How to Collect and Use Patient Feedback to Improve Care

Elena Marquez October 15, 2026 20 min read

A few years into my career in quality improvement, a nurse manager slid a binder across the table to me. It was thick, the kind of thick that makes a clasp strain. Inside were eleven months of patient feedback from her outpatient unit, handwritten comment cards carefully filed by date. So I asked her what the team had done with them. She looked at the binder, then at me, and said, “We read them.”

That moment has stayed with me because it captures the problem better than any statistic. In fact, most healthcare organizations are not short on patient feedback. Instead, they are short on a system that turns what patients say into something that changes on Tuesday morning. Reading comments is not improvement. Similarly, filing them is not improvement. Even reporting them to a committee falls short unless someone walks out of that meeting with a task and a date.

I have spent a long stretch of my working life helping hospitals, clinics, and practice groups close that gap. What follows, therefore, is not theory. It is the approach I use, refined by a lot of trial, a fair amount of error, and many honest conversations with frontline staff who were tired of being handed scores without any help understanding them.

Why Patient Feedback Deserves a Seat at the Quality Table

There is still a quiet assumption in some corners of healthcare that patient opinion is a “soft” measure. In this view, clinical outcomes are real while patient feelings are nice to know. I understand where that comes from; however, I think it is wrong, and the research backs me up.

Patients see things we cannot. For example, they notice that the call light took twenty minutes on the night shift but only four minutes during the day. Many also pick up on the fact that three different people gave them three different explanations of the same medication. Others spot discharge instructions that assumed they had someone at home to help, when in reality they did not. None of these show up in a lab value. Even so, every one of them touches safety, adherence, and whether that patient comes back through the emergency department a week later. Together, they make up what we call the patient experience.

A useful working definition comes from a qualitative study in BMJ Open, which describes patient feedback as recording the patient’s perspective on the care they received so that processes and experience can be improved. I like that definition because it puts the purpose right inside it. In other words, feedback exists to drive learning. If it is not doing that, then it is just paperwork.

The Regulatory and Reputational Stakes

Beyond quality, there is also a regulatory and reputational dimension. In the United States, for instance, the HCAHPS hospital survey has been publicly reported for years, and scores are tied to payment programs. Likewise, in Australia, the RACGP practice standards require general practices not only to collect feedback but also to show patients how it was used. Wherever you work, the expectation is moving in the same direction. Listening is no longer optional, and neither is acting.

Start With a Question Before You Collect Patient Feedback

The most common mistake I see is an organization buying a survey platform before deciding what it actually wants to learn. As a result, they end up with forty questions, a 6 percent response rate, and a dashboard full of numbers nobody trusts.

Before you collect anything, sit down with your team and answer three questions:

What decisions will this feedback inform?
Who will receive the results, and what can they realistically change?
How quickly do we need to know?

If the answer to the first question is “we’re not sure,” stop there. After all, feedback without a decision attached to it becomes noise. When I worked with a mid sized orthopedic practice, for example, we narrowed their focus to a single concern: patients felt rushed during post surgical follow up visits. That one question shaped everything. First, we chose a short survey sent within 24 hours of the visit. Next, we added one open text question about time with the surgeon. Finally, we agreed in advance that the practice manager would review responses every Friday. Within three months, they had restructured their scheduling template. Clearly, a focused question produced a focused change, and it is the same principle behind most efforts to improve patient satisfaction in a practice.

Choosing the Right Methods for Collecting Patient Feedback

No single method captures the whole picture. Each one sees part of the patient’s experience and, at the same time, misses another part. For that reason, the strongest programs I have built use a blend, chosen deliberately.

Validated Surveys

Standardized instruments like the CAHPS family of surveys, developed through the Agency for Healthcare Research and Quality, give you reliable, comparable data. Rigor is their strength, whereas timing is their weakness. By the time quarterly results arrive, the patient has long gone home and the staff on that unit may have changed. So I treat validated surveys as the compass, not the steering wheel. In short, they tell you which direction you are heading over months, not what to fix tomorrow.

Real Time and Point of Care Patient Feedback

This is where the field has moved fastest. Short surveys delivered by text, tablet, a patient portal message, or a quick bedside conversation let you hear about a problem while you can still do something about it for that patient, including after virtual appointments.

A good example comes from Penn Medicine. Their team found that long mailed surveys were producing low response rates and slow turnaround, so they built a system called PMX Feedback that sends brief surveys after scheduling and after visits. In the first year and a half, staff took action more than 110,000 times in response to that feedback, whether through service recovery, patient follow up, or coaching. That number tells you something important: when feedback arrives quickly and reaches the right person, people act on it.

Similarly, Cedars-Sinai took this approach for inpatients, sending short, tailored text surveys within hours of admission rather than waiting for a generic survey after discharge, when details have already started to fade.

Still, a word of caution from experience is in order. Real time feedback creates an expectation. If a patient tells you at 2 p.m. that their room is freezing and nobody comes by 6 p.m., then you have made the experience worse by asking. Therefore, do not launch point of care feedback until you know exactly who responds, and how fast.

Leadership and Nurse Rounding

Structured rounding remains one of the most underrated feedback tools we have. When a charge nurse or unit leader stops in to ask, “Is there anything we could be doing better for you today?” they collect information no survey will ever capture. After all, patients say things to a person standing in front of them that they would never type into a form.

The trick, however, is recording what you hear. Rounding that lives only in someone’s memory cannot be trended. For this reason, give rounders a simple log, even a paper one, and review themes weekly.

Complaints and Compliments

Formal complaints are feedback with the volume turned up. Although they are uncomfortable, that is exactly why they are valuable. I encourage teams to code every complaint by theme (communication, access, environment, clinical care, billing) and then look at them alongside survey results. A theme that appears in both is rarely a coincidence.

Compliments, meanwhile, deserve the same treatment. When a patient names a specific nurse who explained a procedure clearly, that is a behavior you can teach to others.

Patient and Family Advisory Councils

If surveys are how you measure the experience, then advisory councils are how you design it. Bringing former patients and family members into the room when you redesign a discharge process or a waiting area catches problems before they happen. Once, I watched a council member look at a new patient education leaflet and point out, kindly, that it was written at a reading level well above most of our patients. That single comment saved months of confusion, and it is a reminder of why health literacy matters in every piece of material we hand out.

For further guidance, the Beryl Institute has published practical material on both real time feedback and making these councils work, including virtual formats that widen participation.

Online Reviews and Open Text Comments

Public reviews are messy and unrepresentative. Nevertheless, they are not meaningless. Patients who write them often describe access problems (phone wait times, parking, confusing directions) that internal surveys underweight. Accordingly, I recommend a monthly scan for recurring themes rather than reacting to individual posts. When comment volume grows, AI in healthcare tools such as text analytics can help sort themes faster, though a person should still read the comments behind them.

Designing Patient Feedback Questions That Produce Usable Answers

Good questions are specific, behavioral, and tied to something staff can influence. To see the difference, compare these two:

“How satisfied were you with your care?”
“Before you received a new medicine, did a staff member explain what it was for?”

The first produces a feeling, while the second produces a fact you can act on. If the answer to the second question is “no” for a quarter of your patients, you know exactly where to look.

Principles for Writing Better Survey Questions

Here are a few principles I hold to:

Keep it short. For point of care surveys, three to five questions is plenty, because every extra question lowers your response rate.
Always include one open text question. Something like, “What is one thing we could have done better?” works well, since the narrative comments are often where the real insight hides.
Ask about experiences, not ratings, whenever possible. In practice, “did” and “how often” questions beat “how satisfied” questions.
Test the wording with real patients before launch. What seems obvious to clinicians is often confusing to everyone else.
Think about access. Offer surveys in the languages your patients speak, with support from medical interpreter services where needed, in formats that work with screen readers, and through channels that do not assume everyone has a smartphone.

That last point matters more than many organizations admit. If your feedback only comes from patients who are comfortable with technology and fluent in the dominant language, then you are hearing from the people who probably already had the smoothest experience. The broader barriers to accessible healthcare show up in your feedback data too.

Making Sense of Patient Feedback Data

Once feedback starts flowing, the work changes. From that point on, the challenge is interpretation.

A systematic review in the Journal of Patient Centered Research and Reviews developed a framework that I find genuinely useful. Specifically, it groups what organizations need to do after collecting data into three themes: make sense of the data, communicate and explain it, and plan for improvement. Its central conclusion is one I repeat often: running a survey, by itself, will not improve performance.

Four Rules I Follow When Reading Patient Feedback

Look for patterns, not outliers. One furious comment about a rude receptionist deserves a response, but it is not a trend. On the other hand, ten comments over two months mentioning confusion at check in point to a process problem.

Combine numbers with stories. Scores tell you where; comments tell you why. For instance, when one unit’s score on “staff explained things in a way you could understand” dropped, the comments showed that patients were confused specifically about pain medication schedules after surgery. As a result, a vague communication problem became a focused one.

Triangulate with other data. The AHRQ research meeting on CAHPS improvement emphasized complementing survey data with other sources and using dashboards to track change over time. In my own work, I routinely pair patient feedback with incident reports, staffing data, and call center metrics. For patients enrolled in remote patient monitoring, engagement with wearable health devices can add another useful signal. When patients complain about wait times on Mondays and your staffing schedule shows two fewer nurses on Mondays, the conversation gets much shorter.

Respect the limits of small samples. A clinic that receives twenty surveys a month should not panic over a five point drop. Instead, use run charts and look at trends over several months before drawing conclusions.

Getting Patient Feedback Back to the People Who Deliver Care

This is the step that separates programs that work from programs that only produce reports.

I learned this lesson most clearly from a study I still share with every team I train. Researchers at the University of Greenwich ran a pilot randomized trial in which nurses received their own ward specific patient survey results. On some wards, staff got written reports only. Elsewhere, teams had facilitated meetings to discuss the findings together. Remarkably, the written reports alone had no impact on scores, whereas the facilitated discussions had a highly significant one.

To their credit, the researchers were honest about how hard those meetings were. Some nurses doubted negative comments. Others felt hurt or defensive, and a few pushed back against patients. I have seen every one of those reactions in my own work. Over time, I have come to believe they are a necessary part of the process, because people need room to react before they can reflect.

How I Run a Patient Feedback Discussion With Staff

When I facilitate these sessions, I follow a simple structure:

Start with what is going well, using specific patient quotes that name behaviors.
Present one or two themes needing attention, never ten.
Ask the team what they think is causing the issue before offering my own view.
Let the team choose the first change to test.

That last step matters most. Changes designed by the people who have to carry them out last far longer than changes handed down from a committee.

For more real examples, the AHRQ meeting summary is worth reading. It describes, for example, how the Veterans Health Administration rolled out a patient experience toolkit across clinic sites after studying what its survey data revealed. It also explains how dialysis facilities combined grievances and internal surveys to set goals by domain with their clinicians.

Turning Patient Feedback Into Tested Changes

Once a team has chosen a problem, I use the Plan, Do, Study, Act cycle. Admittedly, it is not glamorous. However, it works because it forces small, measurable tests instead of sweeping, untested initiatives.

A Real Pattern From a Medical Unit

I have seen the following approach succeed more than once.

The problem. Patients on a medical unit reported that they did not understand their medications at discharge. In addition, survey scores on medication communication were well below the organization’s average, and comments mentioned feeling “rushed” through instructions.

Plan. To start, the team decided to test the teach back method with one nurse, on one shift, for one week. With this method, the nurse would explain each new medication and then ask the patient to describe, in their own words, what it was for and when to take it.

Do. Over that week, the nurse tried it with every discharge. Along the way, she kept a short tally of how often patients could explain their medications correctly on the first attempt.

Study. Surprisingly, the first attempt rate was lower than expected. Yet the nurse noticed that patients understood much better when she used the actual pill bottles rather than a printed list.

Act. Based on that insight, the team adapted the approach, spread it to two more nurses, and then extended it to the full unit over the following months. Consequently, medication communication scores climbed steadily, and the comments changed tone completely.

What matters here is not the specific intervention. Rather, it is the sequence: feedback identified the problem, the team owned the solution, and small tests built confidence before anything spread.

Closing the Patient Feedback Loop

Closing the loop means telling patients what you heard and what you changed. Unfortunately, it is the most skipped step in patient feedback, even though it is the one that builds trust and lasting patient engagement. There are two loops to close.

The Individual Loop

To begin with, there is the individual loop. When a patient raises a concern, someone should acknowledge it, personally where possible, and follow up. Done well, service recovery can turn a frustrated patient into a loyal one.

The Collective Loop

Then there is the collective loop. Here, you share what you have learned and what changed as a result. Many organizations use a “You Said, We Did” format posted in waiting areas, on websites, or in newsletters. For example, teams in Northern Ireland’s health and social care trusts have described using exactly this approach in their live patient feedback programs, alongside dashboards that ward managers can check in real time.

Consider a simple notice: “You told us the waiting room was too cold in the mornings. We have adjusted the heating schedule.” It sounds small. To patients, however, it says that someone listened.

Sharing Patient Feedback Results With Staff

Closing the loop with staff matters just as much. Whenever feedback leads to a positive change, tell the team. Likewise, when a patient praises a named staff member, pass it on the same day. Recognition is fuel, and feedback programs that only ever deliver criticism will eventually burn people out.

Common Patient Feedback Pitfalls I Still See

Even experienced organizations fall into these traps.

Chasing the score instead of the experience. When survey scores are tied to bonuses, teams sometimes start coaching patients on how to answer rather than improving care. As a result, the data becomes corrupted and the point is lost.

Collecting more than you can act on. Twelve feedback channels with no one responsible for any of them is worse than two channels with clear owners.

Ignoring the silent patients. Low response rates from certain groups are themselves a signal. So find out who is not answering, and why. Older adults are a common gap, especially those who rely on assistive technology to read or respond.

Treating feedback as a patient experience department problem. Feedback is a quality and safety issue. Therefore, it belongs in the same meetings as falls, infections, and readmissions.

Moving too fast. Spreading an untested change across a whole organization because one comment demanded it rarely ends well. Instead, test small first.

Building a Culture Where Patient Feedback Matters

Tools and processes only get you so far. Ultimately, patient feedback improves care when leaders treat it as important and staff feel safe engaging with it.

In my experience, that culture shows up in small, visible behaviors. Picture executives who start meetings with a patient story, or unit huddles that include one piece of patient feedback every morning. Add managers who respond to criticism with curiosity rather than defensiveness, and frontline staff who are given time, not just instructions, to work on improvements.

Notably, the AHRQ CAHPS research meeting series repeatedly returned to this theme: engaging providers and stakeholders early and often, and standardizing how improvement happens. In other words, culture is not a soft extra. It is the infrastructure that every other step depends on.

I often think back to that nurse manager with her binder. A year after we first met, her unit had a weekly fifteen minute feedback huddle, a short real time survey, and a “You Said, We Did” board by the entrance. By then, the binder was gone. In its place was a team that genuinely wanted to know what patients thought, because they had seen what happened when they listened.

That, in the end, is the whole goal. Not more data, but better care, shaped by the people receiving it.

Frequently Asked Questions About Patient Feedback

What is patient feedback in healthcare?

Patient feedback is any information patients or families share about their experience of care, collected through surveys, conversations, complaints, compliments, reviews, or advisory councils. Its purpose is to help organizations learn and improve. For a clear definition and overview, see the BMJ Open study on hospital use of patient feedback.

What is the difference between patient satisfaction and patient experience?

Satisfaction measures whether expectations were met, which varies from person to person. By contrast, experience measures whether specific things happened, such as whether a nurse explained a medication. For this reason, experience questions are generally more actionable. AHRQ explains this distinction through its CAHPS program resources.

How often should healthcare organizations collect patient feedback?

Ideally, continuously, but through different methods at different speeds. Real time surveys and rounding provide daily insight, while validated surveys show longer term trends quarterly or annually. For one practical model, see how Penn Medicine built a real time feedback system.

How can you increase patient feedback survey response rates?

Keep surveys short, send them soon after the visit, use the channels patients prefer, and show patients that their input leads to change. In addition, closing the loop publicly encourages future participation, as highlighted in the RACGP patient feedback standard.

How should negative patient feedback be handled?

Acknowledge it promptly, investigate without blame, respond to the patient where possible, and look for patterns across similar comments. Moreover, sharing results with staff through facilitated discussion works far better than written reports alone, according to research from the University of Greenwich.

What does closing the patient feedback loop mean?

Closing the loop means following up with patients and staff about what was learned from feedback and what changed as a result. A “You Said, We Did” approach is a common example. The Beryl Institute has published guidance on this practice.

Can small practices use patient feedback effectively?

Yes. In fact, small practices often move faster because fewer people need to agree on a change. A short survey, one open text question, and a regular team review are enough to start, as the AHRQ data analysis webcast illustrates through practical examples.

References

Tags:Patient Experience