Environmental Factors Predictive of No-Show Visits in Radiology: Observations of Three Million Outpatient Imaging Visits Over 16 Years

PurposeTo evaluate the impact of environmental and socioeconomic factors on outpatient cancellations and “no-show visits” (NSVs) in radiology.Materials and MethodsWe conducted a retrospective analysis by collecting environmental factor data related to outpatient radiology visits occurring between 2000 and 2015 at our multihospital academic institution. Appointment attendance records were joined with daily weather observations from the National Oceanic and Atmospheric Administration and estimated median income from the US Census American Community Survey. A multivariate logistic regression model was built to examine relationships between NSV rate and median income, commute distance, maximum daily temperature, and daily snowfall.ResultsThere were 270,574 (8.0%) cancellations and 87,407 (2.6%) NSVs among 3,379,947 scheduled outpatient radiology appointments and 575,206 unique patients from 2000 to 2015. Overall cancellation rates decreased from 14% to 8%, and NSV rates decreased from 6% to 1% as median income increased from $20,000 to $120,000 per year. In a multivariate model, the odds of NSV decreased 10.7% per $10,000 increase in median income (95% confidence interval [CI]: 10.3%-11.1%) and 2.0% per 10°F increase in maximum daily temperature (95% CI: 1.3%-1.6%). The odds of NSV increased 1.4% per 10-mile increase in commute distance (95% CI: 1.3%-1.6%) and 4.5% per 1-inch increase in daily snowfall (95% CI: 3.6%-5.3%). Commute distance was more strongly associated with NSV for those in the two lower tertiles of income than the highest tertile (P < .001).ConclusionEnvironmental factors are strongly associated with patients’ attendance at scheduled outpatient radiology examinations. Modeling of appointment failure risk based on environmental features can help increase the attendance of outpatient radiology appointments.     

Implementation and Utilization of a “Pink Card” Walk-In Screening Mammography Program Integrated With Physician Visits

PurposeThe aim of this study was to evaluate the implementation and utilization of the Pink Card program, which links a physician-delivered reminder that a woman is due for screening mammography (SM) during an office visit with the opportunity to undergo walk-in screening.MethodsIn 2016, the authors’ community-based breast imaging center provided physicians from three primary care and obstetrics and gynecology practices located in the same outpatient facility business card–sized Pink Cards to offer women due for SM during office visits. The card includes a reminder that screening is due and can be used to obtain SM on a walk-in basis. The primary outcome measure was the proportion of women who used Pink Cards among all screened women over 2 years. Independent predictors of Pink Card utilization were evaluated using multivariate logistic regression analyses.ResultsAmong 3,688 women who underwent SM, Pink Cards were used by 19.9% (733 of 3,688). Compared with women with prescheduled screening visits, Pink Card users were more likely to be Asian (odds ratio [OR], 1.37; P =.032), Black (OR, 2.05; P = .002), and Medicaid insured (OR, 1.71; P = .013) and less likely to use English as their primary language (OR, 2.75; P = .003). Additionally, Pink Card users were less likely to be up to date for biennial SM compared with women with prescheduled visits (31.9% [234 of 733] versus 66.6% [1,968 of 2,955], P < .001).ConclusionsThe Pink Card walk-in SM program can improve screening access, particularly for racial/ethnic minorities and Medicaid-insured patients. Expansion of this program may help reduce disparities and increase engagement in breast cancer screening.     

Residents’ Perceptions of Usage of the Current Alumni and Attending Network for a Formal Mentorship Program in an Academic Affiliated Community Hospital Radiology Residency

Mentor-mentee relationships within radiology residencies can add significant value to a resident’s overall experience. Studies demonstrate that mentorship programs can increase satisfaction for residents and faculty alike by reducing stress, easing career related decisions, increasing involvement with research, improving teaching and communication skills, and finally increasing leadership roles. In a survey of radiology program directors, 85% of program directors find such a program beneficial but only 57% have a formal program in place. Totally, 42% of program directors believe a structured mentorship program is necessary. Studies have also shown that female residents prefer female mentors. Alumni serve as an ideal group for resident mentorship as they do not face the pressures of internal faculty. No study to date in diagnostic radiology literature uses an alumni network in establishing a formal mentorship program. The objective of this study is to implement a formal mentorship program within an academic affiliated radiology residency by using program alumni and internal attending physicians for potentially increasing faculty engagement, improving resident morale, research opportunities, and networking for fellowship and job opportunities.     

Natural Language Processing of Radiology Reports in Patients With Hepatocellular Carcinoma to Predict Radiology Resource Utilization

ObjectiveRadiology is a finite health care resource in high demand at most health centers. However, anticipating fluctuations in demand is a challenge because of the inherent uncertainty in disease prognosis. The aim of this study was to explore the potential of natural language processing (NLP) to predict downstream radiology resource utilization in patients undergoing surveillance for hepatocellular carcinoma (HCC).Materials and MethodsAll HCC surveillance CT examinations performed at our institution from January 1, 2010, to October 31, 2017 were selected from our departmental radiology information system. We used open source NLP and machine learning software to parse radiology report text into bag-of-words and term frequency–inverse document frequency (TF-IDF) representations. Three machine learning models—logistic regression, support vector machine (SVM), and random forest—were used to predict future utilization of radiology department resources. A test data set was used to calculate accuracy, sensitivity, and specificity in addition to the area under the curve (AUC).ResultsAs a group, the bag-of-word models were slightly inferior to the TF-IDF feature extraction approach. The TF-IDF + SVM model outperformed all other models with an accuracy of 92%, a sensitivity of 83%, and a specificity of 96%, with an AUC of 0.971.ConclusionsNLP-based models can accurately predict downstream radiology resource utilization from narrative HCC surveillance reports and has potential for translation to health care management where it may improve decision making, reduce costs, and broaden access to care.     

Managing a Multisite Academic–Private Radiology Practice Reading Environment: Impact of IT Downtimes on Enterprise Efficiency

PurposeThe aim of this study was to assess whether the complex radiology IT infrastructures needed for large, geographically diversified, radiology practices are inherently stable with respect to system downtimes, and to characterize the nature of the downtimes to better understand their impact on radiology department workflow.MethodsAll radiology IT unplanned downtimes over a 12-month period in a hybrid academic–private practice that performs all interpretations in-house (no commercial “nighthawk” services) for approximately 900,000 studies per year, originating at 6 hospitals, 10 outpatient imaging centers, and multiple low-volume off-hours sites, were logged and characterized using 5 downtime metrics: duration, etiology, failure type, extent, and severity.ResultsIn 12 consecutive months, 117 unplanned downtimes occurred with the following characteristics: duration: median time = 3.5 hours with 34% <1.5 hours and 30% >12 hours; etiology: 87% were due to software malfunctions, and 13% to hardware malfunctions; failure type: 88% were transient component failures, 12% were complete component failures; extent: all sites experienced downtimes, but downtimes were always localized to a subset of sites, and no system-wide downtimes occurred; severity (impact on radiologist workflow): 47% had minimal impact, 50% moderate impact, and 3% severe impact.ConclusionsIn the complex radiology IT system that was studied, downtimes were common; they were usually a result of transient software malfunctions; the geographic extent was always localized rather than system wide; and most often, the impacts on radiologist workflow were modest.