If you took the ancient Greek priestess Sybil of the Temple of Apollo at Delphi and combined her with Nostradamus, Edgar Cayce, Carnac the Magnificent, and Miss Cleo, you would come fairly close to meeting the requirements put on today's healthcare chief information officer (CIO).
Oracle, prophet, visionary, and seer combined, today's healthcare CIOs are asked time and again to perform divination on behalf of the enterprise that employs them. Get it right and they become near demigods. Get it wrong and they can bring an entire healthcare organization down with them.
In years past, CIOs might have one or two major initiatives they would work on, and then move on to the next project. Today, it's not unusual for CIOs to have several major projects and numerous smaller projects running concurrent to each other. This is why most CIOs establish teams and rely heavily on the input of others to help them through the various projects they have on their plates. One person simply cannot do it all. Many CIOs also have physician or clinician champions on board as well to help ensure that the decisions made meet both the short- and long-term objectives of the enterprise.
PACS can represent a particular challenge for CIOs. The technology has evolved from being confined to a silo within the radiology department to an enterprise-wide network that can be a cornerstone of a health system's enterprise data storage strategy -- thus requiring the close attention of the CIO. In this article, I'll address what CIOs should consider when managing PACS.
The evolution of PACS
In the 1990s, PACS consisted primarily of an onsite server to capture, display, and archive DICOM images from a single vendor. Enhanced workflow and the use of thin clients didn't come into play until the early 2000s, with data migration still proving to be a huge challenge.
It really hasn't been until the past five years or so that consolidated storage (aka vendor-neutral archives, or VNAs) started playing a much larger role in the enterprise storage of DICOM and non-DICOM images. Data migration has become a bit easier, but it's still costly and time-consuming, and because of standards adoption and advancement, best-of-breed solutions are being adopted.
PACS consultant Michael J. Cannavo.Today, PACS is much more complex than in years past, consisting of multiple servers with workstations powerful enough to run specialized software applications such as mammography, orthopedics, advanced visualization, products from multiple vendors, and more. Solutions are also being offered onsite, in the cloud, or with hybrid approaches. Most local area networks (LANs) are now up to 1 Gb/sec or more to address the large datasets that PACS deals with, and wide area networks (WANs) can often run at speeds of 50 Mb/sec or more to the radiologist's home and clinicians' offices.
Despite overcoming many of the technological barriers of the past, knowledge of where PACS technology is and where it is going remains key. Obtaining that knowledge, however, is not easy, even though there are many resources available to educate a CIO. Conflicting and confusing information is often presented in an effort to advance a company's agenda based on its product design rather than the actual market direction.
Further compounding the CIO's job is the task of actually understanding what PACS is today and where PACS is going. Papers have been published stating that "PACS is dead," when that is about as far from the truth as it gets.
Yes, PACS has evolved, just like a marriage evolves over time, and it doesn't come close to representing PACS of the past -- but it is very much alive. Call it electronic imaging, PACS, or whatever you want, without PACS radiology and cardiology departments would function at a much lower level than they do today.
The PACS Ball of Confusion
PACS is offered in many different configurations depending on the vendor. Most companies still offer a "generic" PACS configuration, but terms such as deconstructed PACS, reconstructed PACS, virtual PACS, PACS lite, and others show up routinely.
One CIO told me that trying to understand it all is like a take on the early '70s Temptations song "Ball of Confusion": "Deconstructed, reconstructed, RIS and PACS, software hacks, image sharing, who ya' daring, VNA -- who's to say? ... just a ball of confusion."
On top of that, the CIO also has to deal with clinicians who live in a BYOD (bring your own device) world and expect effortless connectivity to their smartphones, tablets, and laptops. Thankfully, the widespread acceptance of zero-footprint viewers has pretty much addressed that issue.
The main advantage of a zero-footprint viewer is the flexibility to launch studies directly in a Web browser, regardless of the type of browser, operating system, or device, as well as not having to install a new application on the computer to open a document or requiring an upgrade. Zero-footprint viewers offer several other benefits, but once again the CIO needs to understand these at least globally and be able to determine how important they are in the overall deployment of a PACS.
When a healthcare CIO looks at PACS, he or she looks at it from a much different perspective than that of the radiology or cardiology departments. Departments typically look at their PACS from a clinical and functional perspective: How will/can it work to improve the overall productivity of the department and hopefully improve patient care as well?
CIOs, on the other hand, look at how the system can be interconnected to other systems, and not just serve radiology but also primary care physicians, patients, and others. They also look at how the data generated by the system (i.e., data analytics) can improve overall department performance; how connecting the systems can help achieve higher stages of meaningful use, and with it related incentive payments; how the consolidation of systems such as archives can be achieved without compromising system performance; and similar areas.
Few would question that CIOs are concerned about the clinical and functional aspects of the systems, but they need to look at the bigger picture as well. It's also not just about existing technology like PACS but also related technologies like telemedicine/telehealth, and the impact these may have on network utilization to physician deployment to patient acceptance.
The CIO is also called on to determine if the resources needed to acquire and install new software or medical equipment are worth the investment from a financial and resource perspective. This includes training, the impact on the patient care workflow, and other areas. The PACS role might also be considered as part of a health information exchange (HIE).
Interoperability
How important is ensuring interoperability? Crucial is probably an understatement, especially as enterprises grow. Over the past 10 years, properties owned by Florida Hospital in Orlando, which is part of the Adventist Health System, more than doubled through acquisition or new construction, rising in value from $610 million to nearly $1.5 billion. Its crosstown rival, Orlando Health, more than tripled the value of its properties from $281 million to over $758 million. Larger entities such as HCA, which owns 168 hospitals and 116 freestanding surgery centers, grew its revenue over $3 billion from fiscal 2014 (to $40 billion from $37 billion), again largely reflecting growth through acquisition, expansion, and new construction.
As facilities grow, the clinical systems they use to provide patient care need to work with each other. This can be through the use of standards such as Integrating the Healthcare Enterprise (IHE), interface engines, or even customized application program interfaces (APIs), although CIOs tend to avoid the latter interfaces whenever possible due to the high costs associated with them and ongoing customization whenever an upgrade is done.
Considerations include where the systems reside (onsite, in the cloud using virtual machines, or a combination of both), how disaster recovery is addressed, whether business continuity is required, and other questions. Addressing organizational control of protected health information (PHI), migrating data storage or other services to the cloud, and buying an analytics product or software are all other areas in which healthcare CIOs have a say.
Lastly, not only do the systems need to work together, they need to be secure and fully HIPAA-compliant as well.
Other considerations
So what else is important to CIOs? As stated before, data analytics is very important to the optimization of clinical systems, but few companies really stress its importance. Radiology has a plethora of data that can be aggregated, digested, and used to create a useful metric, yet rarely is this done.
Data analytics that can help an organization with its future direction typically falls into two areas: business analytics and clinical analytics. Business analytics looks at the use and scheduling of equipment, and it also analyzes variances in radiology reports, monitoring the ordering patterns of clinicians and various parameters to determine turnaround time. Personnel utilization addresses not just radiology department staffing but also radiologist optimization as well.
Clinical analytics helps a location follow up on the actions taken after the patient leaves the department and closes the loop between the facility, radiologist, clinician, and patient. This can also address areas such as peer feedback from fellow clinicians, as well as determine under- or overutilization of a site, how multisite facilities can be optimized, and whether sites should be expanded or merged.
Radiation dose management and tracking is another area where information is king, and dose reporting is now mandatory in several states. The CIO needs to ensure that all state and federal guidelines are met, just as they are with mammography and its requirements to meet the Mammography Quality Standards Act (MQSA).
Radiation dose monitoring software provides an approximation of the radiation dose delivered to patients, but it also provides a detailed look at Web-based radiation dose trends and analytics. These can help identify peaks and establish averages per exam. The software also looks at protocol changes in real-time, generates cumulative patient reports, and much more. More-advanced systems allow monitoring of scanner image volume, volumetric trends, and other complex data, and identify maximum radiation dose in mSv for each exam. This can all help the CIO develop a better-performing and safer department.
The future
Understanding analytics doesn't end with what's available today, either. Works in progress such as the IBM project code-named Avicenna, demonstrated at RSNA 2015, bring imaging analytics to a new level. The futuristic workstation allows a radiologist to augment normal viewing of imaging studies with clinical and imaging summaries by collecting relevant clinical, textual, and imaging patient data from electronic health record systems. It would then analyze this multimodal content to detect anomalies and summarize the patient record, collecting relevant information pertinent to a chief complaint.
The results of anomaly detection would be fed into a reasoning engine using evidence from both patient-independent clinical knowledge and large-scale analysis of data from patients with similar measurements to arrive at a potential differential diagnosis. The resulting clinical summaries, along with the findings from imaging studies, would then be available to further assist the interpretative decision-making process.
Why is it important to know about the future? Because a CIO has one foot in the present and one in the future, while also learning from the past.
American educator and business writer Peter Drucker once said, "Trying to predict the future is like trying to drive down a country road at night with no lights while looking out the back window."
With enough foresight, a solid supporting team, and knowledge of market directions past and present, CIOs who are unafraid to reach for the brass ring while holding on with just one hand to the carousel horse can move their entire enterprise into the future as effortlessly as possible.
In cases like these, instead of becoming the means to an end, technology becomes the basis to establish a new beginning.
