Practice and How-To

Coding Infections in Australia: Sepsis, Pneumonia and Post-op Infection

Infection coding sits at the intersection of three Australian Coding Standards: ACS 0110 Sepsis and septic shock, ACS 1904 Complications of surgical or medical care, and the principal-diagnosis logic of ACS 0001. Sepsis coding is driven by the alphabetic index path “Sepsis / by type”. Pneumonia coding hinges on whether an organism is documented. Post-operative infection coding turns on the causal-relationship rules in ACS 1904. This guide explains the three patterns, walks through worked examples lifted from the standards, and maps the decisions a coder makes on an infection episode.

It is written for HLT50321 students, trainee coders, and career changers who want a grounded introduction to infection coding in Australia. Every code referenced here is drawn directly from ICD-10-AM 13th Edition worked examples printed in ACS 0110 and ACS 1904. As always, verify each code in the current tabular list before finalising an episode, because subdivisions and instructional notes change between editions.

Infection admissions are among the most common episodes an Australian coder sees, and also among the easiest to mis-code. The alphabetic index paths for sepsis, pneumonia and post-procedural infection are narrow, the sequencing rules are distinct, and the documentation can be ambiguous in ways that other chapters rarely are.

Three rules carry most of the weight:

The rest of this article works through the three infection patterns in order: sepsis first (because ACS 0110 is the most consulted infection standard), then pneumonia (the most common admitted infection), then post-operative infection (where ACS 1904 adds a layer of external-cause and place-of-occurrence coding that does not apply anywhere else).

ACS 0110 Sepsis and septic shock is the authoritative standard for sepsis coding in Australia. It defines sepsis as a life-threatening organ dysfunction caused by a dysregulated host response to infection, and septic shock as a subset of sepsis with profound circulatory, cellular and metabolic abnormalities. Those definitions are clinical; the coding rules sit in the directives.

The six general directives in ACS 0110 shape every sepsis episode:

The exception is worth knowing: A41.9 Sepsis, unspecified is not assigned together with O85 Puerperal sepsis or P36 Sepsis of newborn, because it adds no specificity. Every other sepsis episode follows the six directives above.

Documentation reads: “Patient was admitted with Escherichia coli [E. coli] sepsis, without identification of an underlying source of infection.” This is example 1 from ACS 0110.

How a coder applies the standard:

1. 1Identify the documented condition. Sepsis due to E. coli. No source of infection identified.
2. 2Look up the alphabetic index. Follow Sepsis / Escherichia coli (directive 1).
3. 3Assign A41.51 Sepsis due to Escherichia coli [E. coli] as the principal diagnosis.
4. 4Do not assign an additional source-of-infection code. Directive 1 only requires the underlying source code “if known”, and in this episode it is not.

Final code set per ACS 0110 example 1: A41.51. The pattern to notice is how the index subterm for the organism resolves directly to a specified sepsis code without any further additions.

Documentation reads: “Patient was admitted via the emergency department with shortness of breath, tachycardia and fever. The patient was transferred to the ward and administered high flow oxygen, intravenous antibiotics and fluid resuscitation. Septic workup included sputum and blood cultures (sputum positive for Streptococcus pneumoniae), chest x-ray, and serum lactate. Discharge summary stated: sepsis due to pneumonia.” This is example 2 from ACS 0110.

How a coder applies the standard:

Final code set per ACS 0110 example 2: A40.3 + J13. The example shows why organism documentation matters so much. If the sputum culture had come back without organism identification, this episode would code differently.

Documentation reads: “Patient was admitted with community acquired pneumonia. The patient became feverish and sluggish and was transferred to the intensive care unit with a diagnosis of septic shock and acute type I (hypoxic) respiratory failure requiring ventilation.” This is example 3 from ACS 0110.

This episode is useful because it shows how three of directive 1’s sub-requirements (Chapter 1 sepsis code, R57.2 for septic shock, and an additional code for acute organ failure) sit alongside the pneumonia.

Final code set per ACS 0110 example 3: J18.9 (principal) + A41.9 + R57.2 + J96.00. The sequencing insight here is that J18.9 sits as the principal diagnosis because the admission was for pneumonia; the sepsis and septic shock codes follow as additional diagnoses, even though septic shock is clinically the more dramatic event.

Pneumonia coding in ICD-10-AM turns on whether an organism is documented. The worked examples in ACS 0110 already demonstrate two of the most common patterns: J13 Pneumonia due to Streptococcus pneumoniae when the organism is identified, and J18.9 Pneumonia, unspecified when the documentation does not specify one.

Two principles, both evident in the ACS 0110 examples, shape every pneumonia episode:

When pneumonia is the principal diagnosis, ACS 0001 decides whether pneumonia or a coexisting sepsis leads. When pneumonia is an additional diagnosis on another episode, ACS 0002 decides whether it meets the additional-diagnosis criteria for that episode.

Because organism identification depends on sputum culture, blood culture or other laboratory confirmation that may not be available at discharge, coders frequently query the clinician before finalising an episode where pneumonia is documented without an organism. A clinician query is often the correct step before defaulting to J18.9.

Where an infection is caused by a drug-resistant organism, ACS 0112 Infection with drug resistant microorganisms directs the coder to add a supplementary code from the Z14-Z16 block to flag the resistance alongside the infection code itself. The resistance code is assigned in addition to the infection code, never instead of it.

The common AU hospital pattern is Methicillin-resistant Staphylococcus aureus (MRSA). A documented MRSA bloodstream infection codes to the sepsis or infection code for the site and clinical presentation, plus B95.6 Staphylococcus aureus as the cause of diseases classified to other chapters where the base code does not name the organism inherently, plus Z16.11 Resistance to methicillin to flag the resistance. Extended-spectrum beta-lactamase (ESBL) producers and vancomycin-resistant enterococci (VRE) follow the same three-code shape: infection code + B95-B97 organism code + Z14-Z16 resistance code.

The B95-B97 block is worth understanding in its own right. These are supplementary “cause of diseases classified to other chapters” codes that identify the organism when the infection code itself does not name it. They appear across the infection chapters (pneumonia, UTI, sepsis, wound infections) any time the documentation specifies an organism that the primary code leaves unstated. Block-level Excludes notes, for example at B95-B97 itself, block supplementary organism-code assignment where the infection code already contains the agent (the Excludes note at B95-B97 is why B95.7 is not added to a diabetic-foot infection already coded to L03.- with the organism specified).

Specific resistance codes are verified against the current 13th edition ICD-10-AM Tabular List. ACS 0112 is the governing standard; coders read it before assigning.

Post-operative and post-procedural infection coding is governed by ACS 1904 Complications of surgical or medical care. The standard sits in chapter 19 of the ACS (Injury, Poisoning and Certain Other Consequences of External Causes) and classifies a post-op infection as a clinical intervention complication. In clinical and reporting language this overlaps with the concept of a healthcare-associated infection (HAI), sometimes called a hospital-acquired infection: the Chapter 1 infection code interacts with the ACS 1904 T-code and Y-code pattern to flag that the episode arose from the healthcare contact itself.

ACS 1904 uses the phrase “clinical intervention complication” to cover complications of surgical and medical care that are classifiable to Chapters 1 to 19, whether they occur during or after the intervention. Post-operative wound infection is the archetypal example, but the standard extends to infections of devices, implants and grafts, infusion and transfusion infections, and a long list of specified intraoperative and postprocedural disorders (E89, G97, H59, H95, I97, J95, K91, M96, N99) mapped by body system.

The directives that apply most often to infection episodes are:

The external-cause and place-of-occurrence requirement is what distinguishes infection-complication coding from ordinary infection coding. Every post-procedural infection episode typically includes at least one T-code or body-system complication code, one Y-code for the underlying procedure, and a Y92.2- code for the place.

Documentation reads: “Patient was admitted with a postcholecystectomy wound infection. The procedure was performed five days prior at another hospital.” This is example 12 from ACS 1904.

This episode is useful because it demonstrates the full three-code pattern: the T-code for the complication, the Y-code for the underlying cause, and the Y92 place-of-occurrence code.

Final code set per ACS 1904 example 12: T81.4 + Y83.75 + Y92.23. The pattern is generalisable: any post-procedural infection episode follows the same three-code structure, adjusted for the specific T-code (T81.4, T82.74, T82.75, T85.71, T86.4 and so on) and the body system in the Y83-Y84 range.

Documentation reads: “Patient was admitted for treatment of a stroke (cerebral infarction). During the admission burning pain was noted at the site of the intravenous (IV) catheter which was then removed. On day 8 the patient developed fever, rigors and swelling with purulent discharge from the old IV entry site. A wound swab confirmed a coagulase-negative Staphylococcus infection. Blood cultures were positive for the same organism. A diagnosis of sepsis secondary to IV site infection was made and antibiotic treatment commenced.” This is example 4 from ACS 0110.

This example stitches ACS 0110 and ACS 1904 together: the episode has a non-infection principal diagnosis (the stroke), a device-related infection as an additional diagnosis, and a sepsis code driven by a specified organism.

Final code set per ACS 0110 example 4: I63.9 (principal) + T82.75 + A41.1. Note that B95.7 Other Staphylococcus as the cause of diseases classified to other chapters is not added, because the Excludes note at block B95-B97 prevents its use alongside the A41.1 code that already identifies the organism.

Infection coding errors cluster around four patterns. Recognising them is the quickest route to cleaner episode audits.

Every one of those errors is prevented by the same habit: read the record, open the relevant ACS, look the condition up in the alphabetic index, and assign the code the index directs to. The boring rule is the reliable one.

The same five-step workflow handles most infection episodes cleanly. Apply it in order.

1. 1Read the whole record. Identify what the clinician has documented about the infection, the site, the organism (if any), and whether a clinical intervention is implicated. Note what is documented and what is not.
2. 2Apply ACS 0001 to select the principal diagnosis. If the admission is for the sepsis itself, the Chapter 1 sepsis code leads. If the admission is for the source infection (pneumonia, UTI, cellulitis), the source code leads and sepsis sits as additional.
3. 3If sepsis is documented, follow Sepsis / by type in the alphabetic index. Assign the Chapter 1 code, the underlying source code if known, R57.2 for septic shock if applicable, and additional codes for any acute organ failure (ACS 0110 directive 1).
4. 4If a clinical intervention is implicated, apply ACS 1904. Follow Complication(s) / postprocedural in the alphabetic index for assumed causal relationships (directive 1.2). Assign the T-code or body-system complication code, the external cause code from Y83-Y84 or Y60-Y82, and the place-of-occurrence code Y92.23 or Y92.24.
5. 5Query before inferring. If organism, causal relationship, or infection site is ambiguous, raise a clinician query rather than defaulting to unspecified or assigning from clinical cues. Query-first is the habit that keeps audits clean and the AR-DRG accurate.

This is the workflow the HLT50321 diploma teaches through graded exercises, and the workflow working coders apply every day. The repetition makes it fast; the habit of looking up rather than guessing keeps it accurate.

Australian coders learn ACS 0110 and ACS 1904 through our HLT50321 Diploma of Clinical Coding, a 12-month, 100% online qualification built around real Australian coding scenarios.

Infection coding is one of the most worked-through topics in the diploma, because it touches so many admission types and combines three standards (ACS 0001, ACS 0110 and ACS 1904) in almost every episode. Students practise with documented scenarios drawn from the same patterns as ACS 0110 and ACS 1904 printed examples, apply the directives in order, and build the habit of following the alphabetic index rather than relying on memory. Integrated digital coding software replicates the hospital workflow, including the index-to-tabular lookup path the standards require.

Australia’s best-value Diploma of Clinical Coding includes daily intakes 365 days a year, flexible payment plans, and graded assessments that build ACS 0110 and ACS 1904 judgement the same way a hospital coder develops it: by working through episodes, one at a time, with the standards open.

Keep going