Clinical messaging delivery and read receipts are often treated as a single product capability, but the source documentation describes several distinct mechanisms with different evidentiary value. A read receipt can mean a message reached a device, was opened by a named recipient, was viewed by every intended reader, or was acknowledged by an external system. The distinction matters because delivery confirmation and read confirmation answer different operational questions. This article examines what FHIR messaging specifies about reliable delivery, what Microsoft Teams healthcare policy settings control for read receipts, and how patient-facing products from InterSystems and SigmaMD document read status. It closes with a hypothetical validation scenario that an implementation team could adapt without assuming any vendor has passed independent testing.

Delivery Acknowledgment Is Not the Same as Read Confirmation

The FHIR messaging specification separates the transport of a message from the behavior of the receiving application. It states that content is delivered from one application to another by some delivery mechanism, and then one or more responses are returned to the source application [2]. The exact mechanism of transfer can be abstracted, and the specification explicitly ignores interface engines and message transfer agents that exist between source and destination [2]. This means a FHIR message acknowledgment can confirm that a conformant receiver processed a bundle, but it does not by itself confirm that a human clinician read the content.

The specification also distinguishes reliable delivery systems from common implementations that do not provide reliable messaging [2]. When a receiver implements reliable messaging, it must check the incoming Bundle.identifier against a cache of previously received messages [2]. If the identifier has not been received, the receiver processes the message; if it has already been received, the receiver can return the previous response [2]. This deduplication behavior is a delivery guarantee, not a read receipt. A system can confirm that an order was processed once without any evidence that a person opened or understood it.

How Microsoft Teams Policies Make Read Receipts a Tenant-Level Decision

Microsoft's healthcare messaging documentation describes read receipts as a policy setting that lets a sender know when a message was read in one-to-one chats and in group chats of 20 people or fewer [3]. The setting can be user controlled, on for everyone, or off for everyone [3]. The documentation states that read receipts are important in healthcare organizations because they remove uncertainty about whether a message was read [3]. For healthcare applications, Microsoft recommends choosing either user controlled or on for everyone, and it states that the feature is most effective when set to on for everyone [3].

A documented example scenario illustrates the intended workflow. A nurse starts a group chat for a high-risk patient and sends urgent messages through one-to-one and group chat [3]. The documentation describes a situation where the nurse is aware that a message was read by a named recipient before that recipient decides how to respond [3]. This example shows the product's stated purpose: visibility into read status for team coordination. The documentation does not claim that a read receipt proves clinical action, and it does not describe read receipts for channels or larger groups beyond the stated 20-person limit [3].

A Hypothetical Validation Scenario for Read Receipt Semantics

An implementation team could validate read receipt behavior by constructing a controlled scenario with three message types: a routine informational message, an urgent clinical message, and an externally generated patient notification. For the routine message, the team would verify that a delivery acknowledgment is recorded and that a read receipt appears only after the intended recipient views the message. For the urgent message, the team would test whether escalation to an alternate clinician occurs when the primary recipient does not open the message within a defined interval, as the SAFER guide suggests escalation could involve automatic forwarding [1]. For the external notification, the team would confirm whether the sending system receives a read status update, as InterSystems documents for externally initiated messages [4].

The scenario should also test boundary conditions that the documentation leaves open. For example, a team using Microsoft Teams would need to verify whether read receipts behave as documented in group chats of 20 people or fewer and what happens in larger groups [3]. A team using SigmaMD would need to confirm that team chats and automated messages do not show read receipts, as the vendor states [5]. None of these checks would establish clinical efficacy or legal compliance. They would only confirm that the documented mechanics operate as described in a specific deployment.

Patient Portal Read Status in InterSystems HealthShare Personal Community

InterSystems HealthShare Personal Community documentation describes two workflows for communicating with patients: secure messages and direct notifications [4]. Secure messages are bi-directional communication between patients and providers within the Personal Community application, while direct notifications are automatic, workflow-based notifications intended to inform a patient about upcoming events or direct them to complete an action [4]. The documentation states that messages may be automatically generated and sent from a triggering event within an external system [4]. For these messages, the external system has the opportunity to receive updates on the read status of messages and any attachments that have been sent to patients [4].

The documentation frames this read status update as a way for the sending system to receive an accurate account of whether a patient is reading the materials sent to them [4]. This is a vendor-documented capability for externally initiated messages, not a universal claim about all message types. The same product area guide also notes that proxy-only members can send and receive messages only on behalf of their principals and do not have their own inbox in the public application [4]. That constraint means read status for a proxy-mediated message may reflect the proxy's access rather than the principal's direct viewing, an operational distinction the documentation does not resolve.

SigmaMD Documents a Different Read Receipt Model for Patient Chats

SigmaMD's clinician help documentation describes read receipts as checkmarks next to messages in patient chats, visible only to clinicians and not to patients [5]. One checkmark means the message was delivered but not everyone it is meant for has seen it, while two checkmarks mean everyone it is meant for has seen it [5]. The documentation states that a message switches to seen once all of its intended readers have viewed it [5]. For messages to the patient, intended readers include the patient and any caregivers with app access; for internal notes, intended readers are the chat's assignee and followers [5].

SigmaMD explicitly limits the scope of its read receipts. The documentation states that read receipts appear in SigmaMD Clinician on the web, in patient chats, and that team chats and automated messages do not show read receipts [5]. It also states that seen means the message was viewed, not that it was answered [5]. This is a narrower definition than a delivery acknowledgment or an application-level response. The Seen By view lists who viewed the message and when, grouped into patients and clinicians [5]. Internal notes list clinicians only, since patients never see them [5].

What the SAFER Guide Adds About Escalation and Differentiation

The SAFER Clinician Communication guide addresses secure messaging as a recommended practice for EHR-based systems. It states that critical and time-sensitive messages and results to clinicians should be clearly differentiated from routine or information-only communication that does not require immediate attention [1]. The guide also describes escalation as potentially involving automatically forwarding the message to an alternate or supervising clinician if the intended recipient is unavailable [1]. These are recommended practices, not product specifications, and the guide states that its content is not intended to be an exhaustive or definitive source and does not constitute legal advice [1].

The SAFER guide's framing matters for read receipt design because it connects acknowledgment to urgency. A read receipt that confirms viewing may be sufficient for routine information, while a time-sensitive message may require escalation logic that operates independently of whether a recipient opened the message. The guide does not prescribe a specific read receipt standard or vendor. It provides a worksheet with a rationale and examples for implementation, and it notes that users should consult their own legal counsel regarding compliance with Medicare or Medicaid program requirements and other laws [1].