In excellent condition

Most health care providers probably wouldn't feel comfortable referring to their patients as "customers," but in many ways that's what they've become. Changes in the health care industry have forced providers to compete for business in a market whose consumers are making increasingly well-informed choices from the array of "products" available to them.

One outgrowth of this competition likely will be an increase in the number of hospitals and other health care organizations conducting research with their patients to gauge satisfaction levels and identify areas needing improvement.

An example is Bellevue Hospital, a privately-owned 40-bed facility in Schenectady, New York. The hospital, which sees about 2100 babies delivered each year, provides a variety of health care services for women, including obstetrics, gynecology, and outpatient clinical services. It draws its patients from 17 counties in and around the Tri-City area of Albany, Schenectady and Troy.

Bellevue is nearing the end of a yearlong ongoing patient satisfaction survey conducted by Fact Finders, Inc., an Albany, New York-based research company. Each quarter, 500 discharged patients are contacted by telephone two weeks after the end of their stay to take a three-minute survey on their impressions of the hospital's service.

Ellen Kerness, Bellevue's manager of marketing and public relations, says that the survey was commissioned despite the hospital's already strong reputation. "We'd always heard that the perception of Bellevue was that it was a wonderful place to go and that the care was excellent, from food services to nursing. But we wanted to see if what we were hearing through the grapevine was actually what was perceived by patients."

In each quarter of the survey the hospital has earned rave reviews from its patients in all service areas. For example, the nursing staff was rated extremely highly across the board.

Kerness says that the survey is an excellent tool that she would recommend to other hospitals, even a facility with a less solid reputation which might be hesitant to conduct a similar survey for fear of what it might tell them.

"It would be wonderful if they would do it, I think it would help them a lot. But whether they would be willing to do it is another question, because of the expenditure, and also because they might find out too much that's wrong. You might think you're doing OK and find out that you really aren't. We were very fortunate in that we found out that we were as good as we hoped we were."

For the Bellevue patient satisfaction study, an array of categorical (excellent/ good/fair/poor, very satisfactory/somewhat satisfactory/not satisfactory at all) and interval scale measurements are used to quantify patients' impressions of and level of satisfaction with services such as nursing staff, physicians, visiting hours, housekeeping, quality of food, comfort of the room, and hospital rules and procedures.

For example, some questions on the amount of time spent waiting at the hospital include:

Was the waiting area very comfortable, somewhat comfortable, or not comfortable at all?

Would you say you waited in the waiting room for a short, moderate, or long period of time?

Were you given an explanation for the wait?

The section of the questionnaire on nursing services ( developed in conjunction with the hospital's director of nurses) was customized for measurement of satisfaction with nursing services by service type and area (e.g., nursery, operating suite, labor and delivery, floor).

Some open-ended questions are included, asking patients, for example, if there was anything particularly negative or positive about their stay. And in the section on physicians, respondents who expressed dissatisfaction were asked to offer changes they might make to increase satisfaction.

The survey covered many service areas, but the interview length was kept short to avoid inconveniencing the respondents, says Sue Swartz, research associate, Fact Finders. "Our goal was to have an interview length not to exceed three to five minutes. The reason for that is, we were calling patients just after their discharge and we did not want to burden them. We wanted to get a quick yet comprehensive measurement of patients' opinions of the services received and the levels of overall satisfaction with the hospital experience."

The survey has enjoyed a very low refusal rate-of the nearly 2000 respondents so far, fewer than 10 have refused to participate. Swartz says that other healthcare-related telephone market research studies done by Fact Finders have also had relatively low refusal rates. "People are amenable to participate in research surveys with topics that are of interest to them, and health care clearly is of interest to most people, either for financial or personal reasons."

The telephone survey, says Ellen Kerness, "gives patients the opportunity, in an anonymous way, to say what they want to say. The wording of the questions and the shortness of the questionnaire makes it very easy for them to handle."

Patient's privacy

A primary concern during the study has been to preserve each patient's privacy. Patients are not asked about their health condition or the reason they were hospitalized. "Though the questions don't concern the patient's actual health problem, they do concern the services consumed, from which you could infer details about their health history. So we have been exceedingly careful about protecting the anonymity of the patient," Swartz says.

Respondents are chosen randomly from data tapes supplied by the hospital containing the phone numbers-but not the names-of discharged patients.

"The hospital does not know the identities of the individual patients who we have interviewed. We are sent the data in coded form, without the identities of the patients or the physicians. That is something we insisted on," Swartz says.

Interviewers tell each respondent that they are calling on behalf of Bellevue Hospital, that the research concerns patients' opinions about their hospital experiences, that the patient's phone number has been randomly selected, and that no names or identifying information is included.

"In the introduction we give them an adequate amount of information without adding too much length to the interview, so that they know who we are, what our purpose is, why they are being called, and what we will be asking of them."

The quarterly survey reports given to the hospital contain no physician, nurse, or other staff names, with the exception of the verbatim responses. The head of each department at Bellevue receives a copy of the results for use in monitoring department operation and to recognize the positive contributions of individual employees.

"At the request of the hospital, when a staff member is mentioned in a positive light, that person is named (in the report). The hospital can then go back to that person. So for management purposes, when there are positive remarks, we pass them along in the report," Swartz says.

Written evaluations

Kerness says that Bellevue will also use the information from the patient satisfaction study to make changes to the written questionnaire it distributes to patients during their stay at the hospital. "The telephone study has helped us get patients' opinions of some of the services that are not addressed at this point on the written evaluation. (The written evaluation) could be more effective than it is, but we don't want to drop it at this point, so we're trying to refine it a little. We want to get continual feedback to see how we're doing in 1991."

Prior to the telephone survey, the hospital relied on the written evaluations for feedback, but Swartz says that information gathered this way has limited management utility as it can be difficult to compile, analyze, and format.

"Most hospitals, in my experience, are not able to organize and act on the feedback that they get from self-administered questionnaires, so that they often end up with a grand tally of anecdotes and nothing actionable. But our reports give them, in tabular form, the proportions of satisfied/very satisfied, etc. and the verbatims, so that with the quantitative framework, they're able to put those anecdotes into perspective."

Responses from written evaluations also present another problem: they aren't projectable because of the factors that affect who does-and who doesn't-fill them out.

"Self-administered questionnaires appear to provide a feedback mechanism. However, if you understand the motivators that determine which patients fill them out, you realize that you are getting information, but it is not representative, objective information.

"When people have positive experiences as customers, they feel affiliated with the provider and are therefore more inclined to do something that would be helpful to the provider," Swartz says.

By contrast, someone who has had an unsatisfactory experience may decide not to complete a questionnaire because he or she feels it would be helping the hospital.

Hospitals reluctant

Swartz says that many hospitals are reluctant to spend the money on outside research, choosing to rely on in-house questionnaires to monitor patients' opinions. "Telephone interviewing can be an expensive methodology, and hospitals aren't interested in taking on any additional expense."

Still, she says, this type of study is becoming a very active part of the research field. "We view it as a category of customer satisfaction research, because patients clearly are customers of hospitals. In the past, health care consumption was primarily generated by doctors, They made the decisions and directed where the health services would be consumed. But consumers are now asking the doctors more questions and taking more of the responsibility for their own health. They are making their own decisions regarding where to go and how they shop for health care services."