Your comments are very valid. This will certainly challenge most, if not all, organizations to consider the level of service they offer their patients. I can also see your point to using a carrot (bonus based on patient satisfaction) instead of a stick (no reimbursement). The status quo might allow some organizations a few more resources to improve, but it also might prolong poor performance. You are also right that surveys have inherent data collection problems, and I think these will improve over time.

There is also evidence surfacing that high patient satisfaction scores are not necessarily correlated to quality outcomes. If that is the case, then we will have to question whether patient satisfaction is more of a business performance issue than about clinical outcomes? I do believe that we all want to be treated in a professional and empathetic manner, and that the way we feel about our care can impact our healing or the eventual outcome. There are too many anecdotes where providers go the extra mile to help a patient stay motivated or keep hope alive that it saves their life. I think that satisfaction does have an affect out outcomes, but we don’t yet understand that affect nor to the degree of that influence.

Let me ask you a couple questions.

1) Are hospitals compelled/mandated to offer their services? In most cases, not all, they are not mandated to provide certain levels of care. They can choose their status or category, etc if they choose.

2) Should hospitals have the liberty to choose which services are evaluated by a stakeholder for quality purposes?

It could be argued that hospitals generally have a choice about which services they offer and possibly could opt-out of the services that they don’t want to be evaluated. However, they should be held accountable for the responsibility with which they are entrusted.

Unless you’re referring to patients dying, I’ll also have to challenge your perspective that “some things cannot be fixed.” We should have the professional will and personal humility to never give up. We’re doing things today that were never possible or event dreamed of at one time.

Patients generally judge quality against the best possible outcome (as described by the one providing the service) and how they were treated. Because the second measure is not objective should we not measure it? If we do think it is worth measuring what should we do with that data? Tying it to compensation is certainly one way to get action.

Thoughts?

I am not saying that patients should put up with poor care. That is not my point at all. But I do know that taking away support will not make things better as personal get distressed and people are not at their best when they are distressed. Patients are human beings, not machines and some things cannot be fixed. Every person sees situations from different perspectives. It is difficult to be everything to everyone which is what is required to get perfect scores. Also, it is well known that surveys are not the best way to measure as it is purely subjective. How do we know that the patient is the one filling out the survey? It could be an upset family member that staff did everything they could to please and nothing was acceptable. Do we really want to go down this path?

These HCAHPS scores were derived by first assigning each possible survey response a numerical value. The following are the values assigned to the possible responses:

Patients who gave a rating of 6 or lower (low) = 6 points
Patients who gave a rating of 7 or 8 (medium) = 8 points
Patients who gave a rating of 9 or 10 (high) = 10 points
Always = 10 points
Usually = 5 points
Sometimes or Never = 0 points
Yes (Yes Definitely) = 10 points
Yes (Yes Probably) = 5 points
No = 0 points

The percentage of responses (given in the HCAHPS data) for each possible answer is then multiplied by the point value/multiple per answer. The answer scores are then summed per question. That summed value is then multiplied by 10 to put the score on a common 100 point (100%) scale.

Here’s an example:
Survey Question: How often did staff explain about medicines before giving them to patients?

Responses, Point Value Multiple * Answer Percentage:
Staff always explained, 10* 57% = 5.7
Staff usually explained, 5*22% = 1.1
Staff sometimes or never explained, 0 * 22% = 0
Sum = 5.7 + 1.1 + 0.0 = 6.8

Multiply that question sum by 10 to get to the 100 point (percentage) scale: 6.8*10 = 68.0

Each question is then summed and an average is given per state.

You’ll notice that a lot of the answers to questions have ranges (e.g. 0-6), but I give the state the benefit of receiving a 6. I mention this because the average response for that question is probably not the full 6 points, so these state averages might even be a little lower than reported.

I did a little more research and found that Sarasota Memorial’s average is 79.07. That average is above the Florida average (75.40) but below the national average (80.23).

thanks…shawn
twitter: @smhcs