Getting the secondary diagnoses right on a chart is just as important as nailing the selection of the primary diagnosis. A secondary diagnosis is any condition that coexists alongside the main reason for the visit and that requires or affects evaluation, treatment, or management during that encounter. If the primary diagnosis answers “why did the patient come in,” secondary diagnoses answer “what else did we have to deal with while they were here” — and both pieces together are what make the coded record an accurate reflection of the care actually delivered.
Take a patient admitted for pancreatitis who also happens to have diabetes. Pancreatitis is chiefly responsible for the admission, so it’s the primary diagnosis. The diabetes only earns a secondary diagnosis code if it’s actually evaluated or managed during that stay — not simply because it’s sitting in the chart. The same logic goes to the patient admitted for a fractured leg who also has lupus; lupus becomes reportable when it is clinically relevant to healing or care, and it’s actually addressed, not just noted in passing.
Inpatient vs. Outpatient: Different Rules Apply
Setting matters a lot here. On the inpatient side, secondary diagnoses can include conditions present on admission (POA) as well as anything that develops during the stay. Inpatient guidelines even allow certain uncertain diagnoses at discharge to be coded as if they were confirmed, provided the criteria for that are met — a nuance that trips up a lot of coders new to facility coding.
Outpatient coding works differently. The focus shifts toward conditions that establish medical necessity for the services actually performed. And once a definitive diagnosis is documented alongside related signs or symptoms, you code the definitive diagnosis — not the symptoms separately — unless a coding note tells you otherwise. For a deeper look at how this plays out with vague presenting complaints, see this breakdown of coding signs and symptoms correctly.
What Actually Qualifies as a Secondary Diagnosis
Here’s where a lot of over-coding and under-coding happens: a condition doesn’t become a secondary diagnosis just because it’s sitting in the patient’s history or problem list. The documentation has to show the condition was current and clinically relevant to that specific encounter. In practice, a condition earns its place as a secondary diagnosis when it’s monitored, evaluated, assessed, or treated.
- Monitored — labs are ordered or reviewed related to the condition, or the provider documents it as stable, worsening, or uncontrolled.
- Evaluated — it shows up in the assessment and plan, its risk factors or complications are discussed, or it’s part of a differential diagnosis.
- Assessed — a chronic condition is reviewed for its current status or its influence on the encounter, feeding into the provider’s medical decision-making.
- Treated — medications tied to the condition are started, adjusted, or continued, or the provider makes a management decision like a referral, therapy order, or care coordination step.
Together, these 4 criteria are what separate a condition that is actively part of the encounter from one that’s just documentation clutter.
What Coders Should Actually Be Looking For
Determine clinical significance first. Report conditions that are evaluated, treated, monitored, or that genuinely affect the patient’s care, length of stay, or resource use. A condition parked in someone’s past medical history with no relation on the current visit doesn’t belong on the claim.
Get the sequencing right. Identify the condition chiefly responsible for the encounter/visit and list it first — for inpatient stays, that’s the condition confirmed after study to be the reason for admission; for outpatient and professional services, it’s the condition primarily responsible for that visit. Every additional diagnosis should follow the sequencing rules in the ICD-10-CM Official Guidelines.
Push for the highest level of specificity. Read the documentation very closely for site, type, severity, acuity, and laterality, and code to that level. Don’t default to a less-specific code just because it’s familiar when the record clearly supports something more precise.
Watch for etiology and manifestation relationships. Linking language like “with,” “due to,” “caused by,” “associated with,” or “secondary to” often signals that one condition is driving another. When guidelines call for the underlying disease to be sequenced first, don’t let the manifestation jump ahead of it — diabetic chronic kidney disease is the classic example, where the diabetes code leads and the kidney disease code follows. If you need a refresher on the mechanics of this, this guide on “code first” instructions and manifestation codes walks through it in detail.
Know when a combination code does the job. Before reaching for two separate codes, check whether a single combination code already captures the provider’s full diagnostic statement. If it does, use it. If no combination code exists, report the underlying condition and its manifestation separately, sequenced per guidelines.
Don’t overlook comorbidities. Chronic or co-existing conditions like COPD or arthritis deserve a code when they influence treatment, monitoring, or overall risk during the encounter. These conditions also factor into severity and risk adjustment, so accurate capture has real reimbursement consequences — see this overview of CCs, MCCs, and clinical validation for how that connects to DRG assignment.
Skip signs and symptoms once a definitive diagnosis is in place. If the provider has already pinned down the diagnosis explaining a symptom, that symptom generally shouldn’t be coded separately as its own primary diagnosis. The exception is outpatient coding when no definitive diagnosis exists yet — in that case, the sign or symptom itself may be the appropriate first-listed code.
Separate active conditions from historical ones. A cancer currently being treated is not the same as a personal history of cancer, and coding it that way understates the patient’s actual condition. On the flip side, a history code is the right call once a condition is fully resolved but still relevant to current care decisions. This distinction is explained well in this comparison of status codes versus personal history codes, and the companion piece on when to use Z codes as a secondary diagnosis is worth bookmarking for anyone who codes a lot of follow-up visits.
Why This Matters Beyond the Code Itself
Accurate secondary diagnosis reporting isn’t just a compliance checkbox — it’s what makes the coded record match the clinical reality of the encounter. Missed or over-reported secondary diagnoses distort severity of illness, risk adjustment, and reimbursement, and they’re a common source of denials and audit findings. Reviewing conditions against the monitored/evaluated/assessed/treated framework above, and double-checking sequencing and specificity before finalizing a claim, catches most of the errors before they ever reach a payer.
