How to simplify and improve physician referrals with data

There’s nothing wrong with traditional referral methods but decision support technology can help simplify the referral process for both physicians and patients.

It seems so much in healthcare is now “data-driven.” From diagnosis to treatment to tracking disease trajectory, healthcare’s vast data accumulation (growing faster than any other industry) combined with sophisticated analytics platforms are supporting nearly every clinical decision.

The exception to this informatics adoption trend, though, is referrals. Physicians still often arrange referrals for patients—such as a primary care physician to a specialist—the same way they did over a century ago. Referrals are based on the physician’s personal experience and positive relationship with the specialist. More recently, the referral is driven by health plan network restrictions and the patient’s reluctance to spend more for out-of-network care. Although the latter referral method is somewhat driven by data from the patient’s health plan provider directory, is it the best decision for the patient?

In some cases, the answer is “no.” The referral physician may be the right specialty and qualify as in-network, but there could be another specialist with more relevant experience, better outcomes and lower costs and is even closer to the patient’s home.

Certainly, there is nothing wrong with either traditional referral method when it results in positive clinical outcomes. Decision support technology, however, can help simplify the referral process for both physicians and patients. An automated deeper dive into available data on the physician can find the ideal referral physician sooner, which can save time for the clinician and result in a more optimal patient outcome and experience.

Referrals Still Old-Fashioned and Subjective

A fascinating study of the referral decision processes published this year reveals that despite the amount of quality, cost and outcomes data available to drive decisions, little has changed. Consider these quotes from physicians who were interviewed by researchers:

  • “90% or 80% (of my incoming referrals) [are] from a cardiologist who’s seen me in a meeting, or heard of me, or something like that.”
  • “I think I’ve built it [my reputation] based on patients I’ve operated on, who’ve gone back to their cardiologists or PCPs and have said nice things about me.”
  • “Any cardiologist within the system or cardiologists out in the community, they will refer to just [name of surgeons].”
  • “…Not only in terms of their knowledge base, but we’re all humans, and some humans interact better with other people, and [I] think about who would be the best interaction for the patient.”

The final quote raises an important point that not all referral decisions should be based on data alone. However, that comment from a heart surgeon demonstrates the inherent subjectivity of referral decisions. The greater availability of data science tools means this more qualitative assessment of the patient’s personality can supplement a deeper, more objective review of the referral physician’s experience and their clinical outcomes. Such due diligence does not replace the physician’s judgment, but rather adds a layer of certainty to an important clinical decision.

Value-Based Care Adds Referral Complexity

Sub-optimal referral decisions could affect patient outcomes and experience, which are the most important considerations. However, with the growth of value-based care payment models, where reimbursement is based on costs incurred, referrals are an important consideration for financial performance, as well.

Nearly two-thirds of healthcare organizations plan to enter or expand their value-based payment program participation, such as through a bundled payment arrangement or accountable care organization (ACO) structure. ACOs participating in the Medicare Shared Savings Plan are responsible for all costs attributed to their enrolled patients. This means a specialist with historically higher costs and poorer outcomes could negatively affect the ACO’s end-of-year expenditures.

Investigating the referral physician’s performance, which should include an analysis of a broad set of cost, quality, outcome, and other data can reduce unnecessary costs, but can also help identify opportunities where the ACO can support that physician in improving performance. By presenting the physician with objective data on cost and outcomes, the ACO or health system can demonstrate how that doctor could increase their referral volume and improve quality metrics, but also drive stronger financial performance for the overall organization.

As stated earlier, a health plan or ACO network is an important referral consideration for the patient to avoid higher out-of-pocket spending, but also for the healthcare organization’s revenue. The average out-of-network referral equates to $5,000 in lost downstream revenue. If a primary care physician refers only ten patients per month outside of the preferred network, the health system risks losing $600,000 per year—and that is just one referring physician. Health plan or ACO network affiliation, however, should only be one data point of many to consider when conducting a deep dive into potential referral partners.

What is a Data-Driven Referral Decision?

Claims data will offer insight on costs, but health systems and ACOs – if the referral physician is unaffiliated – may need to partner with a third-party data provider to fill gaps in claims data for care delivered outside the network. Such a data partner can also deliver years of non-publicly available quality data for a referral network, based on Healthcare Effectiveness Data and Information Set (HEDIS) and other measures, which are crucial for gauging the historic performance of the referral.

Non-clinical data points should figure into the referral decision as well. Social determinants of health (SDoH) factors of the patient, such as their primary language or proximity to the patient’s home location, are crucial data elements that improve the patient’s experience and could influence the likelihood of appointment completion.

All of this data can be automatically compiled and analyzed using population health management technology that can deliver a numerical score for each physician in the referral network based on patient factors and the physician’s performance across multiple quality and cost metrics. This streamlined process for selecting a referral partner can save time, but also remove uncertainty for the referring physician and the patient if neither has direct experience with that physician. Referring physicians who have historically limited their referrals to a small number of physicians can also expand their referral network using such population health management tools, which may help load balance schedules and reduce appointment wait times for the patient.

Optimizing Referral Networks

Applying data analytics to the referral decision process will not eliminate the collegial relationships physicians have with each other. Rather, conducting a deep dive into the referral physician’s performance can strengthen such relationships by putting some evidence behind their subjective or anecdotal assessment of their colleague’s reputation. As the positive outcomes continue, the referring physician will have even greater confidence in sending their patients to the referral practice or hospital for the foreseeable future.

Matt Cheatham is Lightbeam's Operations Director, ReferralPoint.