We’ve lost our patients’ trust. Here’s how to earn it back.

Remember when healthcare workers were regarded as heroes? During those early days of the Covid-19 public health emergency, when most schools and workplaces were closed, people lined up on their driveways and apartment building balconies, literally cheering clinicians as they went to work.

That seems like a lifetime ago. Today, healthcare suffers from a crisis of trust. Some of this is institutional and political, tied to the fallout from the pandemic. Americans increasingly perceive that healthcare quality has declined. We can give our advice as experts, but patients increasingly are tuning us out. Instead, they’re turning to tools and sources that are not grounded in evidence, sometimes with disastrous results.

What went wrong? How did we fall so far so quickly? And how can we earn back the trust of the patients and communities we serve?

It is tempting to blame the pandemic, politics, or social media. These factors matter, but they do not fully explain what we are seeing. So, a little introspection is appropriate. Clinical work has become more uncertain and more relational, but the underlying design of the care encounter has not kept pace. For decades, healthcare has been engineered around efficiency, standardization, and throughput; these priorities remain foundational but are no longer sufficient. We are asking clinicians to deliver a kind of care the system is not built to support, even though we know outcomes are better when relationships between clinicians and patients are strong.

Trust also has implications beyond clinical outcomes. While the relationship among trust, net promoter scores, and loyalty is still being explored, trust may be an important factor. If you want to keep your patients coming back and bring new patients in through word of mouth, it’s a prerequisite that they trust you.

If trust is to be restored, we have to make it visible. We cannot treat trust as an aspiration; we must regard it as a performance indicator that can be measured and managed.

Trust Performance Measures

We must measure what matters and can only improve what we measure. This applies to trust. Performance metrics in trust can be classified into three categories:

1. Behavior. Vaccine acceptance is an example of a measure that is predicated on trust (as well as an intentional approach from a population health standpoint). Patterns of adherence and continuity in patient-provider relationships provide insight into whether patients believe in the guidance they are receiving.
2. Understanding. Measures of patient activation, comprehension, and engagement offer a window into whether patients feel informed and confident enough to act.
3. Decision quality. Goal-concordant care, particularly in complex or high-stakes situations, indicates whether treatment aligns with what patients and families say matters to them. When that alignment is present, trust is functioning.

These performance categories can be considered proxy measures for trust. They are individually imperfect, but taken together, they provide a practical way to assess how trust operates within a clinical environment.

Trust and Efficiency Are Compatible

Let’s pause to acknowledge a foundational truth: throughput is critically important. Higher throughput improves access and margin, and it often improves patient satisfaction as well.

Unfortunately, there is rarely an equivalent emphasis on relational work. (Hospice and palliative care teams, we’re looking at you.) In most organizations, efficiency is engineered precisely, while trust is treated as an intangible. That distinction shows up in how care is structured. Standard visit lengths and productivity targets assume a level of predictability that no longer reflects clinical reality. When encounters require extended counseling, the system does not flex to accommodate that complexity. Instead, clinicians absorb the burden, often by working longer and delaying or compressing conversations.

The introduction of new immunization CPT codes reflect a partial recognition of this. These codes create a mechanism for stand-alone vaccine counseling, including when a vaccine is not administered. This constitutes a formal acknowledgement that the cognitive and relational demands of these encounters have grown.

This is a partial adjustment and won’t fix everything at once. But it’s a start.

If it is true, as quality improvement pioneer W. Edwards Deming has written, that every system gets the results it was designed to achieve, then we must face an uncomfortable reality: our operating models are not designed to build trust. But if it’s a design problem, then it can be solved. The levers are within reach. Measurement can be expanded, workflows adjusted, and shared decision-making strengthened. With deliberate design, trust can become a reliable outcome of care.