Physician self-referral and imaging use appropriateness: negative cervical spine MRI frequency as an assessment metric

MRI order appropriateness for chronic neck pain: Comparison of ordering practices and treatment outcomes for primary care physicians and specialists

Chronic neck pain is a common reason for doctor visits in the United States. This diagnosis can be evaluated through patient history, physical examination, and judicious use of radiographs. However, possible inappropriate magnetic resonance imaging (MRI) ordering persists. We hypothesized that no difference in ordering practices, ordering appropriateness, and subsequent intervention would be appreciated regarding physician specialty, location, patient characteristics, and history and physical exam findings. A multisite retrospective review of cervical spine MRI between 2014 and 2018 was performed. A total of 332 patients were included. Statistical analysis was used to assess MRI order appropriateness, detail of history and physical exam findings, and intervention decision-making among different specialties. If significant differences were found, multiple linear regression was performed to evaluate the association of MRI order appropriateness regarding physician specialty, location, patient characteristics and history, and physical exam findings. The significance level for all tests was set at <0.05 Orthopedic surgeons ordered MRIs most appropriately with an average American College of Radiology (ACR) score of 8.4 (p < 0.005). Orthopedic surgeons had more comprehensive physical exams as compared to the remaining specialties. The decision for intervention did not vary by physician specialty or ACR score, except for patients of pain medicine physicians who received pain management (p = 0.000). Orthopedic surgeons utilize MRI most appropriately and have more comprehensive physical exams. These findings suggest a need for increased physician education on what indicates an appropriate MRI order to improve the use of resources and further protect patient risk-benefit profiles. Further research elucidating factors to minimize negative findings in "appropriate" MRIs is indicated. Clinical significance: More detailed physical exams may lead to more appropriately ordered MRIs, subsequently resulting in surgery or procedures being performed when appropriately indicated. This suggests the need for increased physician education on when MRI ordering is appropriate for chronic neck pain to improve the use of resources and further protect patient risk-benefit profiles.

Private Equity Backed Radiology Considerations for the Radiology Trainee

Radiology trainees are served well by understanding the financial and operational aspects of the burgeoning phenomenon of private equity (PE) backed radiology and its implications on radiologist roles and remuneration. Consolidation in radiology has two major patterns, namely coalitions and PE-backed corporations, with distinct ownership, remuneration, and clinical decision-making dynamics. PE is defined by stock ownership, reduced base compensation, and greater conflicts of interest with respect to clinical decision-making given private investors' arguably larger appetites for profit. Data on PE's impact on radiology are scarce, but literature in other specialties suggests a potential for negative effects. Early career radiologist data point to growing concerns over this phenomenon's growing presence in radiology. Radiology trainees are encouraged to become more financially literate with respect to PE, as this model has the potential to disrupt the practice of radiology and the competitiveness of future talent pools.

Effects of Physician Experience, Specialty Training, and Self-referral on Inappropriate Diagnostic Imaging

BACKGROUND

Although previous research has demonstrated high rates of inappropriate diagnostic imaging, the potential influence of several physician-level characteristics is not well established.

OBJECTIVE

To examine the influence of three types of physician characteristics on inappropriate imaging: experience, specialty training, and self-referral.

DESIGN

A retrospective analysis of over 70,000 MRI claims submitted for commercially insured individuals. Physician characteristics were identified through a combination of administrative records and primary data collection. Multi-level modeling was used to assess relationships between physician characteristics and inappropriate MRIs.

SETTING

Massachusetts PARTICIPANTS: Commercially insured individuals who received an MRI between 2010 and 2013 for one of three conditions: low back pain, knee pain, and shoulder pain.

MEASUREMENTS

Guidelines from the American College of Radiology were used to classify MRI referrals as appropriate/inappropriate. Experience was measured from the date of medical school graduation. Specialty training comprised three principal groups: general internal medicine, family medicine, and orthopedics. Two forms of self-referral were examined: (a) the same physician who ordered the procedure also performed it, and (b) the physicians who ordered and performed the procedure were members of the same group practice and the procedure was performed outside the hospital setting.

RESULTS

Approximately 23% of claims were classified as inappropriate. Physicians with 10 or less years of experience had significantly higher odds of ordering inappropriate MRIs. Primary care physicians were almost twice as likely to order an inappropriate MRI as orthopedists. Self-referral was not associated with higher rates of inappropriate MRIs.

LIMITATIONS

Classification of MRIs was conducted with claims data. Not all self-referred MRIs could be detected.

CONCLUSIONS

Inappropriate imaging continues to be a driver of wasteful health care spending. Both physician experience and specialty training were highly associated with inappropriate imaging.

Effects of Physician Experience, Specialty Training, and Self-referral on Inappropriate Diagnostic Imaging

BACKGROUND

Although previous research has demonstrated high rates of inappropriate diagnostic imaging, the potential influence of several physician-level characteristics is not well established.

OBJECTIVE

To examine the influence of three types of physician characteristics on inappropriate imaging: experience, specialty training, and self-referral.

DESIGN

A retrospective analysis of over 70,000 MRI claims submitted for commercially insured individuals. Physician characteristics were identified through a combination of administrative records and primary data collection. Multi-level modeling was used to assess relationships between physician characteristics and inappropriate MRIs.

SETTING

Massachusetts PARTICIPANTS: Commercially insured individuals who received an MRI between 2010 and 2013 for one of three conditions: low back pain, knee pain, and shoulder pain.

MEASUREMENTS

Guidelines from the American College of Radiology were used to classify MRI referrals as appropriate/inappropriate. Experience was measured from the date of medical school graduation. Specialty training comprised three principal groups: general internal medicine, family medicine, and orthopedics. Two forms of self-referral were examined: (a) the same physician who ordered the procedure also performed it, and (b) the physicians who ordered and performed the procedure were members of the same group practice and the procedure was performed outside the hospital setting.

RESULTS

Approximately 23% of claims were classified as inappropriate. Physicians with 10 or less years of experience had significantly higher odds of ordering inappropriate MRIs. Primary care physicians were almost twice as likely to order an inappropriate MRI as orthopedists. Self-referral was not associated with higher rates of inappropriate MRIs.

LIMITATIONS

Classification of MRIs was conducted with claims data. Not all self-referred MRIs could be detected.

CONCLUSIONS

Inappropriate imaging continues to be a driver of wasteful health care spending. Both physician experience and specialty training were highly associated with inappropriate imaging.

Development and Performance of the Pulmonary Embolism Result Forecast Model (PERFORM) for Computed Tomography Clinical Decision Support

IMPORTANCE

Pulmonary embolism (PE) is a life-threatening clinical problem, and computed tomographic imaging is the standard for diagnosis. Clinical decision support rules based on PE risk-scoring models have been developed to compute pretest probability but are underused and tend to underperform in practice, leading to persistent overuse of CT imaging for PE.

OBJECTIVE

To develop a machine learning model to generate a patient-specific risk score for PE by analyzing longitudinal clinical data as clinical decision support for patients referred for CT imaging for PE.

DESIGN, SETTING, AND PARTICIPANTS

In this diagnostic study, the proposed workflow for the machine learning model, the Pulmonary Embolism Result Forecast Model (PERFORM), transforms raw electronic medical record (EMR) data into temporal feature vectors and develops a decision analytical model targeted toward adult patients referred for CT imaging for PE. The model was tested on holdout patient EMR data from 2 large, academic medical practices. A total of 3397 annotated CT imaging examinations for PE from 3214 unique patients seen at Stanford University hospitals and clinics were used for training and validation. The models were externally validated on 240 unique patients seen at Duke University Medical Center. The comparison with clinical scoring systems was done on randomly selected 100 outpatient samples from Stanford University hospitals and clinics and 101 outpatient samples from Duke University Medical Center.

MAIN OUTCOMES AND MEASURES

Prediction performance of diagnosing acute PE was evaluated using ElasticNet, artificial neural networks, and other machine learning approaches on holdout data sets from both institutions, and performance of models was measured by area under the receiver operating characteristic curve (AUROC).

RESULTS

Of the 3214 patients included in the study, 1704 (53.0%) were women from Stanford University hospitals and clinics; mean (SD) age was 60.53 (19.43) years. The 240 patients from Duke University Medical Center used for validation included 132 women (55.0%); mean (SD) age was 70.2 (14.2) years. In the samples for clinical scoring system comparisons, the 100 outpatients from Stanford University hospitals and clinics included 67 women (67.0%); mean (SD) age was 57.74 (19.87) years, and the 101 patients from Duke University Medical Center included 59 women (58.4%); mean (SD) age was 73.06 (15.3) years. The best-performing model achieved an AUROC performance of predicting a positive PE study of 0.90 (95% CI, 0.87-0.91) on intrainstitutional holdout data with an AUROC of 0.71 (95% CI, 0.69-0.72) on an external data set from Duke University Medical Center; superior AUROC performance and cross-institutional generalization of the model of 0.81 (95% CI, 0.77-0.87) and 0.81 (95% CI, 0.73-0.82), respectively, were noted on holdout outpatient populations from both intrainstitutional and extrainstitutional data.

CONCLUSIONS AND RELEVANCE

The machine learning model, PERFORM, may consider multitudes of applicable patient-specific risk factors and dependencies to arrive at a PE risk prediction that generalizes to new population distributions. This approach might be used as an automated clinical decision-support tool for patients referred for CT PE imaging to improve CT use.

Comparative effectiveness of convolutional neural network (CNN) and recurrent neural network (RNN) architectures for radiology text report classification

This paper explores cutting-edge deep learning methods for information extraction from medical imaging free text reports at a multi-institutional scale and compares them to the state-of-the-art domain-specific rule-based system - PEFinder and traditional machine learning methods - SVM and Adaboost. We proposed two distinct deep learning models - (i) CNN Word - Glove, and (ii) Domain phrase attention-based hierarchical recurrent neural network (DPA-HNN), for synthesizing information on pulmonary emboli (PE) from over 7370 clinical thoracic computed tomography (CT) free-text radiology reports collected from four major healthcare centers. Our proposed DPA-HNN model encodes domain-dependent phrases into an attention mechanism and represents a radiology report through a hierarchical RNN structure composed of word-level, sentence-level and document-level representations. Experimental results suggest that the performance of the deep learning models that are trained on a single institutional dataset, are better than rule-based PEFinder on our multi-institutional test sets. The best F1 score for the presence of PE in an adult patient population was 0.99 (DPA-HNN) and for a pediatrics population was 0.99 (HNN) which shows that the deep learning models being trained on adult data, demonstrated generalizability to pediatrics population with comparable accuracy. Our work suggests feasibility of broader usage of neural network models in automated classification of multi-institutional imaging text reports for a variety of applications including evaluation of imaging utilization, imaging yield, clinical decision support tools, and as part of automated classification of large corpus for medical imaging deep learning work.

Methods to detect inappropriate use of MRI and CT for musculoskeletal conditions: A scoping review

OBJECTIVE

Identify and evaluate methods suitable for detecting inappropriate use of MRI or CT in the musculoskeletal system.

DESIGN

Systematic review of studies that described methods to measure inappropriate use of MRI or CT in the musculoskeletal system. We used a multi-step strategy to classify identified methods into categories. These categories were then analyzed according to the data needed and their limitations.

ELIGIBILITY CRITERIA FOR SELECTING STUDIES

English or German language studies that measured inappropriate use of MRI or CT in the musculoskeletal system. Articles were also included if they reported a general approach to the measurement of inappropriate imaging regardless of body region. Expert opinions, unsystematic reviews, commentaries, articles without abstracts, and studies on cancer were excluded.

RESULTS

47 studies met the inclusion criteria. The categorization of the studies resulted in seven individual approaches to measure inappropriate use: (1) availability of meaningful diagnostic information; (2) predictors associated with imaging use; (3) comparison with guideline recommendations; (4) assessment by experts; (5) comparison or analysis of patients' paths; (6) comparison with surgery findings; (7) geographic variation. All these approaches have specific data requirements and individual advantages and disadvantages regarding risk of bias and needed data.

CONCLUSIONS

We could not find a single method of choice to detect inappropriate use of MRI or CT in the musculoskeletal system. A combination of different approaches is the preferred strategy to deal with the advantages and disadvantages of the individual methods.

Appropriateness of the Use of Magnetic Resonance Imaging in the Diagnosis and Treatment of Wrist Soft Tissue Injury

BACKGROUND

When diagnosing wrist soft tissue injury, the authors hypothesize that magnetic resonance imaging is used injudiciously and is associated with unnecessary cost.

METHODS

A retrospective review was conducted of patients aged 20 to 60 years who underwent magnetic resonance imaging for possible wrist soft tissue injury at a tertiary care center between 2009 and 2014. Treatment recommendation was classified as nonoperative, operative, or equivocal. If the magnetic resonance imaging-directed treatment recommendation differed from the pre-imaging recommendation, it was noted that the imaging influenced patient care (impact study). The cost estimate of an impact study was calculated by dividing the total studies performed by the number of studies that impacted the treatment recommendation and multiplying this value by the institutional wrist magnetic resonance imaging cost ($2246 in 2016).

RESULTS

One hundred forty patients were included. Magnetic resonance imaging affected treatment recommendation in 28 percent of patients. Independent predictors of impact on treatment recommendation were "question specific injury" (OR, 9.46; 95 percent CI, 3.18 to 28.16; p < 0.001) and "question scapholunate injury" (OR, 2.88; 95 percent CI, 1.21 to 6.88; p = 0.02). The only independent predictor of surgery was ordering physician (hand surgeon) (OR, 3.69; 95 percent CI, 1.34 to 10.13; p = 0.01). The cost of an impact study ordered by a non-hand surgeon versus a hand surgeon was $13,359 versus $6491, respectively.

CONCLUSIONS

The provider must carefully consider the pretest probability of ordering a study that will affect treatment recommendation. Injudicious screening with magnetic resonance imaging ($15,565) incurred a cost nearly seven times the cost of the one imaging scan ($2246) before impacting one treatment recommendation. In the current era of cost containment and bundled payment, diagnostic test probability must be appreciated to guide physician ordering practices.

Performance of a Machine Learning Classifier of Knee MRI Reports in Two Large Academic Radiology Practices: A Tool to Estimate Diagnostic Yield

OBJECTIVE

The purpose of this study is to evaluate the performance of a natural language processing (NLP) system in classifying a database of free-text knee MRI reports at two separate academic radiology practices.

MATERIALS AND METHODS

An NLP system that uses terms and patterns in manually classified narrative knee MRI reports was constructed. The NLP system was trained and tested on expert-classified knee MRI reports from two major health care organizations. Radiology reports were modeled in the training set as vectors, and a support vector machine framework was used to train the classifier. A separate test set from each organization was used to evaluate the performance of the system. We evaluated the performance of the system both within and across organizations. Standard evaluation metrics, such as accuracy, precision, recall, and F1 score (i.e., the weighted average of the precision and recall), and their respective 95% CIs were used to measure the efficacy of our classification system.

RESULTS

The accuracy for radiology reports that belonged to the model's clinically significant concept classes after training data from the same institution was good, yielding an F1 score greater than 90% (95% CI, 84.6-97.3%). Performance of the classifier on cross-institutional application without institution-specific training data yielded F1 scores of 77.6% (95% CI, 69.5-85.7%) and 90.2% (95% CI, 84.5-95.9%) at the two organizations studied.

CONCLUSION

The results show excellent accuracy by the NLP machine learning classifier in classifying free-text knee MRI reports, supporting the institution-independent reproducibility of knee MRI report classification. Furthermore, the machine learning classifier performed well on free-text knee MRI reports from another institution. These data support the feasibility of multiinstitutional classification of radiologic imaging text reports with a single machine learning classifier without requiring institution-specific training data.

Patient Attitudes Toward Orthopedic Surgeon Ownership of Related Ancillary Businesses

BACKGROUND

Physician ownership of businesses related to orthopedic surgery, such as surgery centers, has been criticized as potentially leading to misuse of health care resources. The purpose of this study was to determine patients' attitudes toward surgeon ownership of orthopedic-related businesses.

METHODS

We surveyed 280 consecutive patients at 2 centers regarding their attitudes toward surgeon ownership of orthopedic-related businesses using an anonymous questionnaire. Three surgeon ownership scenarios were presented: (1) owning a surgery center, (2) physical therapy (PT), and (3) imaging facilities (eg, Magnetic Resonance Imaging scanner).

RESULTS

Two hundred fourteen patients (76%) completed the questionnaire. The majority agreed that it is ethical for a surgeon to own a surgery center (73%), PT practice (77%), or imaging facility (77%). Most (>67%) indicated that their surgeon owning such a business would have no effect on the trust they have in their surgeon. Although >70% agreed that a surgeon in all 3 scenarios would make the same treatment decisions, many agreed that such surgeons might perform more surgery (47%), refer more patients to PT (61%), or order more imaging (58%). Patients favored surgeon autonomy, however, believing that surgeons should be allowed to own such businesses (78%). Eighty-five percent agreed that patients should be informed if their surgeon owns an orthopedic-related business.

CONCLUSION

Although patients express concern over and desire disclosure of surgeon ownership of orthopedic-related businesses, the majority believes that it is an ethical practice and feel comfortable receiving care at such a facility.