Characteristics of COVID-19 Community Practice Declines in Noninvasive Diagnostic Imaging Professional Work

PURPOSE

The operational and financial impact of the widespread coronavirus disease 2019 (COVID-19) curtailment of imaging services on radiology practices is unknown. We aimed to characterize recent COVID-19-related community practice noninvasive diagnostic imaging professional work declines.

METHODS

Using imaging metadata from nine community radiology practices across the United States between January 2019 and May 2020, we mapped work relative value unit (wRVU)-weighted stand-alone noninvasive diagnostic imaging service codes to both modality and body region. Weekly 2020 versus 2019 wRVU changes were analyzed by modality, body region, and site of service. Practice share χ2 testing was performed.

RESULTS

Aggregate weekly wRVUs ranged from a high of 120,450 (February 2020) to a low of 55,188 (April 2020). During that -52% wRVU nadir, outpatient declines were greatest (-66%). All practices followed similar aggregate trends in the distribution of wRVUs between each 2020 versus 2019 week (P = .96-.98). As a percentage of total all-practice wRVUs, declines in CT (20,046 of 63,992; 31%) and radiography and fluoroscopy (19,196; 30%) were greatest. By body region, declines in abdomen and pelvis (16,203; 25%) and breast (12,032; 19%) imaging were greatest. Mammography (-17%) and abdominal and pelvic CT (-14%) accounted for the largest shares of total all-practice wRVU reductions. Across modality-region groups, declines were far greatest for mammography (-92%).

CONCLUSIONS

Substantial COVID-19-related diagnostic imaging work declines were similar across community practices and disproportionately impacted mammography. Decline patterns could facilitate pandemic second wave planning. Overall implications for practice workflows, practice finances, patient access, and payment policy are manifold.

National Trends in Oncologic Diagnostic Imaging

OBJECTIVE

To characterize national trends in oncologic imaging (OI) utilization.

METHODS

This retrospective cross-sectional study used 2004 and 2016 CMS 5% Carrier Claims Research Identifiable Files. Radiologist-performed, primary noninvasive diagnostic imaging examinations were identified from billed Current Procedural Terminology codes; CT, MRI, and PET/CT examinations were categorized as "advanced" imaging. OI examinations were identified from imaging claims' primary International Classification of Diseases-9 and International Classification of Diseases-10 codes. Imaging services were stratified by academic practice status and place of service. State-level correlations of oncologic advanced imaging utilization (examinations per 1,000 beneficiaries) with cancer prevalence and radiologist supply were assessed by Spearman correlation coefficient.

RESULTS

The national Medicare sample included 5,051,095 diagnostic imaging examinations (1,220,224 of them advanced) in 2004 and 5,023,115 diagnostic imaging examinations (1,504,608 of them advanced) in 2016. In 2004 and 2016, OI represented 4.3% and 3.9%, respectively, of all imaging versus 10.8% and 9.5%, respectively, of advanced imaging. The percentage of advanced OI done in academic practices rose from 18.8% in 2004 to 34.1% in 2016, leaving 65.9% outside academia. In 2016, 58.0% of advanced OI was performed in the hospital outpatient setting and 23.9% in the physician office setting. In 2016, state-level oncologic advanced imaging utilization correlated with state-level radiologist supply (r = +0.489, P < .001) but not with state-level cancer prevalence (r = -0.139, P = .329).

DISCUSSION

OI usage varied between practice settings. Although the percentage of advanced OI done in academic settings nearly doubled from 2004 to 2016, the majority remained in nonacademic practices. State-level oncologic advanced imaging utilization correlated with radiologist supply but not cancer prevalence.

Implications of achieving TICI 2b vs TICI 3 reperfusion in patients with ischemic stroke: a cost-effectiveness analysis

BACKGROUND

The benefit of endovascular thrombectomy (EVT) in stroke patients with large-vessel occlusion (LVO) depends on the degree of recanalization achieved. We aimed to determine the health outcomes and cost implications of achieving TICI 2b vs TICI 3 reperfusion in acute stroke patients with LVO.

METHODS

A decision-analytic study was performed with Markov modeling to estimate the lifetime quality-adjusted life years (QALY) of EVT-treated patients, and costs based on the degree of reperfusion achieved. The study was performed with a societal perspective in the United States' setting. The base case calculations were performed in three age groups: 55-, 65-, and 75-year-old patients.

RESULTS

Within 90 days, achieving TICI 3 resulted in a cost saving of $3676 per patient and health benefit of 11 days in perfect health as compared with TICI 2b. In the long term, for the three age groups, achieving TICI 3 resulted in cost savings of $46,498, $25,832, and $15 719 respectively, and health benefits of 2.14 QALYs, 1.71 QALYs, and 1.23 QALYs. Every 1% increase in TICI 3 in 55-year-old patients nationwide resulted in a cost saving of $3.4 million and a health benefit of 156 QALYs. Among 65-year-old patients, the corresponding cost savings and health benefit were $1.9 million and 125 QALYs.

CONCLUSION

There are substantial cost and health implications in achieving complete vs incomplete reperfusion after EVT. Our study provides a framework to assess the cost-benefit analysis of emerging diagnostic and therapeutic techniques that might improve patient selection, and increase the chances of achieving complete reperfusion.

Invasive Procedural Treatments for Symptomatic Uterine Fibroids: A Cost Analysis

PURPOSE

The aim of this study was to evaluate the contemporary use of procedural interventions to treat symptomatic uterine fibroids and assess associated health care system costs.

METHODS

Using the IBM Watson MarketScan Commercial Claims and Encounters database for 2009 to 2015 and relevant International Classification of Diseases diagnosis codes, women aged 18 to 55 years with clinically significant uterine fibroids were identified. Using Current Procedural Terminology codes, relevant procedural interventions were identified (hysterectomy, endometrial ablation, myomectomy, and uterine fibroid embolization [UFE]). Costs were defined as total actual payments by insurers and patients (per procedure and per episode of care) and were adjusted and compared using generalized linear models.

RESULTS

Of 241,757 invasive procedures for fibroids, hysterectomy was most common (76.5%), followed by endometrial ablation (14.5%), myomectomy (4.7%), and UFE (4.3%). Hysterectomy was more common in older women and those in rural areas (65.2% of patients <40 years of age, 77.6% of those 40-49 years of age, and 83.6% of those 50-55 years of age; 83.9% of patients outside versus 75.3% within metropolitan statistical areas). Per procedure, adjusted mean costs were $3,188 (95% confidence interval [CI], $3,114-$3,264) for hysterectomy, $2,781 (95% CI, $2,695-$2,870) for ablation, $4,436 (95% CI, $4,256-$4,623) for myomectomy, and $6,161 (95% CI, $5,736-$6,617) for UFE. Adjusted mean costs for entire episodes of care were $14,676 (95% CI, $14,496-$14,858) for hysterectomy, $6,702 (95% CI, $6,534-$6,875) for endometrial ablation, $14,791 (95% CI, $14,465-$15,125) for myomectomy, and $13,873 (95% CI, $13,182-$14,599) for UFE.

CONCLUSIONS

Of invasive procedures for symptomatic uterine fibroids, hysterectomy was used more frequently than endometrial ablation, myomectomy, and UFE combined. Per procedure and per episode, ablation was least costly. Costs per episode were similar for hysterectomy, myomectomy, and UFE.

Changes in Current Procedural Terminology Coding and Its Effect on Specialty-Level Utilization of Musculoskeletal Ultrasound

PURPOSE

Concerns regarding increasing utilization of non-vascular extremity ultrasound (US) imaging led to the Current Procedural Terminology (CPT) Editorial Panel separating a singular billing code into distinct comprehensive and focused examination codes with differential reimbursement. We explore this policy change's temporal association with utilization.

METHODS

Using Physician/Supplier Procedure Summary Master Files, we identified all nonvascular extremity US services billed for Medicare fee-for-service beneficiaries between 1994 and 2017. These included generic (CPT code 76880 from 1994 to 2010), complete (code 76881 from 2011 to 2017), and limited (code 76882 from 2011 to 2017) examinations. Annual utilization per 100,000 beneficiaries was computed and stratified by billing specialty. Compound annual growth rates were calculated.

RESULTS

Radiologists and podiatrists were the top 2 billing specialties for nonvascular extremity US examinations. From 1994 to 2010, radiologist services increased 6.1% annually. Following the 2011 code separation, radiologists' utilization increased 2.7% annually for complete and 12.3% for limited exams. Between 1994 and 2017, radiologists' market share decreased 72.8% to 40.4%. From 1994 to 2010, podiatrist services increased 87.1% annually. Following the code separation, podiatrists' annual utilization growth stabilized 0.4% for complete and 0.6% for limited exams. Podiatrists' market share was 9.1% in 2001, peaked at 31.3% in 2009, and declined to 14.3% in 2017.

CONCLUSIONS

Prior rapid growth in extremity nonvascular US for podiatrists slowed considerably following CPT code separation in 2011. Subsequent service growth has largely been related to less costly, focused examinations performed by radiologists. Further study may help better understand how CPT coding changes alter imaging utilization more broadly.

Leveraging the Mammography Setting to Raise Awareness and Facilitate Referral to Lung Cancer Screening: A Qualitative Analysis

PURPOSE

Despite compelling support for the benefits of low-dose CT (LDCT) screening for lung cancer among high-risk individuals, awareness of LDCT screening and uptake remain low. The aim of this project was to explore the perspectives of ACR mammography screening program directors (MPDs) regarding efforts to raise LDCT screening awareness and appropriate referrals by identifying high-risk individuals participating in routine mammography.

METHODS

MPDs were recruited from ACR-accredited mammography facilities to participate in semistructured interviews after the completion of an online survey. Interviews were conducted over the telephone, recorded, transcribed, and subsequently reviewed for accuracy. Twenty MPDs were interviewed, and 18 interviews were transcribed and included in the thematic analysis. A theme codebook was developed, and all interviews were coded using NVivo by two trained reviewers.

RESULTS

Key themes were organized into four broad domains: (1) general attitudes toward the integration of LDCT screening, (2) identifying mammography patients at high risk for lung cancer, (3) counseling about LDCT screening, and (4) strategies to identify high-risk women and increase awareness and knowledge of LDCT screening. Overall, MPDs recognized the benefits of integrating mammography and LDCT screening and were receptive to educating and referring women for LDCT screening. However, training and workflow changes are needed to ensure successful implementation.

CONCLUSIONS

Qualitative data suggest that MPDs are amenable to leveraging the mammography setting to engage women about LDCT screening; however, additional tools, training, and/or staffing may be necessary to leverage the full potential of reaching women at high risk for lung cancer within the context of mammographic screening.

Changing National Medicare Utilization of Catheter, Computed Tomography, and Magnetic Resonance Extremity Angiography: A Specialty-focused 16-Year Analysis

To assess changing utilization of extremity angiography from 2001 to 2016, focusing on modalities and provider specialties. Medicare PSPS Master Files from 2001-2016 and POSPUF from 2016 were used to determine overall and specialty utilization of diagnostic catheter angiography (DCA), CT angiography (CTA), and MR angiography (MRA). From 2001 to 2016, extremity angiography increased from 1107 to 1590 extremities imaged per 100,000 beneficiaries, with rapid expansion of CTA (22 in 2001 to 619 in 2009; plateau of 645 in 2016), but declines in DCA (1039 to 914) and MRA (45 to 30). Over time, extremity angiography shifted from 94% DCA, 4% MRA, and 2% CTA to 58% DCA, 41% CTA, and 2% MRA. For radiologists, extremity angiography increased slightly (741 to 767) with increases in CTA (20 to 595) and large decreases in DCA (681 to 145), with MRA remaining low (40 to 27). Extremity angiography increased for cardiologists (197 to 349) and vascular surgeons (87 to 351), both overwhelmingly performing DCA. Radiologists' share of all extremity angiography shifted from 67% to 48%, with interventionalists (47%), generalists (43%), and abdominal radiologists (7.4%) providing most radiologist services in 2016. Throughout, radiologists were the dominant providers of CTA (89% to 92%) and MRA (89% to 90%). Extremity angiography utilization in Medicare beneficiaries increased nearly 50% from 2001 to 2016, largely related to CTA performed by radiologists. Of radiologists, interventionalists and generalists together render most services. Cardiologists and surgeons assumed a large share of DCA previously performed by radiologists.

Use and Cost of Actinic Keratosis Destruction in the Medicare Part B Fee-for-Service Population, 2007 to 2015

Importance

Actinic keratosis is prevalent and has the potential to progress to keratinocyte carcinoma. Changes in the use and costs of actinic keratosis treatment are not well understood in the aging population.

Objective

To evaluate trends in the use and costs of actinic keratosis destruction in Medicare patients.

Design, Setting, and Participants

A billing claims analysis was performed of the Medicare Part B Physician/Supplier Procedure Summary Master Files and National Summary Data of premalignant skin lesion destructions performed from 2007 to 2015 among Medicare Part B fee-for-service beneficiaries.

Main Outcomes and Measures

Mean number of actinic keratosis lesions destroyed and associated treatment payments in 2015 US dollars estimated per 1000 Medicare Part B fee-for-service beneficiaries. Data analysis was performed from November 2017 to July 2018.

Results

More than 35.6 million actinic keratosis lesions were treated in 2015, increasing from 29.7 million in 2007. Treated actinic keratosis lesions per 1000 beneficiaries increased from 917 in 2007 to 1051 in 2015, while mean inflation-adjusted payments per 1000 patients decreased from $11 749 to $10 942 owing to reimbursement cuts. The proportion of actinic keratosis lesions treated by independently billing nurse practitioners and physician assistants increased from 4.0% in 2007 to 13.5% in 2015.

Conclusions and Relevance

This study's findings suggest that actinic keratosis imposes continuously increasing levels of treatment burden in the Medicare fee-for-service population. Reimbursement decreases have been used to control rising costs of actinic keratosis treatment. Critical research may be warranted to optimize access to actinic keratosis treatment and value for prevention of keratinocyte carcinoma.

Disparities over Time in Volume, Day of the Week, and Patient Complexity between Paracentesis and Thoracentesis Procedures Performed by Radiologists versus Those Performed by Nonradiologists

PURPOSE

To compare the disparities between the paracenteses and thoracenteses performed by radiologists with those performed by nonradiologists over time. Variables included the volume of procedures, the days of the week, and the complexity of the patient's condition.

MATERIALS AND METHODS

Using carrier claims files for a 5% national sample of Medicare beneficiaries from 2004 to 2016, paracentesis and thoracentesis examinations were retrospectively classified by physician specialty (radiologist vs nonradiologist), day of the week (weekday vs weekend), and the complexity of the patient's condition (using Charlson comorbidity index scores). The Pearson chi-square and independent samples t-test were used for statistical analysis.

RESULTS

Between 2004 and 2016, the proportion of all paracentesis and thoracentesis procedures performed by radiologists increased from 70% to 80% and from 47% to 66%, respectively. Although radiologists increasingly performed more of both services on both weekends and weekdays, the share performed by radiologists was lower on weekends. For most of the first 9 years across the study period, radiologists performed paracentesis in patients with more complex conditions than those treated by nonradiologists, but the complexity of patients' conditions was similar during recent years. For thoracentesis, the complexity of patients' conditions was similar for both specialty groups across the study period.

CONCLUSIONS

The proportion of paracentesis and thoracentesis procedures performed in Medicare beneficiaries by radiologists continues to increase, with radiologists increasingly performing most of both services on weekends. Nonetheless, radiologists perform disproportionately more on weekdays than on weekends. Presently, radiologists and nonradiologists perform paracentesis and thoracentesis procedures in patients with similarly complex conditions. These interspecialty differences in timing and complexity of the patient's condition differ from those recently described for several diagnostic imaging services, reflecting the unique clinical and referral patterns for invasive versus diagnostic imaging services.

Transcatheter Dialysis Conduit Procedures: Changing National and State-Level Medicare Use Patterns over 15 Years

PURPOSE

To evaluate the changing use of transcatheter hemodialysis conduit procedures.

METHODS

Multiple Centers for Medicare & Medicaid Services datasets were used to assess hemodialysis conduit angiography. Use was normalized per 100,000 beneficiaries and stratified by specialty and site of service.

RESULTS

From 2001 to 2015, hemodialysis angiography use increased from 385 to 1,045 per 100,000 beneficiaries (compound annual growth rate [CAGR], +7.4%)]. Thrombectomy use increased from 114 to 168 (CAGR, +2.8%). Angiography and thrombectomy changed, by specialty, +1.5% and -1.3% for radiologists, +18.4% and +14.4% for surgeons, and +24.0% and +17.7% for nephrologists, respectively. By site, angiography and thrombectomy changed +29.1% and +20.7% for office settings and +0.8% and -2.4% for hospital settings, respectively. Radiologists' angiography and thrombectomy market shares decreased from 81.5% to 37.0% and from 84.2% to 47.3%, respectively. Angiography use showed the greatest growth for nephrologists in the office (from 5 to 265) and the greatest decline for radiologists in the hospital (299 to 205). Across states in 2015, there was marked variation in the use of angiography (0 [Wyoming] to 1173 [Georgia]) and thrombectomy (0 [6 states] to 275 [Rhode Island]). Radiologists' angiography and thrombectomy market shares decreased in 48 and 31 states, respectively, in some instances dramatically (eg, angiography in Nevada from 100.0% to 6.7%).

CONCLUSIONS

Dialysis conduit angiography use has grown substantially, more so than thrombectomy. This growth has been accompanied by a drastic market shift from radiologists in hospitals to nephrologists and surgeons in offices. Despite wide geographic variability nationally, radiologist market share has declined in most states.

Determining the Patient Complexity of Head CT Examinations: Implications for Proper Valuation of a Critical Imaging Service

PURPOSE

The head-computed tomography (CT) exam code was recently identified by policy makers as having a potentially overvalued resource value units (RVU). A critical aspect in determining RVUs is the complexity of patients undergoing the service. This study evaluated the complexity of patients undergoing head-CT.

METHODS

The 2017 Medicare PSPS Master File was used to identify the most common site for performing head-CT examinations. Given the most common location, the 5% Research Identifiable File, was then used to evaluate complexity of patients undergoing head CT on the same day as an emergency department (ED) visit based on the Evaluation & Management (E&M) "level" of these visits (1-least complex to 5-most complex patient) and the ICD-10 diagnosis coding associated with the billed head CT claims.

RESULTS

56.1% of head CT examinations were performed in the ED. Seventy percent of noncontrast exams performed in the ED were ordered in the most complex patient encounters (level 5 E&M visits). The most common ICD-10 code for head-CT without intravenous contrast billed with a level 5 E&M visit was "dizziness and giddiness," and for head-CT without and with intravenous contrast was "headache."

CONCLUSION

Head-CT is not only most frequently ordered in the ED, but also during the most complex ED visits, suggesting that the ICD-10 codes associated with such exams do not appropriately reflects patient complexity. The valuation process should also consider the complexity of associated billed patient encounters, as indicated by E&M visit levels.

Management of Unruptured Intracranial Aneurysms in Older Adults: A Cost-effectiveness Analysis

Background Unruptured intracranial aneurysms (UIAs) are relatively common and are being increasingly diagnosed, with a significant proportion in older patients (˃ 65 years old). Serial imaging is often performed to assess change in size or morphology of UIAs since growing aneurysms are known to be at high risk for rupture. However, the frequency and duration of surveillance imaging have not been established. Purpose To evaluate the cost-effectiveness of routine treatment (aneurysm coil placement) versus four different strategies for imaging surveillance of UIAs in adults older than 65 years. Materials and Methods A Markov decision-analytic model was constructed from a societal perspective. Age-dependent input parameters were obtained from published literature. Analysis included adults older than 65 years, with incidental detection of UIA and no prior history of subarachnoid hemorrhage. Five different management strategies for UIAs in older adults were evaluated: (a) annual MR angiography, (b) biennial MR angiography, (c) MR angiography every 5 years, (d) coil placement and follow-up, and (e) limited MR angiography follow-up for the first 2 years after detection only. Outcomes were assessed in terms of quality-adjusted life-years (QALYs). Probabilistic, one-way, and two-way sensitivity analyses were performed. Results Imaging follow-up for the first 2 years after detection is the most cost-effective strategy (cost = $24 572, effectiveness = 13.73 QALYs), showing the lowest cost and highest effectiveness. The conclusion remains robust in probabilistic and one-way sensitivity analyses. Time-limited imaging follow-up remains the optimal strategy when the annual growth rate and rupture risk of growing aneurysms are varied. If annual rupture risk of nongrowing aneurysms is greater than 7.1%, coil placement should be performed directly. Conclusion Routine preventive treatment or periodic, indefinite imaging follow-up is not a cost-effective strategy in all adults older than 65 years with unruptured intracranial aneurysms. More aggressive management strategies should be reserved for patients with high risk of rupture, such as those with aneurysms larger than 7 mm and those with aneurysms in the posterior circulation. © RSNA, 2019 Online supplemental material is available for this article. See also the editorial by Cloft in this issue.

Characteristics of MR Neuroimaging Services Billed by Radiologists versus Nonradiologists

BACKGROUND AND PURPOSE

Although most neuroimaging examinations are interpreted by radiologists, many nonradiologists provide interpretation services. We studied day of the week, site of service, and patient complexity differences for common Medicare MR neuroimaging examinations interpreted by radiologists versus nonradiologists.

MATERIALS AND METHODS

Using carrier claims files for a 5% sample of Medicare beneficiaries from 2012 to 2014, we identified all claims for brain and lumbar spine MR imaging examinations. Services were categorized by physician specialty, day of the week, and the site of service. Patient complexity was calculated using Charlson Comorbidity Indices. The χ² was performed to test statistical significance.

RESULTS

A provider specialty could be identified for 568,423 brain and lumbar spine MR imaging examinations. Of weekday examinations, radiologists interpreted 475,288 (92.3%), and nonradiologists, 39,510 (7.7%). Of weekend examinations, radiologists interpreted 52,028 (97.0%) and nonradiologists 1597 (3.0%). Radiologists interpreted 145,904 (98.7%) examinations in the inpatient hospital and emergency department settings versus 1882 (1.3%) by nonradiologists. Of all examinations, 44,547 of those interpreted by radiologists (8.4%) were on the most clinically complex patients versus 2139 (5.2%) for nonradiologists. All interspecialty differences for day of the week, the site of service, and patient complexity were statistically significant (P < .001).

CONCLUSIONS

Although radiologists interpret most common MR neuroimaging examinations for Medicare beneficiaries, in contrast to nonradiologists, they disproportionately render those services on weekends, in higher acuity sites, and on more complex patients. To optimize access and minimize disparities in necessary neuroimaging, quality metrics should consider such service characteristics.

Use of Breast Cancer Screening and Its Association with Later Use of Preventive Services among Medicare Beneficiaries

Purpose To retrospectively assess whether there is an association between screening mammography and the use of a variety of preventive services in women who are enrolled in Medicare. Materials and Methods U.S. Medicare claims from 2010 to 2014 Research Identifiable Files were reviewed to retrospectively identify a group of women who underwent screening mammography and a control group without screening mammography in 2012. The screened group was divided into positive versus negative results at screening, and the positive subgroup was divided into false-positive and true-positive findings. Multivariate logistic regression models and inverse probability of treatment weighting were used to examine the relationship between screening status and the probabilities of undergoing Papanicolaou test, bone mass measurement, or influenza vaccination in the following 2 years. Results The cohort consisted of 555 705 patients, of whom 185 625 (33.4%) underwent mammography. After adjusting for patient demographics, comorbidities, geographic covariates, and baseline preventive care, women who underwent index screening mammography (with either positive or negative results) were more likely than unscreened women to later undergo Papanicolaou test (odds ratio [OR], 1.49; 95% confidence interval: 1.40, 1.58), bone mass measurement (OR, 1.70; 95% confidence interval: 1.63, 1.78), and influenza vaccine (OR, 1.45; 95% confidence interval: 1.37, 1.53). In women who had not undergone these preventive measures in the 2 years before screening mammography, use of these three services after false-positive findings at screening was no different than after true-negative findings at screening. Conclusion In beneficiaries of U.S. Medicare, use of screening mammography was associated with higher likelihood of adherence to other preventive guidelines, without a negative association between false-positive results and cervical cancer screening.

Can You Do Health Disparities Research with Publicly Available Datasets?

RATIONALE AND OBJECTIVES

Given the growing importance of identifying and reducing health disparities, it is important for radiologist researchers to engage in this space to promote evidence-based imaging disparities policy. However, researchers are often hindered by access to appropriate data to perform quality research.

MATERIALS AND METHODS

This paper reviews existing publicly available data sets that may be useful for performing imaging disparities research.

RESULTS

Multiple data sources are publicly available and have been used by previous researchers to examine imaging disparities.

CONCLUSIONS

This paper provides an overview of publicly available data sources that radiologists can use for imaging disparities research. Appropriate use of these data sources will require researchers to carefully consider the overall research question and level of analysis.