The Process of Structured Reporting: Adding Value and Quality

Travis Browning

Structured reporting is a tool, and like any other tool expertise comes with education and practice. The purpose of this tool is to standardize and codify the content of radiologic interpretations to facilitate communication—communication to referring clinicians, communication to future comparing radiologists, and communication for the purposes of quality improvement and research.

The three elements that constitute structured reporting include consistent headings (indication, comparison, findings, etc.), itemized reports with discrete short descriptions, and use of a standard lexicon (1). While there is yet no universal agreement on the construction of these elements, the Radiological Society of North America (RSNA) has championed the process and led the way with first developing RadLex as a standard lexicon and second organizing the first versions of “best practice” radiology report templates in the Radiology Reporting Initiative.

There has been resistance to the use of structured reports. Many users cite the perceived increased length of time of report generation and loss of the “personalized” component of the report. Some feel that the prose style of reporting is more akin to consulting and actually improves communication with the ordering provider. However, it has been shown that the referring clinicians actually prefer the itemized report content over prose (2). It is likely related to clearer, more consistent and objective evaluation and presentation of the imaging findings.

The itemized style can be paralleled to other medical documentation, such as the History & Physical (H&P). The H&P is always structured in similar format with discrete sections in the same order: Details of the presenting complaint, followed by past medical history, then review of symptoms, the vitals and physical examination, and so forth. The physical examination too is ordered in the same way with a general appearance followed by HEENT examination, cardiothoracic, abdominal, and so on. The structured radiology report can be looked at in this same fashion, equating the Impression section condensing the thoughts and recommendations of the radiologist to the clinical physician’s assessment and plan.

Even with the first iteration of reports organized by the RSNA, there is much room to improve. While all individual reports and reporting divisions are structured, there is little agreement across the entire spectrum of published templates on standard headings and element organization. Examples include use of “Clinical Indication” versus “Clinical Information” versus “Clinical History,” the location of the Impression section at the beginning or the end of the report, inclusion of date or time of examination, Observation versus Findings section, and inclusion or not of a Recommendation section. In addition, there is even greater variation in the clinical information content and organization across divisions and across modalities within the same division.

For the next step in evolution, it falls to a single entity or governing body to begin to pull together standardization across the spectrum of reports. One purpose of this manual is to provide such an effort for the entirety of the division of musculoskeletal imaging, to either lead the way for the rest of imaging or facilitate rapid conversion to any future adopted industry standard. Either way, it is strongly encouraged that individual institutions and practices organize their own internal efforts when pursuing structured reporting.

An example of this process is to put together a governance committee, comprising physician leaders for administrative, key clinical service, information technology, and quality assurance/quality improvement components of the practice. Such a group can provide standardization for service branding and billing assurance while maintaining a unified vision for the effort and identifying IT needs and optimization opportunities.

Organized beneath this governance group are the division leaders to champion and drive clinical content of the reports. These leads take the direction back from governance to their respective divisions and then work directly with IT resources to build the structured report content into the locally installed electronic reporting applications. It is important that these leads be direct users of the reports and reporting applications, so that they can identify pressure points of the systems. With this, they can then again work with IT to tweak the report templates to drive continuous gains in efficiency and accuracy within the limits of utilized reporting applications.

End user involvement is the first step to develop generalized buy-in and adoption. The users must have a voice and process to provider user feedback. Disregarding their opinion will not only limit buy-in but could also miss valuable efficiency gains and victories—only the end users can truly know how to best improve the system. Along that same line, it is important to facilitate the close interaction of clinical content experts with IT resources so that technical application knowledge can be leveraged to generate improvements that might not be readily apparent to a clinical end user. After all, each reporting application is different: Some with basic and some with advanced template tools to assist with structured reporting. Toward this, having a radiology or medical informatics position is becoming more and more valuable to physician groups and institutions.

After implementation, rapid cycle change processes should be engaged as often variant uses and short cuts can be developed that the application vendors would never have envisioned. If a template is never changed after implementation, then issues are likely being missed such as punctuation, misspelling, template field optimization, or even simple nonuse of the templates. Every effort should be made to continuously engage and reengage the users and IT sources of knowledge. In addition, it should be determined whether the reporting application or medical record can be leveraged to track template usage.

By completing these integral organizational components, focus may now be put on the main purpose of this manual—the codification of the clinical content. When reviewing available first-generation templates on the RSNA Radiology Reporting site, two basic organizational forms for findings/observations can be identified—issue centric and anatomy centric.

Issue centric reporting describes the most pertinent issue findings first, then subsequently “everything else.” The benefit of this style is that often the content most important to the original ordering provider is listed first whereas items that a referring specialist may not be concerned about come second. The issue centric style tends to presume that the pathology is known to some degree before imaging—which is not always true. The primary issue with this style is that the organization of the report content for the same examination, like an MR of the abdomen, can be strikingly different based on the reason the examination was done. In an age when patients with multiple medical problems may be followed by multiple specialists organized by a primary care physician, tracking findings across multiple reports can become problematic. Imagine a person imaged for abdominal pain found to have a renal mass, a liver mass, and an adrenal mass who then undergoes a renal mass imaging protocol, a liver mass imaging protocol, and an adrenal mass imaging protocol all with reports, although structured, organized differently—it could quickly become confusing trying to track any changes.

In contrast, the anatomy centric style handles this issue in a much more standardized fashion. The liver section is always in the same location. The renal section is always in the same location. And so on. The contents of each section may be more abbreviated or detailed based on the imaging protocol utilized, but the order is the same. Such organization then facilitates tracking findings across multiple reports and is the methodology utilized here.

Standardizing the format and content in this way will encourage the more thorough reading of the Findings sections of imaging reports. With familiarity to the layout, a referring or consulting provider can more rapidly refer to specific sections of their concern, even if not specifically discussed in the Impression section. This then allows the reading radiologist to tailor the Impression section to elements reflective of the reason for examination or urgent incidental discoveries while still facilitating identification of other findings of interest to other consumers of the report who may be referring to it for reasons entirely different than the original indication. Such structure also supports the generation of more complete reports, preventing indication or satisfaction of search from glossing over important elements of information ultimately eliminating the “you didn’t say anything about the xyz” phone call.

Implementing this process will create a standardized and codified report, with report details highlighting anatomy and pathology rather than the reporting radiologist preferences and style choices. More complete reports can be created while improving the ability of the clinical providers to consume the entirety of the report. Specific elements of interest can more readily be tracked through serial reports, even if not part of the original reason for examination. But potentially of greatest impact, this establishes the framework for computerized processing and sorting of report elements. The electronic medical record and imaging create enormous volumes of information; however, with vast components in a format not interpretable by a computer. By facilitating computerized consumption of the report, we set the stage for large-scale (even population based) evaluations pushing the future of medicine from anecdotal to true outcome-based practice.

The following chapters will detail the proposed best practice report templates for musculoskeletal imaging reports. Adhering to the above principles, maximizing reporting application abilities, engaging the reporting and clinical users, and applying the structure and content of these provided templates will advance the specialty of radiology into the next step of standardized structured reporting and toward the future of medicine.

REFERENCES

1. Danton GH. Radiology reporting, changes worth making are never easy. Appl Radiol. 2010;39(5):19–23.

2. Naik SS, Hanbidge A, Wilson SR. Radiology reports: Examining radiologist and clinician preferences regarding style and content. AJR Am J Roentgenol. 2001;176(3):591–598.

Only gold members can continue reading. Log In or Register to continue