EHR records, claims, and patient-reported outcomes each reveal a different part of care. Bringing them together can make quality measures more valid, but teams must reconcile definitions, dates, and patient groups. Clear data rules and routine checks help turn those differences into a more reliable view of care.

Multiple Choice

What is a key consideration when combining multiple data sources (EHR, claims, PROs) for quality reporting?

Combining multiple data sources for quality reporting can strengthen the validity of measures because each source captures different aspects of care: the EHR provides detailed clinical information, claims reflect utilization and costs, and patient-reported outcomes offer the patient’s perspective on results. This broader view helps validate whether a metric truly reflects quality across what was done, what was billed, and how the patient actually experienced care. However, this approach brings real complexity. The sources use different definitions for events (for example, what counts as a complication or a readmission), they record timing differently, and they may cover different patient populations or time windows. To make sense of the combined data, you must reconcile these differences—align metric definitions, map data to common vocabularies, agree on timing windows, and resolve conflicts when sources disagree. Governance and data quality checks are also essential to ensure the reconciled data are reliable for reporting. That’s why this concept is best described as improving validity while adding complexity and requiring reconciliation of definitions and timing. Using a single source can seem simpler but risks missing important information and may bias results; relying on one source for consistency ignores the benefits of a more complete, multi-source view. Limiting use to administrative purposes misses the broader quality-improvement value of incorporating clinical detail and patient perspectives.

A Fuller Picture of Care: Bringing Healthcare Data Sources Together

A hospital’s electronic health record may show that a patient received a treatment. An insurance claim may show when care was billed and what it cost. A patient survey may reveal whether the treatment made daily life easier—or left important concerns unresolved. Each source tells part of the story. Together, they can offer a more useful view of healthcare quality, though bringing them together takes careful work.

Quality reporting often draws on several kinds of information, including electronic health records (EHRs), insurance claims, and patient-reported outcomes (PROs). These sources weren’t built to answer exactly the same questions. That’s both their strength and their challenge. Combining them can improve the validity of a quality measure, but only when health organizations reconcile differences in definitions, timing, and coverage.

Three sources, three different views

EHRs capture details about care inside a health system. They may include diagnoses, lab results, medications, procedures, and notes from clinicians. This can make an EHR a rich source for understanding what happened during a visit or hospital stay. But the record may not show care delivered elsewhere, and data entered by people can be incomplete or uneven.

Claims offer another angle. They’re created when a provider submits a bill to an insurer, so they often capture services across multiple facilities. Claims can help track hospital admissions, procedures, follow-up visits, and costs. They’re useful for understanding how people move through the healthcare system. But a claim is mainly a record of billing and payment. It may not reveal a patient’s symptoms, the reasoning behind a treatment decision, or whether the patient felt better afterward.

PROs add the patient’s voice. These measures may ask people to report pain, mobility, fatigue, mood, or satisfaction with care. A patient’s account can fill gaps that clinical records and billing data leave behind. A scan can show a joint’s condition, for example, while a person’s report can explain whether pain still makes it hard to climb stairs.

Put these views together, and the picture becomes broader: what care was delivered, what services were used, and how the patient experienced the outcome. Still, more data doesn’t automatically mean better data. The sources need to connect in a meaningful, reliable way.

The same event may have different meanings

One common challenge is that data sources may define the same event differently. Consider a readmission. A hospital might count any return to its facility within 30 days. An insurer’s claims data might identify a return to any hospital in its network. Those definitions can produce different results, even when both are handled correctly.

Complication measures raise similar questions. Does a complication count if it appears during the hospital stay, or only if it leads to another visit? Should a condition that existed before treatment be included? How should an organization treat a diagnosis that appears in a billing record but not in the EHR?

These aren’t small details. A definition shapes the number being reported, and the number can influence how leaders understand quality. If one source includes events that another leaves out, comparing their results without checking the rules can lead to a misleading conclusion.

The solution starts with clear measure specifications. Organizations need to define each event, describe which records count, and agree on how to handle exceptions. They may also map different codes and terms to a shared vocabulary. This work can feel fussy, but it’s the foundation beneath a fair comparison.

Timing can change the story

Data sources also run on different clocks. An EHR may record the date a clinician entered information. A claim may arrive weeks later, after the service has been billed and processed. A patient survey might be sent soon after treatment, or months afterward.

That gap matters. Suppose a quality measure looks at follow-up care after a hospital discharge. The discharge date may be clear in the EHR, while a claim for a follow-up visit arrives later. If reporting staff use the date a claim was processed instead of the date care occurred, the visit may appear to fall outside the measure’s time window.

PROs bring a different timing issue. A survey collected two weeks after treatment may capture early recovery. One collected six months later may reflect a more lasting result. Neither is automatically the better measure; they answer different questions.

To make data comparable, teams need to set consistent time windows and distinguish service dates from entry or processing dates. They also need rules for missing information and delayed records. Think of it like lining up clocks before comparing race times: even accurate clocks won’t help if they’re set to different moments.

Coverage gaps and conflicting records

A combined dataset may also contain different patient populations. An EHR usually reflects people who received care within a particular organization. Claims may cover members of an insurer’s plan, while PRO surveys may include only patients who responded. Some people may appear in all three sources; others may appear in just one.

That mix can affect reported results. People who respond to surveys may differ from those who don’t. Claims may be unavailable for people whose insurance changed. An EHR may miss treatment received at another health system. If these gaps aren’t understood, a measure may describe only a slice of the population while appearing to represent everyone.

Records can disagree, too. A person’s date of birth might be entered differently across systems, or a procedure may appear in a claim but not in the local EHR. Reliable record linkage helps identify which records belong to the same person, but matching must be handled with care. Poor matching can merge unrelated records or split one person’s history into fragments.

When sources conflict, teams need a documented way to assess the difference. Sometimes one source is better suited to a particular field. Claims may be useful for tracking paid services across organizations, while the EHR may provide stronger detail about lab values. The aim isn’t to declare one source superior in every case. It’s to understand what each source can and can’t support.

Reconciliation is the quiet work behind good reporting

Reconciliation means making data sources comparable without hiding their differences. It may involve standardizing terms, agreeing on event dates, checking duplicate records, and deciding how to resolve conflicting values. Data teams, clinicians, quality leaders, and patient representatives can all contribute. Their perspectives help keep technical decisions connected to the realities of care.

Good data governance supports this work. A governance group can set definitions, assign responsibility for data quality, and document changes to measures. Routine checks can flag missing fields, unusual shifts, and mismatches between sources. For example, if the number of readmissions suddenly drops, the cause could be a real improvement—or a change in claims coverage or data processing.

Transparency matters, too. Reports should explain which sources were used, how key events were defined, and where important gaps remain. This gives leaders a better basis for interpreting results. It also helps avoid false certainty, which can be just as harmful as having too little information.

More complete reporting, with care

Using one source can make data collection simpler, but simplicity has a trade-off. An EHR alone may miss care received elsewhere. Claims alone may miss the patient’s experience. PROs alone may not capture the full course of treatment. A multi-source view can reduce those blind spots and strengthen the validity of quality measures.

Yet complexity is part of the bargain. Combining sources requires time, shared definitions, consistent timing rules, and ongoing checks. Without that work, the extra data can create noise rather than clarity.

When healthcare organizations handle these details well, multiple data sources can support a more balanced account of quality. They show not only what was recorded or billed, but also how care unfolded and whether it mattered to the person receiving it. That fuller view is worth the effort—provided the differences between sources are made visible, examined, and reconciled.