Improving patient satisfaction: Use data to reduce readmissions and lower costs

In the course of my customer loyalty work with Fortune 500 companies across a broad variety of industries over the last three decades, I have identified a series of behaviors that are exhibited by loyal (i.e., highly satisfied customers):

• Greater repurchase intent
• Greater recommendation intent

These behaviors are simply a manifestation of the customer’s trust in your company. Loyal customers recommend you to friends and repurchase from you because they trust that you will consistently deliver on your value proposition.

In addition, there are a series of other behaviors exhibited by loyal customers. We have seen highly satisfied customers modify their behavior to optimize your economics. For example, in branch banking loyal customers will alter their shopping schedules in order to visit the bank at off-peak hours because they have learned that the lines are shorter. This reduces the need for the branch to schedule additional tellers, which helps lower the bank’s cost structure.

However, nowhere is behavioral modification more important than in healthcare.

Consider this common scenario: “Mr. Larson, you have just had a heart attack. When we discharge you I strongly recommend that you stop smoking, modify your diet and begin to exercise.” That recommendation constitutes some substantial behavioral modification on my part. Not following this advice increases the likelihood of readmission for the same health problem, which raises hospital operating costs and reimbursement penalties.

Since highly satisfied patients have a trust-based relationship with you if our theory is correct, we should expect to see that hospitals that provide the highest levels of patient satisfaction have:

• Significantly better rates of patient compliance (e.g., lower rates of readmission within 30 days)
• Lower cost to serve

To test these hypotheses, we went to the Medicare Hospital Compare Website, which is maintained by the Centers for Medicare & Medicaid Services (CMS), and created a large database of patient satisfaction and readmissions rates for each of the 2,709 reporting hospitals, including:

1. Patient satisfaction scores (Percent 9 or 10) on overall patient satisfaction
2. Readmission rates (within 30 days) for three separate disease states

• Heart Attack (445,268 cases)
• Heart Failure (1,054,186 cases)
• Pneumonia (1,269,912 cases)

We subdivided the 2,709 hospitals into four groups by level of patient satisfaction (Percent 9 or 10 on the HCAHPS Satisfaction scale):

1. 80-plus percent: 276 hospitals
2. 70 percent to 79 percent: 1,329 hospitals
3. 60 percent to 69 percent: 918 hospitals
4. Less than 60 percent: 186 hospitals

For each of these four groups, we then computed readmission rates for each of the three disease states listed above. Readmission rates within 30 days are generally considered the most accurate measure of patient compliance available.

For purposes of simplicity, we will focus on the first and fourth groups. In what follows, we will refer to the 80-plus percent group as the high-satisfaction hospitals and the less than 60 percent group as the low-satisfaction hospitals.

For each of the above disease states we found that readmission rates were substantially lower at the high-satisfaction hospitals than at the low-satisfaction hospitals. For example, a heart attack patient admitted to a low-satisfaction hospital was 23 percent more likely to be readmitted within 30 days than those treated at a high-satisfaction hospital. A heart failure patient was 17 percent more likely to be readmitted and a pneumonia patient was 20 percent more likely to be readmitted. All these differences are statistically significant at the 99-plus percent level of confidence.

The Medicare Hospital Compare Website has another useful piece of data called Medicare Spending Per Beneficiary. For every Medicare patient a hospital serves, Medicare computes the total payments to the hospital starting three days prior to admission and going through to 30 days after discharge. These payments are then adjusted for factors such as disease severity (it costs more to treat a heart attack than it does the flu) and geographic cost of service differentials.

Again, when we compare the high-satisfaction hospitals to the low-satisfaction ones we see that the former have costs per beneficiary that are near 4 percent less than their low satisfaction counterparts. This difference is also statistically significant at the 99-plus percent level of confidence.

This data demonstrates that improving patient satisfaction should be taken very seriously by healthcare providers. It is an operational imperative. In my experience, there are three things that hospitals can do to improve patient satisfaction, lower readmissions, and reduce cost to serve:

• Share these findings on the impact of patient satisfaction with physicians and other hospital staff. Once they recognize the link between patient satisfaction and positive outcomes, they will approach the subject more seriously.
• Analyze your patient satisfaction data to identify the two or three questions that have the greatest impact on patient satisfaction. Most hospitals focus on too many things. Determine the critical drivers of patient satisfaction and focus on them every day.
• Share ongoing results broadly across the entire hospital. Identify departments in your hospital that are strong performers and learn from them.

Available data suggests that we’ve been approaching patient satisfaction in the hospital setting with the wrong mindset. Instead of weighing the cost of cultivating highly satisfied patients, we should be looking at the bottom-line benefits of committing to a superior patient satisfaction program. If the patient feels positive about their experience and the hospital improves their financial position in the process, everyone wins.