Answers are grounded in this paper and cite their sources.
Photo ID 136523299 © Designer491 | Dreamstime.com Abstract Patient portals were introduced to increase communication between patients and their care teams. However, rising message counts have contributed to physician burnout. Some health systems have started billing patients for portal messages that require substantial clinical judgment and time. This policy is often ethically justified through beneficence by decreasing provider burnout. However, it also introduces uncertainty into the patient-provider relationship. This paper argues that because patients cannot know whether a message will be billed, they cannot give informed consent, undermining their autonomy. It further reexamines the beneficence argument, finding that reductions in message volume have been modest and often offset by increased phone calls and added physician workload. The paper concludes with recommendations on informed consent and potential future changes. Introduction Patient portals have emerged as part of a broader shift toward giving patients direct digital access to their own health information and care teams. These portals function as a channel to allow patients to view test results, request refills, and reach their physicians between visits. Messaging a physician through the portal was understood as an extension of the ordinary patient-physician relationship and not a significant burden on physician workloads. That understanding began to break down when patient messaging surged during the COVID-19 pandemic. This increase in portal messaging has become a sustained, ongoing source of physician work rather than a temporary disruption, with an estimated 153 percent increase in patient-authored portal messages continuing between 2020 and 2025.[1] This rise was not accompanied by any decrease in telehealth or office visits, which suggests messaging has become an expansion of between-visit care rather than a substitute. Physicians now receive substantially more portal messages than they did before the pandemic, a volume consistently linked to physician exhaustion. Physicians in the top quartile of messages received more than six times greater odds of burnout.[2] As open access to clinicians has evolved into a second, largely uncompensated shift, health systems have sought ways to address the burden. One response has been billing for portal messages. In 2020, the Centers for Medicare & Medicaid Services allowed billing patient portal messages requiring medical decision making and at least five minutes of clinician time as e-visits.[3] Charging for clinically substantive messages seems like a straightforward correction as it compensates clinicians for work that has long gone unpaid while discouraging low value messages from burdening inboxes. Over 470 healthcare organizations have begun billing portal messages meeting the appropriate criteria, with approximately 100,000 billed messages per month.[4] However, this approach assumes that billing simply assigns a price to an established service. In practice, billing for portal messaging means health systems are creating a new billable communication service that patients have not historically understood as a paid service. Autonomy in the Setting of Billing Uncertainty Prior ethics work on billing for patient portal messages has focused on the strain on the provider-patient relationship, financial access to care, and disparities linked to technological literacy.[5] This paper builds on that foundation, but places focus on patient autonomy, an area that has received comparatively less attention. Specifically, it examines how the uncertainty of not knowing whether a physician will bill for the message shapes patients’ ability to make a truly informed decision. Respect for autonomy requires that patients possess sufficient information to make decisions about their care in accordance with their own values and preferences. Financial costs are related to such decisions. For patients managing limited financial resources, chronic illness, or high-deductible insurance plans, the cost of care may be a material consideration in determining whether to seek care at all.[6] This argument doesn’t support requiring complete certainty of the specific dollar cost of care. Rather, it addresses the insufficient certainty in the current piecemeal system. Respect for autonomy does not require that patients possess perfect information, nor does it eliminate the inherent ambiguity that characterizes medical decision making.[7] Rather, the relevant question is whether patients have sufficient information about pricing to make decisions regarding messaging their clinicians that meaningfully reflect their own preferences and interests. In many cases, clinicians decide only after a message thread is complete whether it qualifies for e-visit billing and patients are told beforehand simply that a message may result in a charge. This means that they message a clinician without knowing whether their action will result in a bill. The patient receives only a generic notice that a charge is possible at the beginning of the encounter. The treating physician could respond, but a delegated nurse or other advanced practice provider could as well.[8] The response may be brief or include a substantive clinical assessment and it may require a prolonged exchange. Consent in this setting is therefore price-blind, and the nature of the content is unpredictable and largely based on whether the response is long or generates additional back-and-forth conversation. Requiring patients to guess beforehand whether their question will take a provider five minutes is a judgment inappropriate for the medical literacy level of the patient. Offloading this responsibility to patients only furthers inequity by making less informed and less well-off patients question reaching out for care. A qualitative study of patients and physicians across several institutions confirms that physicians as well as patients report uncertainty about whether messages cross the billing threshold.[9] If the clinicians responsible for applying billing criteria cannot reliably predict its application in advance, it is difficult to argue that patients who cannot see how the criteria apply to their own case are positioned to consent to it prospectively. The role of unpredictable financial consequences forcing patients to make medical decisions parallels the history of retrospective denials of emergency department care for visits deemed to not be a true medical emergency. In 2018, Anthem, the second largest insurance carrier in the United States, expanded a policy denying reimbursement for visits deemed not a true medical emergency.[10] In contrast to the process of patient portal messaging, however, there is the concept in emergency medicine of the prudent layperson. If a layperson was having symptoms which he could reasonably consider an emergency, then the care must be covered for the patient even if further clinical evaluation does not find a time-sensitive critical or emergency diagnosis.[11] Applying this idea to patient portal messages, some would argue that the primary issue concerns coverage rather than billing. Like insurance companies should not punish the decision to seek emergency department-level care, payors should cover portal messages over the threshold whether patients knew they would surpass the threshold or not. At minimum, in partial-capitation-based models, patient portal messaging should be seen as an essential component. Some degree of certainty would help patients who may have concerns about deductibles and insurance coverage. Reexamining Reducing Workload, Burnout, and Efficiency Decreasing provider workload often serves as an ethical argument for billing patient portal messages. Proponents argue that billing reduces the message volume which contributes to physician burnout. Despite this widely held belief, current evidence warrants further examination. Patient portal billing has coincided with a decrease in message counts. However, after implementing billing for portal messages, the Mayo Clinic saw the six-month total message count decrease by only 8.8 percent.[12] After implementing billing patient portal messages, UCSF’s weekly patient message count decreased by approximately 1700 a week, only around a 2.9 percent decrease.[13] Though billing reduces messages, there is no evidence that this single-digit percentage decrease is enough to meaningfully ease burnout. 91.5 percent of physicians found that billing for portal messages led to either an increase or no change in the time they spent responding to messages.[14] In addition, although messages have been down, the decrease has been offset by an increase in telephone calls. In one qualitative study, patients who found portal message billing confusing or frustrating said they would return to calling the office rather than sending a message.[15] This means that the reduction in portal messages is just a transfer of the same work onto an unbillable channel that takes longer to triage and pulls front-desk staff into tasks a written message could have handled with less overall work. Billing itself takes time. Physicians are no longer able to simply answer a message. For each exchange, they must decide whether it meets the billing threshold. This policy, meant to reduce off-hours work, adds new work to the original task. Without a clear reduction in clinician time spent responding to messages or a decrease in provider burnout, burnout reduction does not justify billing. While billing is typically framed as a mechanism to reduce message volume, it can equally be understood as a deterrent to contacting physicians due to fear of cost. Naturally, the patients with the most chronic conditions are significantly more likely to send messages through the patient portal.[16] The uncertainty around billing may lead these patients to delay reaching out to their physician. In the case of patients with congesti
No takes yet. Share an insight, caveat, or question.
Arnav Goel (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: