Implementing Shared, Standardized Imaging Protocols to Improve Cross-Enterprise Workflow and Quality

Abstract

Value-based imaging requires appropriate utilization and the delivery of consistently high-quality imaging at an acceptable cost. Challenges include developing standardized imaging protocols, ensuring consistent application by technologists, and monitoring quality. These challenges increase as enterprises grow in geographical extent and complexity through mergers or partnerships. Our imaging enterprise includes a university hospital and clinic system, a large county hospital and healthcare system, and a pediatric hospital and health system. Studies across the three systems are interpreted by one large academic radiology group with expertise in various subspecialties. Our goals were as follows: (1) Standardize imaging protocols; (2) adapt the imaging protocols to specific modalities and available equipment; and (3) disseminate this knowledge across all of the sites of care. Our approach involved three components: (1) facilitation of imaging protocol definition across subspecialty radiologist teams; (2) creation of a database which links the clinical imaging protocols to the scanner/machine specific acquisition protocols; and (3) delivery of a protocol library and updates to all users regardless of location. We successfully instituted a process for the development, implementation, and delivery of standardized imaging protocols in a complex, multi-institutional healthcare system. Key elements for success include (1) a Project Champion who is able to articulate the importance of protocol standardization in improving the quality of patient care, (2) strong, effective modality-specific operational committees, (3) a Project Lead to manage the process efficiently, and (4) an electronic publishing of the protocol database to facilitate ease of access and use.

Background

Value-based imaging requires the delivery of high-quality, consistent, and appropriate imaging at an acceptable cost [1]. Common challenges include variation in the protocolling of imaging examinations and inconsistent implementation and execution of these protocols at the time of image acquisition. Sample imaging protocol sets have been developed by the American College of Radiology and many subspecialty groups, although these are not uniformly followed, and do not span every available scanner make and model in all possible configurations. In the case of complex computed tomography (CT) and magnetic resonance imaging (MRI), and even ultrasound (US) exams, there is potential for considerable variability in the interpretation of these recommendations by radiologists. Moreover, this variability may occur ad hoc with little opportunity to identify and decrease these variations which could result in value to the patient (such as improved diagnostic ability and outcomes) or value to operations (such as imaging efficiency and utilization). A process to create local consensus and to identify, develop, and implement site-specific optimal protocols is required.

The issue of protocol variability is further compounded in large, geographically distributed healthcare systems in that even if a recommended protocol has been created, there may not be a mechanism to effectively update and distribute it to the various points of care. Thus, a patient might undergo imaging with different image acquisition protocols based simply on where the imaging took place or which machine was used, rather than the nature of the clinical indication. A mechanism is needed to provide consistent, up-to-date imaging protocols at the point of care and to adapt them to the specific equipment available at each location.

Finally, effective communication with all technologists is essential to achieving the desired outcome (effective and consistent application) of the imaging protocol. This includes engagement of the technologist leaders in the design of protocols, the efficient dissemination of protocols, and training and implementation of processes for quality control.

Our imaging enterprise includes a state university hospital and clinic system, a large county hospital and healthcare system, and a pediatric hospital and health system. Studies across the three systems are interpreted by one large academic radiology group with expertise in various subspecialties with the mission to offer a single, uniform, high-level standard of care independent of visit site and/or time of the day. Our goals were as follows: (1) Standardize imaging protocols; (2) adapt the imaging protocols to specific modalities and available equipment; and (3) disseminate this knowledge across all of the sites of care.

Methods

General Approach

Creation of standardized protocols was identified as a Quality Improvement project for the Department of Radiology. The initial focus of the project involved CT, MR, Ultrasound, and Nuclear Medicine protocols.

Environment

The Department of Radiology provides professional services for three institutions: a university hospital, a public hospital system, and a children’s hospital system. The university hospital system includes the hospital with 460 beds and five outpatient imaging centers. The public hospital system consists of a hospital with 862 beds and nine outpatient facilities. The children’s hospital system includes two hospitals with a total of 674 beds and two outpatient facilities. The Department of Radiology has 133 clinical faculty members, 52 residents, and 27 fellows. The clinical volume for fiscal year 2017 was as follows: plain films 508,795, CT 177,860, ultrasound 122,693, nuclear medicine 13,440, positron emission tomography (PET) 2844, and MR 70,908.

Approach to standardizing clinical imaging protocols: A single individual with an extensive background as a MR scientist, MR technologist, and radiology manager, designated to be the “Project Lead,” was tasked with reviewing existing protocols and creating radiologist and technologist workgroups to develop standardized protocols (Fig. 1). These were reviewed and approved by the lead radiologist responsible for each modality and subspecialty area using a structure similar to that recently described [2].

Fig. 1.

Protocol standardization process: The overall process is shown above

Detailed processes were created for both initiating a new protocol and modifying an existing protocol. The process for creating a new CT protocol is shown in Fig. 2. During the process, overlapping protocols were combined, duplicative/outdated protocols were eliminated, and new protocols were developed to fill perceived clinical care gaps.

Fig. 2.

a Input form for initiating protocol change. b Flowchart illustrating process to initiate new CT protocol

Approach to developing machine specific protocols

The second step involved the creation of a database (Microsoft Access) linking clinical imaging protocols to machine-specific CT acquisition protocols. The clinical imaging protocol reflects clinical elements of the process (such as IV access, contrast dosing, electronic order details, and other special guidance instructions) and the intended scan output (imaging phases, radiation dose, technologist post-processing); these elements were largely intended to be as standard as possible regardless of imaging location and equipment (Fig. 3a). The machine protocol addresses the machine-specific settings to accomplish the actual imaging, which would vary appropriately based on manufacturer and modality (Fig. 3b).

Fig. 3.

a Clinical imaging protocol providing the generalizable guidance for completion. The hyperlink (arrow) facilitates navigation to the equipment specific machine protocol. Other hyperlinks link out to associate guidance documents. Output from the database showing the contrast administration and other details for technologist. b Machine-specific protocol example. Output from the database showing more detailed information for the technologist

The table structure implemented for the Microsoft Access database utilized a master table with joined tables for those elements in which fixed options were required. One of the most important was the official list of Radiology Information System (RIS) orders, which allowed the technologists to accurately match the clinical imaging protocol to the billable order. The RIS order names largely follow the CPT code language structure, and many protocols could be performed for any given order; however, only a specific order would match any single given protocol. Other elements standardized in this fashion included contrast agents, phase naming, and reconstruction algorithm. The applicable machine protocol name also utilized its own table to allow cross-linking from the clinical imaging protocol (with details on how the exam was performed and how the order would be processed) with the machine protocol (with details on the modality specific machine settings which might vary from one piece of equipment to another). Discrete fields were used for important standard elements such as rate/volume of contrast, field of view, and desired reconstruction planes/slice thicknesses. Finally, database queries were utilized for tracking the annual protocol review cycle and other such investigative concerns.

Implementing delivery system for information to the point of care

To facilitate sharing the information and protocols across multiple sites, the protocol library was implemented on a Microsoft SharePoint platform, which was made available to medical and technical staff at all institutions. At launch, all healthcare systems used this library as the “source of truth” for protocols, although web navigation to the library would originate in each system’s internal systems (Fig. 4).

Fig. 4.

SharePoint screen at the point of care: The top panel shows the protocol library table which is managed by the Project Lead and is not seen by the end user. The lower panel shows the presentation of the filtered data to facilitate ease of use

Process for change management and quality control

Modality-specific operational committees were developed to institute change management controls for ongoing protocol maintenance and to lay the foundation for quality assurance practice for utilization and adherence to established protocols. These committees included representation from the radiologist professional group, administrative and technologist members from the healthcare systems, and medical physics; this collaborative approach facilitated institutional buy-in that aided implementation. Furthermore, a process was implemented to maintain competencies among technical staff that takes into account the periodical optimization of protocols.

Results

Evaluation of existing state of protocols

A total of 606 imaging protocols were developed and updated using this process. All protocols in use prior to initiation of the project were collected for review and analysis (899), and duplicates were removed resulting in 651 protocols for subsequent redesign (Table 1).

Table 1.

Protocol characteristics

Modality	Number pre-implementation	Number post-implementation
MR	372	341
CT	222	142
US	84	63
XR	80	51
NM	141	54

For example, in nuclear medicine, there were initially 141 protocols. Forty-two protocols were identified as no longer applicable in our current clinical environment and were retired. Forty-eight were removed by combining differently named but similar protocols into new standardized protocols. Other less frequent reasons included changes in referring provider mix, changes in imaging hardware, and standardization of contrast agents or radiopharmaceuticals.

Protocol redesign process

In order to understand the existing process, the Project Lead met with the different modality managers to collect the clinical acquisition protocols. Despite explaining the goals of the project, the initial response to collecting protocols varied from highly cooperative to highly resistant. At times, the intervention of modality directors or divisional chiefs was necessary to collect existing protocols. The management of clinical protocol documentation varied among sites. In some cases, clinical protocols existed only in the scanner for the technologist to follow, while at other times, they existed only on paper. Electronic delivery of protocols was inconsistent and primarily through pdf or similar scanned documents.

Resistance arose from portions of our practice which had worked as independent, siloed efforts. The Department Chair championed standardizing our practice so that patients experienced the same care regardless of site, breaking down such barriers. In addition, the project lead reached out to individuals and educated them regarding the overall goals and scope of the project. Thus, by “winning each heart and mind,” engagement was improved and resistance melted away.

Once the protocols were collected, they were examined and used in the standardization process:

1. Outdated protocols (those no longer used or related to completed research projects) were eliminated and an annual review process for removal of protocols established.

2. Potential duplicates were carefully reviewed for any machine specificity. If no clear specificity was present, then a standard single protocol was developed and implemented.

3. The final set of protocols was reviewed by individual modality work groups for final approval.

Implementation of standardized imaging protocols

The protocols were stored in a SharePoint library with metatags to facilitate multiple ways of filtering and display. For example, protocols can be filtered by modality, body part, or radiology sub-specialty to support differing needs on different pages in the SharePoint site. Displaying the same source content in multiple ways ensures a single source of truth and minimizes effort when changes in protocols occur.

An added advantage of the use of SharePoint is the ability to utilize standard SharePoint and Google analytics to evaluate the use of the site. Total number of page views on the imaging protocol library grew to 11,207 (7% of the total site traffic) in the first 6 months. The protocol Library was utilized 53,717 times from July 1, 2015, to October 21, 2016, averaging 3357 clicks per month. The number of unique users varied from 54 to 121 over the past 6 months. Interestingly, analytics indicated that complicated protocols (e.g., MR Brachial Plexus) receive highest page views. In addition, new or recently changed protocols (e.g., CT Chronic Aortic Dissection/TEVAR) were also accessed frequently. During this transition time, the average number of protocol changes was approximately two per month.

Implementation of a standardized protocol change process

Ongoing maintenance of the clinical protocols required two processes:

1. All minor changes go through a “fast” change process which involves presentation during the modality protocol committee meeting with the approval of the modality Director. Our average turnaround time for fast change protocol implementation is about 48 h.

2. Any fundamentally new protocols are discussed at length for clinical necessity, patient time in the scanner, radiation dose/MR safety, Specific Absorption Rate (SAR) consideration by the physicists, and any other technical limitation identified by the modality technologists. The review and approval process for a new protocol could vary from 1 to 4 weeks depending on the complexity of the review.

Discussion

Our work indicates that it is possible to (1) create a responsive, sustainable process to standardize protocols in a large, multi-institutional environment; (2) instantiate machine protocols to facilitate consistent imaging across multiple imaging devices and vendors; and (3) develop an electronic tool to help support day-to-day operations and quality control.

The major challenges involved were organizational and not technical. Collecting the various modality protocols across the covered hospital systems was a significant but required task in order to standardize the imaging practice. Previous protocols had been developed locally and often reflected historical preferences and patterns of practice. Locating and organizing the protocols to reflect the most current versions stored in the scanner were quite challenging due to incomplete documentation and multiple system updates.

The Project Champion was the Director of Quality for the Department of Radiology who provided focus and facilitated involvement by Division and Modality Service chiefs and hospital administration. Moreover, the emphasis on quality improvement ensured the acceptance of the three institutions to have a single source of truth—the database. A critical success factor for the protocol library project was a Project Lead who had knowledge of the technologists’ needs and the ability to communicate effectively with radiologists and clinical operational committees.

The use of a database to help organize the machine-specific details for the imaging protocols was driven by the increasing number of CT scanners with multiple machine protocols which could potentially be chosen by a technologist. This can be a source of image acquisition variability which could increasingly impact image quantification and application of machine learning. Creating this database required detailed knowledge of technologist workflows and machine protocols which was greatly facilitated by the Project Lead’s prior experience as a technologist and technologist manager.

The choice of SharePoint for delivery of the protocols was driven by its flexibility and relative ease of maintenance compared to a dedicated webpage. The protocols library structure with its metatags is easy to filter and is presently maintained by individuals without formal IT training. We are planning to migrate the protocols library to Office 365 SharePoint in the near future to facilitate delivery of the library to mobile devices with minimal additional effort. Importantly, this approach will allow us to maintain a single library as the source of truth while expanding the availability of the most up-to-date protocols.

Our project was driven by previous work that demonstrated the importance of protocol standardization in providing a consistent radiologist, patient, and referring provider experience [3]. Implementing protocol standardization has been shown to require the appropriate organizational structure and stakeholder commitment, which we also found to be critical [2]. An important aspect of our work is the delivery of standardized, up-to-date protocol information electronically so that it is readily available in the reading room, office, imaging suite, and referring physician as needed. This has important implications as imaging services become more geographically distributed [4]. Moreover, it has potential to facilitate collaborative branding in partnered health systems in radiology [5].

Limitations

The standardization of clinical imaging protocols has been well received by both technologists and radiologist trainees due to the clearer understanding of the imaging intent and reduction of individual faculty radiologist variance. For example, MRI protocols standardized imaging planes, slice thickness, and acquisition techniques, resulting in more efficient protocolling for the radiologists and technologists with reduction in variability. Unfortunately, the series naming standardization has not yet made it into the PACS DICOM headers for hanging protocol standardization. However, the radiologist knows the series and reconstructions which are part of the standardized protocol, reducing calls to the technologist at the scanner for additional reconstructions or acquisitions.

This study does not attempt to measure the direct impact of our protocol standardization approach on workflow or patient outcomes. We plan future studies to assess the impact of our approach on these metrics.

In addition, future efforts will be focused on increasing availability of protocolling tools to mobile devices (particularly ultrasound in remote locations) and optimization of protocols for specific indications. More long-term will be the investigation of radiologist and referring physician decision support tools to facilitate the choice of the optimal standardized imaging protocol on the basis of patient characteristics.

Conclusion

We successfully instituted a process for the development, implementation, and delivery of standardized imaging protocols in a complex, multi-institution healthcare system. Key elements are [1] a Project Champion who is able to articulate the importance of protocol standardization in improving the quality of patient care; [2] strong, effective modality-specific operational committees; [3] a Project Lead to manage the process efficiently; and [4] electronic publishing of the protocol database to facilitate ease of access and use.