New client satisfaction tool offers better data on improving care

NHPCO provides benchmarking services with new tool

This month’s cover story is the first of a two-part series on client satisfaction in hospice care. The first story in the series delves into a new satisfaction survey that will help hospices spot areas where improvement is needed. In the next issue, we will provide additional client satisfaction tips offered by hospice organizations.

Too many hospice client satisfaction surveys do more to boost hospice staff morale than to generate suggestions for improvements. The National Hospice and Palliative Care Organization (NHPCO) in Alexandria, VA, used such a self-congratulatory survey for years before deciding to revamp the survey with the goal of giving more useful feedback to hospice organizations.

"As consumer evaluation of health care research has evolved, it has become clear to us that the tool wasn’t helping us improve quality," says Stephen Connor, PhD, vice president of NHPCO. "It had a leniency bias, so people were not inclined to answer questions in a way that would admit there was a problem," Connor says. "It was a back-patting exercise that made everyone feel good but didn’t really help us improve care."

NHPCO asked for help in revising the tool from researcher Joan Teno, MD, MS, professor of community health and medicine at Brown Medical School and associate medical director of Home and Hospice Care Rhode Island in Providence. Teno and co-investigators spent seven years developing a tool for family evaluation of hospice care that could be validated and compared for national benchmarking purposes. "It’s been a lot of work," Teno says. "We shortened the tool, put it into a format to be self-administered, and helped to build a Web site that NHPCO has further redesigned to allow hospices to send in their satisfaction data and obtain a report with national benchmarks."

Previously, satisfaction tools used ratings ranked on a scale from "very good" to "very dissatisfied," Teno says.

"Everybody either said care was excellent or good or they were very satisfied," she says. "The classic example is when the family member reports that the patient was in moderate pain during the last three days of life and then for the next question would say they were very satisfied with the level of pain control."

So the new tool asks bereaved family members about specific processes of care that they observe. For example, a question might be, "Did the doctor tell you how to administer the pain medication in a way you can understand?"

This is very different from asking how satisfied a person is with pain control, and it’s more likely to produce useful information, because if a person doesn’t understand how to administer pain medications, then there likely will not be effective pain control, Teno says. "We tried to come up with the key important areas to measure end-of-life care in the United States," Teno says.

The model researchers developed is based on these tenets:

  • Patients desire a certain level of pain control and emotional support.
  • All health care providers should promote shared decision making.
  • Health care workers should treat a dying person with respect and dignity.
  • Health care workers should attend to family members’ need to have information about what to do and should provide emotional support both before and after the patient’s death.

After researchers field-tested the satisfaction tool, they addressed various concerns expressed by hospice providers, including a concern that family members might not want to fill out a long survey form, Teno says.

Four domains of information

The tool originally had eight domains, but it was shortened to these four domains:

  • informing and communicating about the patient’s care;
  • providing information about symptoms;
  • providing coordinated care;
  • tending to the family’s needs.

The new and revised survey has the same response rate as the previous one (about 50%), but it provides more valuable information to hospice providers, Teno says. When investigators spoke with hospice providers, they learned that some would let the survey results pile up on a corner of their desks and not fully analyze the results, mostly because they weren’t sure how to do the analysis, Teno says. "So we gave them an Internet tool that allows them to analyze it and provides them with benchmarks, as well," she says.

Since January 2004, the tool has been used to collect data that are uploaded to NHPCO each quarter and analyzed for a benchmarking report, Connor says. "There is an individual hospice report with comparisons on each question and domains for their state and national percentiles and significance," he says. "Right now we have over 500 hospices using it, and that amount increases every quarter."

Some hospices already have made important program changes based on the feedback they’ve received from the family satisfaction tool results, Connor says.

For example, the Hospice of St. Mary’s in Leonardtown, MD, created a new volunteer position that involves the volunteer meeting with each hospice family and patient to find out how they are doing with the specific areas covered by the satisfaction tool. This proactive measure has resulted in improved satisfaction ratings and results that are better than the benchmark averages, says Charlene Taylor, RN, BSN, hospice director. "We have discussed the idea of identifying programs that are particularly exemplary and seeing if they wished to share information about how they are doing it," Connor says. "But at this point we’re very early in the process."

One of the chief findings resulting from the tool’s development and use is that patients are often referred to hospice when it’s too late to provide optimal end-of-life care, Teno says. "From my standpoint, what’s frustrating is how late hospices get people referred to them, so that it’s difficult to do anything except acute symptom management," she says. "Hospices don’t get the opportunity to provide emotional support to family members."

Teno and colleagues published results of their research in the January 2004 issue of the Journal of the American Medical Association. The study found that more than two-thirds of Americans die in an institution. Of the people who died at home, 38.2% did not receive nursing services, while 49.3% had home hospice services.1

The study, which interviewed family members or other people who were knowledgeable about the deceased person’s end-of-life care, found that many people who are dying in institutions have unmet needs for symptom amelioration, emotional support, and physician communication. "We used the survey results from people who died at home with hospice as benchmarks," Teno says. "Some of the consistent findings in the study are that there are opportunities to improve care across all sectors of the health care system, even hospice."

However, hospice uniformly had a lower rate of reported concerns and higher overall satisfaction, Teno adds. "Another big picture is the emerging rate at which nursing homes are now being the last place of care for older frail persons and the significant concerns that bereaved family members have with that last place of care," Teno says. "Nursing homes had some of the higher rates of concerns and some lower satisfaction."

These findings are a reflection of how the United States finances its health care system, Teno says. Nursing homes are not adequately financed and staffed to provide end-of-life care, which has led to staff shortages and high rates of turnover in those facilities. This means hospice has an important role to play in providing help, especially given that increasing numbers of people are going into nursing homes, she adds.

National statistics indicate that about one in four people dies in a nursing home, up from fewer than one in five in the late 1980s, Teno says. "Health care providers don’t do a good job of effectively communicating with patients and families about the possibility that the patient is dying," Teno says. "I think hospices need to partner with nursing homes or hospitals and really try to figure out how to create systems that facilitate communication and earlier referral."


1. Teno JM, Clarridge BR, Casey V, et al. Family perspectives on end-of-life care at the last place of care. JAMA 2004; 291:88-93.