Original article
Automatically Inserted Technical Details Improve Radiology Report Accuracy

https://doi.org/10.1016/j.jacr.2011.02.009Get rights and content

Purpose

To assess the effect of automatically inserted technical details on the concordance of a radiology report header with the actual procedure performed.

Methods and Materials

The study was IRB approved and informed consent was waived. We obtained radiology report audit data from the hospital's compliance office from the period of January 2005 through December 2009 spanning a total of 20 financial quarters. A “discordance percentage” was defined as the percentage of total studies in which a procedure code change was made during auditing. Using Chi-square analysis we compared discordance percentages between reports with manually inserted technical details (MITD) and automatically inserted technical details (AITD). The second quarter data of 2007 was not included in the analysis as the switch from MITD to AITD occurred during this quarter.

Results

The hospital's compliance office audited 9,110 studies from 2005-2009. Excluding the 564 studies in the second quarter of 2007, we analyzed a total of 8,546 studies, 3,948 with MITD and 4,598 with AITD. The discordance percentage in the MITD group was 3.95% (156/3,948, range per quarter, 1.5- 6.1%). The AITD discordance percentage was 1.37% (63/4,598, range per quarter, 0.0-2.6%). A Chi-square analysis determined a statistically significant difference between the 2 groups (P < 0.001).

Conclusion

There was a statistically significant improvement in the concordance of a radiology report header with the performed procedure using automatically inserted technical details compared to manually inserted details.

Introduction

Since 1996, the US Department of Health and Human Services (HHS) and its Office of Inspector General (OIG) have heightened their investigation of hospital and physician billing practices. This elevated vigilance has resulted in increased denials of improper claims and reduced payments to institutions and providers. Even so, in 2001, CMS improperly paid an estimated $12.1 billion in claims. The Department of Health and Human Services and OIG have put physicians on alert: it is necessary to fully and accurately document services and initiate compliance plans to reduce coding and billing errors. Failing to file accurate claims may result in investigation by both federal and state enforcement agencies [1].

The evolution of medicine requires physicians to adopt new skills. Now more than ever, the modern practice landscape requires physicians not only to master the art and science of medicine but to firmly understand the reimbursement and compliance issues surrounding the delivery of care [2]. The benefits of a compliance program include reducing fraud and abuse, enhancing health care providers' operations, improving health care services, and reducing overall health care costs by preventing the submission of false or inaccurate claims or cost information to federal and state health care programs [3]. The OIG recommends that every institution adopt and implement compliance programs that adhere to applicable federal and state laws and the program requirements of federal, state, and private health plans [4]. The OIG views compliance programs as core aspects of a health care system [5].

In radiology, formal reports are at the core of practice. These documents serve not only the primary purpose of clinical communication but are also the documents on which coding and billing are based. If a service is ever audited for regulatory compliance, the formal report will be the single strongest determining factor. If a service is not documented, it is difficult to prove that it occurred [6]. It is as important that the formal radiology report accurately describe the imaging findings as that it accurately document the procedure performed [7].

Historically, the guidelines for what constitutes a complete radiology report have not been uniform [8]. The structure and content of reports has varied depending on any number of factors, including the intended audience, case complexity, and the experience of the authors [9]. However, at the 2007 Intersociety Conference, participants recommended the use of structured reports. In addition to uniform communication, structured reports facilitate the extraction of key information necessary to document quality metrics linked to financial incentives requested by the Medicare program [7]. The development of an effective and automatic structured report templating system based on procedure codes has the potential to eliminate human oversights in providing the correct header for a report. Such an improvement yields positive effects in both clinical and billing accuracy.

The objective of this study was to evaluate the effect of the automatic insertion of technical details on the discordance between a reported performed radiologic study and the actual performed study. Before 2007, our institution's radiology department manually inserted procedure headers into reports for descriptive and billing purposes. Our dictation system is RadWhere, (Commissure, Inc, New York, New York). This software allows a procedure description to be automatically inserted into a report on the basis of accession coding. Individual staff radiologist report concordance data were compiled over a 4-year period, with approximately half of the data from before and half from after the hospital-wide use of the RadWhere reporting system. We evaluated the concordance of the reports before and after the inception of the automated report system. To our knowledge, this investigation is the first to evaluate the relationship between automated radiology reports and compliance.

Section snippets

Methods

This study was approved by the institutional review board. The requirement for informed consent was waived. Data were reviewed from the radiology billing compliance office between January 2005 and December 2009. The office reviewed random radiology reports on a quarterly basis. A total of 20 quarters were included. The migration from manual to electronic insertion of procedure details took place during the second quarter of 2007. This quarter was excluded from the analysis.

Before the switch,

Results

This investigation included 9,110 radiology examinations that were audited by the hospital compliance office. The numbers for the pre-AITD and post-AITD eras and the second quarter of 2007 were 3,948, 4,598, and 564, respectively. We excluded the examinations from the second quarter of 2007, when the switch occurred.

The percentage of discordance in the pre-AITD era was 3.95% (156 of 3,948; range per quarter, 1.5%-6.1%). According to the discordance of the pre-AITD period, it was expected that

Discussion

Given the rapid changes both inside and outside American medicine, as well as rising health care spending in the United States, a major focus of HHS is the heightened investigation of hospitals to detect fraudulent activity. This effort has reportedly resulted in increased denials of improper claims and reduced payments to providers [1].

The mission of OIG, as mandated by Public Law 95-452 (as amended), is to protect the integrity of HHS programs, as well as the health and welfare of the

References (10)

There are more references available in the full text version of this article.

Cited by (8)

  • Automated Structured Reporting for Thyroid Ultrasound: Effect on Reporting Errors and Efficiency

    2021, Journal of the American College of Radiology
    Citation Excerpt :

    Consequently, radiologists have begun focusing on increasing efficiency by using structured reporting. Methods to facilitate structured report construction have been described such as automated insertion of quantitative measurements [11] or technical details [12] into reports, and even automated structuring of the entire report [13]. If creation of the structured report becomes easier, adoption rates and radiologist satisfaction might both increase.

  • Autopopulation of Intravenous Contrast Type and Dose in Structured Report Templates Decreases Report Addenda

    2017, Journal of the American College of Radiology
    Citation Excerpt :

    Report addenda have previously been used as a “self-reporting” method for determining error rates in radiology reports, and erroneous techniques (including contrast errors) accounted for 14% of all addenda [2]. A previous study demonstrated the successful addition of technical details (eg, number of radiographic views) into the radiology report using a voice recognition software package [3]. For each examination code in the RIS, a phrase describing the technique was set to autopopulate into the radiology report.

  • The ACR intersociety committee: History, activities, and membership

    2013, Journal of the American College of Radiology
    Citation Excerpt :

    The ISC has drafted multiple letters over the years, including to the ABR, the Diagnostic Radiology Residency Review Committee, the Association of Program Directors in Radiology, the RSNA, and so on. Intersociety Committee meetings have also resulted in surveys (including the one reported herein) and have spurred new research [7,8]. The title of the 2010 ISC summer conference was “The Radiology Conglomerate: Optimizing the Structure and Function of the 50-Plus Radiology Organizations.”

  • Using auto population of X-ray procedure exam type in radiology reports to decrease reporting errors

    2018, Clinical Imaging
    Citation Excerpt :

    In addition to improving communication structure, there are additional benefits which can also be gained with radiology report standardization using templates [21–24]. Incorporating the use of configurable data elements (“tokens”) that can be prepopulated automatically into a report has obvious efficiency gains, since it offloads tasks that the radiologist would perform manually [25,26]. An example of this is the use of a token which can pull the procedure statement of an examination from the Radiology Information System (RIS- a software system which manages medical imaging and its associated data).

View all citing articles on Scopus
View full text