Surgical wound outcomes data is not scarce, but almost none of it links a specific wound product to what happened after the patient left the operating room. Registries capture the procedure and the complication, claims capture billed items, and EHRs capture whatever the clinician documented. Each answers a different question, and each has blind spots that matter when the question is whether a dressing, NPWT unit, or biologic changed an outcome.
For HEOR, market access, and value-analysis teams, the practical issue is fit. A dataset that supports a hospital quality program may be unusable for a coverage dossier. This article maps the main sources, explains where they fail on product-level questions, and describes how real-world wound care data can narrow the gap, along with its own limits.
What Counts as Surgical Wound Outcomes Data?
The common endpoints are surgical site infection (SSI), dehiscence, time to closure, readmission, reoperation, and wound-related cost. Definitions are not uniform. CDC/NHSN uses standardized surveillance criteria for SSI, classified as superficial incisional, deep incisional, or organ/space. A clinician-documented “wound infection” in an EHR may reflect a treatment decision, such as starting antibiotics, without meeting any surveillance definition. The same patient can be an SSI in one dataset and not in another.
Dehiscence and “time to closure” are even less standardized. Some sources record dehiscence as a diagnosis code, others as a free-text note, and few record the date of closure at all.
Data types and the question each is built to answer
- Registry: how do procedure-specific complication rates compare across participating hospitals?
- Administrative claims: what services were billed, for which patients, at what cost?
- EHR: what did clinicians in one system document during care?
- Prospective trial: does an intervention cause a difference under controlled conditions?
- Product-level real-world data: how are specific products used, on which wounds, in which settings, and what do serial measurements show?
Acute incisions versus wounds that stay open
A closed incision that heals by primary intention is a different population from a wound that dehisces, is left open, or heals by secondary intention. The second group behaves like a chronic wound: it needs serial assessment, often moves to outpatient or home settings, and is where advanced products, NPWT, and cellular and/or tissue-based products (CTPs, also called CAMPs) are most likely to enter. That overlap is why surgical wound data and chronic wound utilization data often describe the same patients at different points in time.
Where Does Surgical Wound Outcomes Data Come From?
Four sources dominate, and each program updates its scope and reporting rules periodically.
- CDC NHSN: national healthcare-associated infection surveillance, including SSI by procedure category. Superficial incisional SSIs are followed for 30 days; deep incisional and organ/space SSIs are followed for 30 or 90 days depending on the procedure category, per the NHSN SSI protocol.
- ACS NSQIP: a clinical registry in which trained reviewers abstract perioperative data from participating hospitals, with 30-day outcomes for many measures. It is risk-adjusted and granular on patient factors.
- CMS claims and hospital quality programs: national billing data, plus programs that tie SSI and other hospital-acquired conditions to public reporting or payment. Program composition has changed over time; as of FY 2027, the HAC Reduction Program still includes a colon and abdominal hysterectomy SSI measure.
- Single-institution EHR datasets: rich in clinical notes and measurements, but limited to one system’s patients and documentation habits.
Compared side by side:
- NHSN: strength is standardized definitions and national scale; blind spot is limited patient-level detail and a fixed surveillance window.
- NSQIP: strength is risk adjustment and clinical abstraction quality; blind spot is the short follow-up window and participation limited to enrolled hospitals.
- CMS claims: strength is population breadth across settings; blind spot is that diagnosis and billing codes stand in for clinical detail.
- EHR datasets: strength is depth within a system; blind spot is generalizability and care that happens elsewhere.
Note what is missing from every row: the product. Most of these sources record the operation and the complication but not the dressing, NPWT unit, or biologic applied afterward. Two misconceptions follow from this. Registry SSI rates are often read as if they reflect post-discharge outcomes, when they reflect only what the program’s window and case-finding methods captured. And claims are often assumed to identify the dressing used, which they generally do not.
Why Do Surgical Wound Datasets Often Fail Product-Level Questions?
The product identification gap
Claims identify products through HCPCS and Q-codes. Some CTPs and NPWT items carry codes that make them visible, depending on setting and payer. Dressings are coded more coarsely. When billed under Medicare’s surgical dressings benefit, HCPCS A-codes identify the dressing category (alginate, foam, hydrocolloid) rather than the brand. In facility settings dressings are typically bundled into a payment, so a claim can show a wound was treated without showing what covered it.
The post-discharge blind spot
Wounds that dehisce or become infected after discharge are frequently managed in SNFs, home health, or outpatient wound centers. Those encounters sit in different data systems, under different billing rules, from the index surgery. A dataset anchored to the operation may lose the patient at exactly the point the wound becomes a product question.
Confounding and selection bias
Patients who receive advanced products differ from those who do not. Comorbidity burden, BMI, diabetes status, smoking, wound class, and surgeon all influence both product selection and outcome. Unadjusted comparisons between product users and non-users are unreliable, and even adjusted comparisons are limited by whatever variables the dataset actually holds.
A worked illustration
Suppose an analyst uses claims alone to compare NPWT with standard dressings on closed incisions. Patients with NPWT billed may be those a surgeon judged high risk: obese, diabetic, or with a contaminated field. If those patients show more complications, the device looks harmful. If they show fewer, it may look better than it is, because the claims lack the wound class, BMI, or glycemic control needed to adjust. Standard dressings would not appear in the data at all, so the comparator group is defined by absence of a billing code, not by documented care. The result cannot separate device effect from patient risk.
How Does Real-World Wound Care Data Fill the Gap?
Product-level real-world data records what the other sources omit: the product documented, wound type, size, duration, care setting, and serial measurements over time. When that record spans SNFs, home visits, and hospital-based wound centers, it can follow a wound through the settings where dehisced or open surgical wounds often end up.
Defensive value: coverage, audits, and PMCF
Comparative effectiveness studies built on this kind of data can support a coverage case, provide documentation when utilization is questioned, and feed post-market clinical follow-up. Under EU MDR, for example, PMCF is an ongoing obligation for many devices, and real-world data collected in routine care is one way to meet it. Intellicure Analytics offers PMCF services and comparative effectiveness studies for this purpose.
Offensive value: how products are really used
Utilization data can show a product being used on wound types or in settings the manufacturer did not plan for, such as a product launched for one indication spreading into pressure ulcers or home visits. It can also show that the actual competitor is a different product category. A CTP team may assume its rival is another CTP when the data show NPWT or a conventional advanced dressing taking the same wounds. Intellicure’s Wound Care Industry Dashboard and Product Ranking Reports are built to surface these patterns across CTPs, CAMPs, and wound types including DFUs, VLUs, and pressure ulcers.
Limits you should state up front
- The data are observational, so confounding remains even after adjustment.
- Documentation quality varies by site and clinician, and missing fields are common.
- Intellicure’s dataset skews toward chronic wounds (DFU, VLU, pressure ulcers), not primary surgical incisions. It is most relevant to surgical wounds that remain open, dehisce, or move into wound center care, and it should not be used to estimate SSI incidence after routine closed incisions.
What Should You Check Before Relying on a Surgical Wound Dataset?
Before using any source for a claim, ask these questions:
- Outcome definition: is SSI defined by NHSN criteria, diagnosis codes, or clinician judgment?
- Follow-up length: does the window cover when the outcome occurs, and does it match the procedure?
- Loss to follow-up: how many patients disappear after discharge, and is that rate reported?
- Product capture method: is the product documented, inferred from a billing code, or absent?
- Risk adjustment: which covariates are available, and are diabetes, BMI, and wound class among them?
- Care settings covered: hospital only, or also SNF, home health, and outpatient wound centers?
Which kind of data does each decision require?
A coverage dossier needs a controlled comparison with risk adjustment and a defensible comparator. Market sizing or competitive intelligence can tolerate descriptive utilization data, provided the denominators are clear. A PMCF plan needs prospectively defined endpoints, because retrospective data rarely contain the exact measures a regulator expects.
Red flags
- Percentages with no named source or year.
- SSI rates quoted without procedure type or surveillance window.
- Healing rates that omit the denominator or the definition of “healed.”
Which Surgical Wound Dataset Fits Your Decision?
The useful question is not whether surgical wound outcomes data exists, but which dataset fits the decision in front of you. Registries and surveillance programs answer population-level complication questions. Claims show billed activity. Product-level real-world data is what you need when the question is how a specific product performs on documented wounds across care settings.
If you are weighing a coverage submission, a PMCF plan, or a competitive assessment, start by listing your evidence gaps against the checks above, then compare them with what product-level real-world data can supply, whether through the Intellicure dashboard or a comparative effectiveness study. Explore our comparative effectiveness studies