The Index gives them nothing else to assign. The information is lost one step earlier, in the phrase you chose.

Sources: Journal of General Internal Medicine, 2022 · ICD-10-CM Official Guidelines FY2027 · NCHS ICD-10-CM Index and Tabular List

When can F19 create a coding or billing problem?

Selecting it because it seems appropriate. This is the big one, and it's not a clinician failing. If "polysubstance dependence" appears in your diagnosis picklist, busy people will select it, because it's there and it's true enough. Removing it from the list does more than any training session.

Using F19 when intake didn't know the complete history. Reasonable on day one. The problem is nobody returns to update it on day two when toxicology resolves the question.

Assuming multiple substances means one combined code. The hierarchy rule in § I.C.5.b.2 limits you to one code per substance at the highest documented pattern: abuse beats use, dependence beats both. It says nothing at all about how many different substances you may report.

Naming substances in the narrative but not the diagnosis. Your assessment lists four substances beautifully and the diagnosis field carries F19.20. The narrative isn't what travels with the claim.

Treating specificity as pedantry. It's the opposite. Coding each substance is how the next clinician knows what they're walking into.

How do you code someone using four substances?

Four codes, one per substance, each at the severity you documented.

Within any one row, the hierarchy still binds. If the chart documents both abuse of and dependence on alcohol, only the dependence code goes on the claim.

The practical test is a single question:

Can you name it? Then code it.

F19 is for situations where the documentation supports an "other" or genuinely unidentified substance-related diagnosis, which may be less common than the word "polysubstance" in clinical notes suggests.

What should you write instead?

The fix is in the assessment line, and it's not longer than what you were writing.

Weak: "Polysubstance dependence."

Better: "Alcohol dependence, daily use. Opioid dependence, IV heroin, withdrawal present on admission. Cocaine abuse, weekend pattern. Nicotine dependence, cigarettes."

That version takes about the same time to write and does considerably more work. It tells the next clinician what to expect, it supports the level of care, and it means nobody has to guess later which substance was driving the admission.

What does it cost to code all substances together?

Diagnosis problems rarely announce themselves as diagnosis problems. Insurers on HealthCare.gov denied 19% of in-network claims in 2024, and the rate ran from 3% to 36% depending on the insurer (KFF). Only 5% of those denials were for lack of medical necessity. A quarter were administrative.

The specific trap with F19 is that whoever handles your prior authorizations may have obtained approval by discussing an opioid dependence diagnosis on the phone, while the claim went out under F19.20. The service was authorized. The claim wasn't, at least not as billed.

Fewer than 1% of denied claims were appealed in 2024, and insurers upheld their original decision in 66% of the appeals that were filed (KFF).

None of that reaches you as a denial. It reaches you as a request for records.

Sources: KFF: claims denials and appeals in ACA marketplace plans in 2024

How to prevent diagnosis-documentation mismatches

The current problem is that your intake team collects a detailed history and it gets compressed into one imprecise code somewhere between the assessment and the claim, where it becomes a CO-50 denial nobody traces back. If you are not sure how often it is happening, start by calculating your denial rate by diagnosis rather than in aggregate.

Nobody does it on purpose. It just happens.

The useful question is not simply whether F19 was selected. It's whether the diagnosis still matches what the clinical record says.

Supa can surface that mismatch before the chart closes. When intake documents four substances and the diagnosis field carries F19.20, that gap can be flagged for review. It can also surface unspecified codes left in place after toxicology was resolved, and diagnoses that stopped matching the narrative weeks ago.

It doesn't diagnose. It doesn't decide which substance use disorder the clinician should document.

It notices when the record says less than the clinical documentation already established.

If polysubstance appears on more than a handful of your admissions, it is worth an hour. Book a demo and we will look at your intake records together.

FAQ

What is the ICD-10 code for polysubstance abuse?

F19.10 is other psychoactive substance abuse and F19.20 is dependence. Neither should automatically be used simply because a client uses multiple identified substances. When you can identify the substances, consider the applicable specific F code for each one.

Can I use F19.20 for someone using alcohol and opioids?

You should not automatically use F19.20 simply because the client uses both substances. Both alcohol and opioids have their own categories, so the chart may carry F10.20 and F11.20 when dependence is documented for both.

Is polysubstance dependence still a diagnosis?

Not in DSM-5, which removed it in 2013 in favor of diagnosing each substance use disorder separately. The term survives in EHR templates and clinical shorthand, which is why it keeps reaching claims.

How many substance codes can I put on one chart?

As many as you documented, one per substance. The hierarchy rule restricts you to a single code per substance at the highest pattern recorded, and places no limit on the number of substances.

What if we genuinely don't know what they took?

F19 may be appropriate when the documentation genuinely supports an other or unidentified psychoactive substance-related disorder. Note that the unspecified use codes, F19.90 and the rest of the F1x.9- range, carry an extra condition: the guidelines allow them only where the provider has documented a link to a substance-related disorder or medical condition. Either way, build the habit of revisiting the diagnosis once toxicology returns, because an unspecified code left in place after the substance is identified is the version that causes problems.

Is there an ICD-10 code that means polysubstance?

Not as a code title, no. But the Alphabetic Index does carry one entry, Dependence, polysubstance, which points to F19.20, so a coder who looks the word up will land there. There is no Index entry for polysubstance abuse or polysubstance use. That is why the phrase in your assessment matters more than the code your biller picks.

What is the difference between F19.10, F19.20 and F19.90?

Abuse, dependence and use, in that order. F19.90 carries an extra condition that the other two do not: the guidelines allow the unspecified use codes only where the provider has documented that the substance use is associated with a substance-related disorder or a medical condition.

Can I code nicotine abuse?

No. ICD-10-CM gives nicotine dependence subcategories only, so there is no F17 abuse code to choose. If the documentation supports nicotine dependence, code the product: F17.210 for cigarettes rather than F17.200 for unspecified.

Do these codes change on October 1?

The code set is replaced every October 1, and FY2027 took effect on October 1, 2026. Nothing in the F10 to F19 block changed in that update, but the habit worth keeping is checking the annual addenda rather than assuming a code you used last year is still current.

Why does it matter clinically, not just for billing?

Because the next clinician reads the code. Alcohol withdrawal and stimulant withdrawal need different management, and a chart that says polysubstance gives whoever picks up the case less to prepare with. The same specificity problem shows up across behavioral health billing.

Sources

1. Centers for Disease Control and Prevention, National Center for Health Statistics. ICD-10-CM Official Guidelines for Coding and Reporting, FY 2027 (effective October 1, 2026), § I.C.5.b. https://ftp.cdc.gov/pub/Health_Statistics/NCHS/Publications/ICD10CM/2027/ICD-10-CM-October-1-2026-FY27-Guidelines.pdf

2. Centers for Disease Control and Prevention, National Center for Health Statistics. ICD-10-CM FY2027 Tabular List and Index. https://ftp.cdc.gov/pub/Health_Statistics/NCHS/Publications/ICD10CM/2027/

3. KFF. (2026). Claims denials and appeals in ACA marketplace plans in 2024. kff.org

4. Oesterle, T. S., et al. (2022). Rethinking the use of "polysubstance" to describe complex substance use patterns. Journal of General Internal Medicine. pmc.ncbi.nlm.nih.gov

Coding note: This article is educational and is not a substitute for the current ICD-10-CM Index and Tabular List, your payer's policies, or your own documentation. Codes and guidelines change every October 1. Every source above was opened and checked on September 25, 2026.