EHR vendors today may have limited capability to index and tie metadata to medical images, perform image lifecycle management functions, or share images outside of the EHR system. Document management systems may be functionally limited in relating priors for comparison, may not offer multidisciplinary relativity capabilities, and may not support video and associated audio. Thus, EHRs and document management systems can not be the single location for storage of all medical images today; a separate vendor neutral archive or picture archiving and communication system (PACS) remains necessary. However, EHRs or document management systems may be the single storage location for a given specialty within a health system if the images are not DICOM, if metadata indexing is not critical to that specialty, comparison across acquisition dates is not desired, availability of a continuous image record across specialties is not essential, or if resulting workflows heavily favor having result text and images consumed together clinically on a given user interface.
In EHRs, using an orders-based workflow requires orders creation and distribution to the lists of commonly requested or performed procedures of involved clinical staff. Providers may not want the extra responsibility of placing an order for the image-based workflow to support image storage. Most providers would prefer to encourage their staff to adopt the order placement similar to how they may encourage their in-clinic staff to populate an ultrasound modality with demographic metadata in preparation for image capture. Often, the non-provider staff that the provider would prefer place the order do not have clearance to place EHR orders because this level of system security would also provide them with access to order many unrelated procedures. Designing an imaging workflow must additionally consider the access securities of those who would use it.
Automated methods of spawning an image order from appointments, operative case requests, and procedure encounters may not be technically available in all systems. Automated methods may present further challenges in specialties acquiring clinical photos such as dermatology and emergency services, as the visit does not specifically identify the body region where photos may be acquired. Confusion with compliance, IT, or operations staff may also be noted, believing that the order to store images actually represents an order to perform a procedure; for example, a scope camera image storage order may be misperceived as the order to perform an endoscopic procedure.
The support for DICOM modality worklist capabilities, both on modalities and on image review software applications, among vendors outside of cardiology and radiology is inconsistent. Workgroups and development is ongoing in specialties such as ophthalmology and pathology . For both encounters- and orders-based image capture and storage, modalities may not have DICOM licenses, DICOM store capabilities, adequate modality storage capabilities, or networking to support the workflow. Additional challenges exist where DICOM MWL and IHE profiles, including scheduled work flow , support order placers which do not exist in an encounters-based imaging workflow.
In an encounters-based image archive model, solutions are expected to interpret and utilize a wide array of ADT events including A01s (Pt. Admit), A04s (Pt. Registration), A02s (Pt. Transfer), and A10s (Pt. Arrived) . Logic supporting workflows tied to a visit, such as inpatient movement/transfers throughout a hospital, are also required for distribution of patient demographics tied to the unit(s) where each device is used . When images are stored to the archive, patient validation criteria can only be gainfully applied based upon an open encounter for that patient, as in the instance of storing scope video associated with an otolaryngology procedure using an encounters-based methodology. For outpatient workflows, the encounter, or attending, provider associated with an ADT event is included within the HL7 message and is fundamental in generating provider worklists within an image viewer . Traditional PACS and vendor neutral archives (VNA) may not store physician data beyond the referring, reading, and occasionally ordering, provider.
In contrast with EHRs, many PACS do not support non-DICOM image storage and distribution, leading organizations to look to VNAs for Cross-Enterprise Document Sharing (XDS) content support or requiring the purchase of vendor-specific systems to manage content and output DICOM formatted image files to the larger, centralized archive . Although non-DICOM clinical images and multimedia may be managed more effectively through XDS workflows, VNAs are lacking in image workflow management tools like exceptions handling and content validation, serving solely as storage repositories for XDS content and images . Resolving challenges associated with specialties acquiring non-DICOM images still requires archive product development supporting encounters-based imaging if organizations are looking to manage visible light imaging without requiring orders, regardless of the format acquired or stored .
Depending on the method of storage, images from encounters-based workflows may be more difficult to locate or share outside of the institution. Some EHR and ECM systems have imaging content storage that does not easily accommodate record sharing or transfer of images outward. A more extensive image exchange discussion is detailed in the HIMSS-SIIM Joint Workgroup white paper “Considerations for Exchanging and Sharing Medical Images” . The most notable EHR and technical architecture considerations and implications of orders-based and encounters-based approaches are outlined in Table 3.