September 2019

Information Blocking: Is It or Isn’t It?
By Elizabeth S. Goar
For The Record
Vol. 31 No. 8 P. 18

Industry experts discuss the complicating factors involved with defining the term.

Efforts by the Office of the National Coordinator for Health Information Technology (ONC) to end information blocking have revealed an interesting consensus among those collecting, sharing, and utilizing patient data: The problem is real, but identifying intentional blocking is complicated.

“First, there’s the matter of what it means, precisely, which varies depending on one’s point of view,” says Peter Embí, MD, MS, FACP, FACMI, FAMIA, president and CEO of the Regenstrief Institute and board chair of AMIA. “Then there’s the matter that while there are the obvious cases where information is refused or overtly blocked upon request, [which is] probably less common, there are also more subtle types of blocking, like making data sharing challenging, either technically or through bureaucracy or policies, that are probably more common.

“I don’t think there’s any doubt that it’s happening, and it could even be common depending how you define it, but it’s hard to quantify today.”

Embí, the Leonard Betley Professor of Medicine and associate dean for informatics and health services research at Indiana University School of Medicine, notes that few studies have been conducted to explore the issue. Therefore, while information blocking does exist and may even be prevalent, “to really know, and I think this is critically important, I believe we need more research to fully characterize and quantify the extent to which it is happening.”

When Is It Blocking?
Embí says that while information blocking was likely happening long before EHRs entered the picture, it has evolved with the widespread adoption of EHR systems. The ease with which data can now be shared has also led to an increase in perceived blocking situations. With information exchange now the norm, “when it doesn’t happen it can be due to ‘blocking’ in one form or another,” Embí says.

The challenge is that exactly what constitutes intentional information blocking is murky. For example, if a hospital’s policy is not to share mental health records without the patient’s express written consent for each request, is it blocking to refuse a request from the patient’s primary care physician because consent has not been obtained?

Technology and its associated costs add another layer of uncertainty. If a small physician practice can’t afford to connect to the local health information exchange (HIE), is it blocking when information from that practice isn’t shared? If a vendor is slow to connect its new customers to an HIE, is it intentionally blocking the flow of information from those groups?

“We do have gaps in data. It’s from entities not connected. It’s from technical issues. It sometimes can be a cost barrier or a business issue. Whether intentional or not, there are gaps in the data” flow, says Teresa Rivera, president and CEO of the Utah Health Information Network (UHIN).

Rivera predicts that the challenges distinguishing true information blocking vs situational or unintended instances will only worsen as providers begin to realize the value of new data types, particularly those that support value-based care such as social determinants of health (SDOH) data. While SDOH data are not protected health information, they represent valuable information that providers and payers alike are starting to demand to help improve patient and population health outcomes.

SDOH will expand requirements for data exchange. For example, information will need to be shared with the social services organizations responsible for filling nonmedical gaps such as transportation and home companions.

There are also issues around standardization that can hamper fulfillment of information requests. Clinical notes are a prime example. Rarely standardized, yet crucial to a patient’s medical history, clinical notes can be difficult to share as part of the patient record. Is it blocking if this information is not interoperable?

Conflicting state regulations governing data sharing further complicate the ability to differentiate between intentional and unintentional information blocking. Rivera points to her own state as an example. Utah is an opt-out state, while Nevada is opt-in. If UHIN requests information from a Nevada HIE, is it blocking if the latter declines because of conflicting requirements? Whose rules apply?

“States have come up with their own local governance on whether there needs to be consent to share,” Rivera says, noting that’s a situation UHIN deals with by sharing data. Through the Strategic Health Information Exchange Collaborative’s Patient Centered Data Home, the HIEs have released data “based on the policy of the HIE pushing the data. [We’ve] created a fabric to respect each other’s consent laws.”

However, she says, it remains “one of the things we are most concerned about as an HIE. We’ve worked hard with the community to pull together and figure out when data are used. … Communities have created a trusted framework [and] we don’t want to disrupt that, so we need to talk about how to respect local governance” without the risk of information-blocking accusations being leveled.

In addition to dealing with the various state laws and how to manage data sharing between and across state lines, other issues such as proprietary technologies and contractual restrictions such as gag clauses further muddy the information blocking waters, says Joe Fisne, MS, CPHIMS, CHCIO, associate CIO at Geisinger Health System.

“You can have certain issues where you’re trying to meet needs without information blocking, but the vendor may have some sort of gag preventing access,” he says. For example, if an EHR system is going through an upgrade and it’s causing delays in data sharing, the gag clause could prevent the provider from explaining why the information is delayed and, as a result, the provider could be accused of blocking.

Embí says resolving the information blocking challenge is a manifold problem that requires addressing technical, regulatory, and business solutions.

“The truth is that solutions to information blocking will have to involve what we often refer to as sociotechnical solutions,” he says. “While there are sometimes technical barriers, such as the absence of easy-to-access or -use interfaces and connectivity solutions between systems, there are also regulatory and business considerations that have to be addressed to resolve this.

“It will be critically important that we also develop and encourage processes to learn more about the real-world examples of information blocking, measure when and how it’s happening, and study it so that we can better address it,” he adds. “We just don’t know enough today, and only through better knowledge of the extent and nature of the problem will we be better able to ensure that any rules we put in place help and don’t have unintended consequences.”

Regulatory Relief?
The rules to which Embí refers are contained within proposed rulemaking from the ONC and the Centers for Medicare & Medicaid Services—rules the agencies hope will bring clarity to what constitutes nefarious information blocking and put a stop to the practice. The proposed changes to the “21st Century Cure Act: Interoperability, Information Blocking, and the ONC Health IT Certification Program” seek to regulate information blocking through civil monetary penalties and other disincentives.

Along with defining what constitutes likely information blocking, the proposed rule identifies several exceptions, which fall under the following seven categories:

• preventing harm;
• promoting the privacy of electronic health information;
• promoting the security of electronic health information;
• recovering costs that are reasonably incurred;
• responding to requests that are infeasible;
• licensing of interoperability elements on reasonable and nondiscriminatory terms; and
• maintaining and improving HIT performance.

Fisne notes that the new ruling hopes to clear up the uncertainty surrounding information blocking, in part by streamlining standardization. However, the success of some elements will depend upon providers’ willingness—and ability—to adapt practices and technologies.

One example is the proposed replacement of the Common Clinical Data Set (CCDS) definition with the United States Core Data for Interoperability standard, which establishes a minimum set of data classes, including structured data, that are required for interoperability. The Core Data for Interoperability adds Clinical Notes and Provenance as data classes, which will require updates to the Consolidated Clinical Document Architecture (C-CDA) standard and several certification criteria.

While acknowledging that it’s a good idea to remove the CCDS definition to standardize information sharing from a continuity of care document, Fisne says such a document “is not one size fits all. Some are more heavily detailed [and] physicians say they can’t use them.”

Nor can all systems handle the amount of discrete data that would be required without being upgraded, he adds.

Geisinger Health collaborated with the College of Health Information Management Executives (CHIME) and fellow members to submit a single set of comments on the proposed rule. One area of concern the group cited was the compliance timeline, which requires organizations to be compliant 24 months after the final rule is set.

Pointing to requirements such as the C-CDA, Fisne says there is a lot currently happening in the industry that may make meeting such an aggressive timeline impossible for many. “We want to do it all, but we want to do it the right way, in a very structured manner,” he says. “We need time to execute.”

Adds CHIME’s vice president of federal affairs, Mari Savickis: “We also asked for an appeals process and hope there will be some ability to negotiate that. The [Office of Inspector General] hasn’t come forward yet with a plan for providers, but having due process will be very important.”

According to Mariann Yeager, CEO of The Sequoia Project, the goals of the proposed rulemaking align closely with what her organization had identified as being in the best interest of information sharing and will continue “the positive trajectory we’re on” in terms of interoperability and data exchange. Among the positives within the proposed rulemaking are establishing standards around application program interfaces and Fast Healthcare Interoperability Resources, as well as the categories of exceptions, she says.

“They are pretty much in keeping with practical implementation of the issues you’d want around information sharing,” Yeager says. “They did focus on patient matching, which is great, and applying lessons learned in the field.

“Our litmus test is if it is of value and usable,” she continues, adding that it’s “a very complex topic and very nuanced. When we look at practicality, there are aspects that will be challenging.”

Those include an overly broad definition of what would be subject to the scope of oversight. The same is true for the definitions on information access, exchange, and use—does it apply to all data or only data held within repositories?

The net cast by the rule, Yeager says, would be broad. “You’d have to get into extensive scenarios to figure out what exceptions apply,” she says. “It creates a complex compliance and enforcement process. Health care organizations will have to determine, document, and demonstrate that all practices are reasonable and appropriate, and go through every exception and element to prove it, which may be an unintended consequence.”

Time and Understanding
Two things that will make or break compliance with any new rules on information blocking: sufficient time before enforcement begins so necessary changes can be implemented and sufficient guidance to ensure the entire health care sector is “mindful that just because information isn’t flowing the way you want it to, that doesn’t mean it’s information blocking,” Yeager says. “[Otherwise] it will result in a flurry of complaints that may or may not be founded and rushes to judgment.”

Savickis concurs, noting that while information blocking does exist and must be stopped, it’s not always the case. It’s too convenient to “lump everything into a bucket of information blocking,” she says.

“The whole combination of rules that’s coming out will reshape our industry,” Savickis adds. “Everyone wants it to be done thoughtfully and keep the momentum around interoperability.”

— Elizabeth S. Goar is a freelance writer based in Wisconsin.