Value-Based Radiology: New Measurements Needed to Show Radiology's Contribution to Better Patient Health

Translation from Läkartidningen, Jörgen Nordenström & Peter Aspelin, February 2017.

Traditionally, health care has evolved through medical innovation and research, but the disruption of health care now underway is governed primarily by the forces and innovations beyond medicine. Today health care develops from information and communications technology (ICT) that creates new opportunities to measure, analyze and communicate. Health care now takes the path that has for so long been the benchmark in banking, retail and transportation. ICT and big data analytics has the potential to contribute to creating a health care that adapts to patient health care needs and preferences, improves coordination among providers, contributes to the quality development of a safer and more equitable system, provides greater accessibility, and drives evidence-based, efficient care.

This development takes place in all areas of health care; but here we will focus on the opportunities and challenges in radiology. Technical development in recent decades has witnessed explosive growth with the development of technologies such as CT, MRI, PET, MR-PET, PET-CT and ultrasound. This has not only changed the we way we make diagnoses but has also enabled a far more advanced interventional radiology. Radiology today is beneficial for diagnosis but also for treatment, forecasting, and understanding disease at the molecular level.

The introduction of digital radiology has led to improved processes. There are reductions in staff and better communication of both radiographs and radiology reports. In this environment, we are not only able to diagnose diseases that were previously difficult to diagnose but also able to treat patients with minimally invasive interventions. Such improvements generate increased costs which are in part driven by the utilization of various examination modalities for single, specific conditions. Variations across units and regions demonstrate the reality of a problematic notion that more expensive investigations are optimal.

The utilization of various examination modalities makes it difficult for the referring clinician to know with certainty which test is optimal for a specific task, leading to the use of several examinations instead of directly choosing the single most reliable test. This year, the annual Swedish Radiology Week featured Value-Based Radiology as its main theme. The concept has been developed in order to move from a dialog solely considering accomplishments and performance in radiological technology to one including the benefits that radiology creates for the patient. 

The concept of Value-Based Radiology has the aim to move the focus from radiological technique and efficiency to patient benefit, and consequentially, from production to value.

Reasoning for previous focus on production rather than on patient value has been the remuneration system which most commonly is based on the number of examinations, the volume, for a specific type of investigation. Such remuneration can be described as counterproductive as health care should be producing as little as possible, but with the highest quality outcomes. A compensation system based on patient benefit is preferable.

Radiological production volume is simple to measure. But a more difficult task, one of critical importance, is to clarify the benefits generated by each examination modalities utilized. Extraneous investigative examinations are driven by both patient related and provider related factors. For providers, multiple examination methods may deliver a more complete observation indicating further that confidence is related to expensive, sophisticated methods that provide the most information possible. Anxieties related to potential mistakes and consequential reporting for incorrect, incomplete investigations may be another source. Patient related factors may include a desire to receive the best treatment without an awareness of associated costs.

There are several initiatives that aim to improve the indications within radiological studies. The Choosing Wisely Initiative (2) identifies studies related to specific diseases across various specialties that have a low or questionable value, refuting many common patient and physician led beliefs. A list on the organization's website includes over 400 measures that are of questionable value – about one third represent radiology. Examples include CT scan of the hip in chronic pelvic pain, CT without intravenous contrast in chronic headaches, CT scan of the lungs for suspected pulmonary embolism with a negative D-dimer, and urography in cases of suspected kidney tumor in patients with blood in their urine.

A more systematic initiative to achieve greater value and better utilization of resources is the development of specialty criteria, or Appropriate Use Criteria (AUC) (3). Various measures and treatments are scored based on treatment recommendations and best practices classified into three groups: ‘appropriate’, ‘may be appropriate’ and ‘rarely appropriate care’.

Measures such as Appropriate Use Criteria have the potential to streamline indications of many measures and to reduce regional variations in diagnostic imaging. Although an adaptation for Swedish practice would be necessary.  In the US, AUC has been used in some regions as a basis for reimbursement wherein the caregiver usually must justify procedures conducted within the category of ‘rarely appropriate’ before receiving payment. We do not seek such use in Sweden, but the principles underpinning AUC may be worth our interest. The starting point for Value-Based Radiology should follow evidence-based care programs, utilizing structured examinations and multidisciplinary collaboration whereby the appropriate methodology to a particle question is determined.

Information and communications technology will facilitate the use of artificial intelligence and provide better opportunities for quantified radiology reports. In the future it will also be possible to use structured, standardized reports as an integral part of care mapping.

A significant way to improve the quality and thus value of X-ray examination is to create better follow-up procedures. ICT can help to make comparisons between X-ray findings, surgical findings and pathology reports. Already IT-based systems compare primary and secondary radiology reports utilizing automatic feedback systems for identifying incorrect statements. Radiology will also lead forecasting with use of biomarkers.

Value-Based Radiology and value-based purchasing face many challenges – relevant methods of measurement must be developed (4). Key outcome indicators such as survival and other long-term dimensions are less suited to assess radiology value. Better outcome and process measures are correlated to improve patient value. Such indicators are those that: 1) are directly related to health (i.e. increased survival in lung or breast cancer screening), 2) measure reductions in health care costs (i.e. faster and more reliable diagnosis in acute appendicitis), and 3) measure hospital lengths of stays (faster diagnosis, time to initiate a treatment, evaluation of treatment efficacy, etc.). Other relevant measures include complication rates for interventional radiology, radiation doses, diagnostic yield from biopsies, etc.

Generally, the best strategy to assess radiology’s contribution to better patient value is one which ensures strategic adherence to evidence-based care and best practices, conducted in multidisciplinary teams. In this environment, unjustified variations in diagnostic imaging can be reduced and a better value can be created for patients.

References
1. Wiener DH. Achieving high-value imaging: challenges and opportunities. J Am Soc Echocardiogr 2014; 27: 1-7.
2. www.choosingwisely.org
3. http://www.acr.org/quality-safety/appropriateness-criteria
4. A Sarwar, G Boland, Monks A Kruskal JB. Metrics for radiologists in the era of value-based healthcare delivery. Radio Graphics 2015; 35: 866-878.

View fullsize

Incomprehensible Criticism Toward Health Care's Ongoing Transformation

Translation from Svenska Dagbladet Debatt, Jörgen Nordenström, Magna Andreen Sachs & Nina Rehnqvist, January 2017

Swedish health care is now undergoing a rapid transformation . Many traditional health care practices are challenged – health care´s DNA is being transformed. The development of information and communications technology and patients' active contributions of medical data therein are forces driving future health care toward improved patient-related value.

A debate last year considered the future of health care in Sweden and the future organization of care at Karolinska University Hospital. The focus has been on procurement issues, infrastructure costs, how decisions are taken and consultant fees. The most important aspect has, however, been overlooked – the need to improve quality in health care. Information and communications technology create new opportunities to measure, analyze, store, and utilize the myriad of data that health care produces. Health care can now be transformed to the extent that other industries, namely banking, retail and transportation, have long since achieved.

 Like it or not, new technology will increase transparency and patient empowerment, creating a structure that is driven by patient needs and demands rather than health care’s organizational design and available infrastructure. Several doctor associations, including the Swedish Medical Association and the Swedish Society of Medicine, have been critical to the development of care that challenges the traditional doctor-patient relationship and questions doctor authority. Those associations are also critical to the changes in care processes that now are being implemented at Karolinska University Hospital, claiming Value-Based Health Care is void of a scientific evidence base that proves the model actually works. Those statements are not true. An abundance of data exists validating that Value-Based Health Care initiatives can improve quality, and at the same time reduce costs. There seems, however, to be some degree of confusion regarding the meaning of this concept. This is perhaps not surprising because the focus has for so long concerned budgets and costs. The reference to the existence of Value-Based remuneration at Karolinska is completely incorrect. Karolinska has, like other acute care hospitals in Stockholm, a reimbursement system with no relationship to the care processes now being introduced.

It has been stated that the quality of care in Sweden is consistently good. Closer scrutiny shows, however, there is significant room for improvement. 10% of patients treated in hospitals develop an avoidable injury. 30-40% of health care is not processed in accordance with current scientific evidence, and many patients with chronic illness will be re-hospitalized within three months. There are efficiency and coordination problems and a 2-3 fold variation in care according to data in National Quality Registers. The potential for improvement and more efficient use of health care resources is thus great.

The paradigm shift that is now transforming health care is the emphasis on value for patients over care production. In the absence of quality data, patient value has been estimated. Reimbursement has been based on production data without regard to what care actually achieves in terms of medical outcomes, quality of life and patient experience. It is against this background that the concept of Value-Based Health Care has been developed. Founder Michael Porter, Harvard professor, maintains that instead of assessing the cost of an operation  or investigation we should shift to assessing patient related value. Value-Based Health Care is neither a strategy nor approach but rather a framework for actions, measures and methods in health care leading to better value for the patient. Achieving patient related quality is the core goal. Sweden, according to the Economist Intelligence Unit, has made more progress than other country toward Value-Based Health Care.

The once passive patient can have better access to medical data in this rapidly developing technological environment. Increased knowledge and engagement builds an entirely new person/patient-centered health care, one with involved and active patients. The health care of tomorrow is no local affair. Web services, mobile consultations, mobile apps and telemedicine can be used beyond the patient's immediate community. Already today, patients produce data from applications linked to wireless mobile sensors. Such sensors measure heart rhythm and disease activity in Parkinson's disease, allowing many treatment and follow-up procedures to be managed by patients themselves with assistance from caregivers. Patients will increasingly contribute to the creation of new medical knowledge, demanding a more impactful patient-physician relationship and moving medicine toward greater democratization and reduced paternalism.