Tracking Radiation Exposure

I was pleased to see that the forward-looking Intermountain Healthcare system has made the following announcement.
Intermountain Healthcare has launched a program that measures cumulative radiation doses patients receive over their lifetime for medical treatment. Intermountain hospitals and clinics are first in the country to compile the cumulative radiation patients receive from about 220,000 higher-dose procedures and imaging exams each year, starting with exams performed in the last quarter of 2012.

I have discussed some of the issues about cumulative radiation exposure in prior blogs: Cumulative Radiation Exposure; Radiation Exposure; Joint Commission Alert on Radiation Risk.

California Senate Bill 1237 requires that exposures from CT radiation be calculated for a specific patient and to a specific organ. Software like Radimetrics will assist in capturing and calculating the delivered dose to the radiologists as part of the reading workflow. Modern imaging equipment is capable of sending the exposure settings along with the images. This data is then manipulated to be specific to the type of study and the physical characteristics of the patient.

We still have an overarching problem which is part and parcel of the fact that as one of the foremost developed nations we still do not have a personal health identifier that can be used nationwide. In fact we have short-sighted laws in place that forbid the federal government from proposing or developing a national identifier. As a result, all the valuable data we collect is located in individual data storage systems which at this point do not have a standard for sharing this data. Unless an individual or their healthcare proxy is extremely organized and persistent in obtaining the information and placing it in a personal health record (PHR) such as MyMedicalRecords, we are still essentially stuck where we were before electronic health delivery systems where nobody knows how much radiation a patient has received to date.

Joint Commission Alert on Radiation Risk

In June 2009 I blogged about measuring cumulative radiation exposure and yesterday the Joint Commission issued Sentinel Event Alert No. 47 addressing radiation risks of diagnostic imaging.

This Alert focuses on diagnostic radiation and does not cover therapeutic radiation or fluoroscopy. The Centers for Medicare & Medicaid Services (CMS) will require the accreditation of facilities providing advanced imaging services (CT, magnetic resonance imaging (MRI), positron emission tomography (PET), nuclear medicine) in non-hospital, freestanding settings beginning January 1, 2012.

The recommended approach to eliminate avoidable radiation dosing is:

• Right Test
• Right Dose
• Effective Processes
• Safe Technology
• Safety Culture

Thoughts on EHR Difficulties and Failures – #2

The transformational character of introducing an electronic health record (EHR) into an enterprise must be well planned and based on foundational work. Because an integrated environment is so interdependent, the absence of policies coordinating the building of interrelated modules and applications can lead to wasted effort and create rework once the unworkability or excessive cost of supporting the configuration is recognized.


Another aspect of dissipating effort and resources are the adoption of deadlines without having contingency plans that allow for the flexible deployment of the EHR to occur after it has successfully passed testing and user acceptance. Often the deadlines are linked to some other events whose prime focus in not the EHR. Similarly, an organization’s executive management must ensure that the project scope is maintained and vigilantly stop or downgrade unassociated projects or unnecessary scope creep.


Establishing the methods for change management (in the integrated environment), coordination and scheduling of build and testing environments, and establishing a sercurity model for user and application access can lead to signigicant simplification of security maintenance while safeguarding protected health information (PHI). Planning for the vetted release of information to indivuals and other healthcare providers should be addressed as part of the architectural design of the EHR. The trend line is that health information exchange (HIE) is becoming one of the main tentants of ARRA HITECH Meaningful Use requirements.

Thoughts on EHR Difficulties and Failures – #1

Having engaged in a number of EHR implementations, I am continually struck by the lack of forethought exhibited by many layers of healthcare management. This is all the more disconcerting considering the initial and long-term cost of an integrated electronic medical/health record (EHR) which usually runs into the millions of dollars.

The lack of planning starts with or even before the inception of an EHR project and lays the seeds for increasing the overall cost over the lifetime of the product by limiting its abilities in handling new practice areas, making it difficult to scale up (or down) in size, and ensure that the application is supportable with a reasonable number of IT and clinical staff.

Most facilities embarking on their first integrated EHR do not comprehend how transformational and even disruptive the implementation will be. The EHR is not primarily an IT project but social and business transformation of the organization. Just like communications and technology have flattened the earth, the EHR flattens the organization. Suddenly, a huge amount of sensitive information is just a click away and a sensible and comprehensive policy needs to be in place to ensure data safety while ensuring authenticated accessibility. Such a security policy needs to be mapped out and model adpoopted that ensures clinical usability while complying with the Sarbanes-Oxley Act, HIPPA and HITECH ARRA and a slew of specialty, regional and state requirements.

Getting the EHR to go-live is just the start of the journey. Operational funding for increasing data storage, increased auditing requirements, continuing process and application improvement, enhance change control for updating the core application and all the attached ancillary systems, increased training requirements for new and temporary employees, downtime and recovery policies, procedures and testing need to have long-term funding guaranteed.

Radiation Exposure

Although there are efforts to create a technological and procedural solutions (see California Bill 1237 and FDA efforts) for determining exposure to ionizing radiation exposure during diagnostic imaging procedures, these will fall short for many different reasons. Some can be traced to calculating the effective absorbed dose which is a function of multiple parameters ranging from the imaging system (amperage, voltage and efficiency of the system generating the radiation), the patient (size, obesity and exposed tissues/organs), and the expertise of the technologist setting up the patient.

The shortcommings should not stop the process of getting imaging vendors to include data from the machine in the Dicom header but they also need to focus more on the whole process of taking images. Radiologist can have a lot of input into the selecting the most appropriate modality, protocol, and use of contrast media. The technologists can ensure that the region of interest is not exceeded and the amperage and voltage are tweaked to deliver a good image with low radiation exposure. The vendors can provide training and built in protocols to optimize the image and area of interest.

The overarching philosophy governing the imaging group can ensure that radiologist and technologist agree on making the patient’s welfare central to their imaging service. Sharing the images with other facilities (Health Information Exchanges (HIE) and RHIOs) can reduce unnecessary re-imaging. Only participation with groups such as LeapFrog and regional HIEs will help move groups and organizations to to better and truly patient-centered care.

Getting the Images to the Emergency Room

Normally, attention is focused on workflows that provide the emergency department (ED) physicians with images and radiology reports in a timely fashion. There are a number of discrete workflows (including communications) that address the interplay of patient, radiology and the ED staff. This can usually be handled by fairly straightforward means using the available technologies.

Things get more complicated when an outlying emergency room needs to transfer the patient to a trauma center. Whether an ambulance or helicopter is used, it will usually take valuable time to get the patient to the new location, very often along with a CD/DVD or hardcopy films of radiology images that were taken in the initial ER. More time is then spent in re-evaluating the images and the patient.

If we leverage the Internet to securely transmit such images before the patient arrives at the trauma center, then surgeons and radiologist can evaluate, plan and prepare treatment ahead of the patient’s arrival. At Hennepin County Medical Center (HCMC), I set up VPN connections with a few of the top referring ERs. This allowed these facilities to send diagnostic quality Dicom images directly to HCMC where radiology, the ER, and the surgeons could review them. Since its inception, this service has been expanded to half a dozen referring facilities.

New Mexico has a system called IMEDCON (NMsoftware) that allows for instant digital file sharing without giving access to PACS. The system has been a success, allowing hospitals to manage head trauma cases without transfer. It is currently set up for 12 or so hospitals around the state to be able to share images with the regional Level One trauma center. A similar coverage system for stroke patients is currently underway.

As we move to expanding health information exchanges (HIE) we must also look at how we can quickly route such time critical data. Ideally the image set can be bundled into a single file using lossless compression and sent to the destination facility where it would automatically notify the trauma center of its availability. The facility would then open the study with their PACS tools and proceed with evaluating the incoming patient.

National Patient Identifier

We are moving towards an environment that will allow us to exchange important patient information with all providers and payers of healthcare. The key to successful health information exchange requires an accurate means for uniquely identifying an individual to ensure privacy and safety. Even last name, first name and date of birth do not guaranty uniqueness and the use of social security numbers is not allowed. Compounding this, congress passed shortsighted legislation which prohibits government from taking any action to bring about a national patient identifier.

On one hand we have HIPAA and high concern for patient privacy while on the other, a burgeoning number of sites where people freely trade personal data and their experience regarding various types of medical treatment. If the government has its hands tied by laws, the free market and people’s interest in obtaining the best healthcare may drive the adoption of a voluntary national patient identifier.

For instance, a patient could subscribe to a service that provides a unique identifier analogous to purchasing the rights to a domain name on the Internet. They could then provide that ID to any health provider (perhaps when providing their HIPAA releases) and the provider would then link this number to their own patient ID (frequently referred to as Medical Record Number or MRN in hospitals). This would allow providers anywhere in the US to query data under the health information exchange rules currently being established.