Patient experiences

In clinical trials, it is important not to leave the patient out of the human equation.

When people think of clinical trials, they tend to think of controlled experiments in which clinicians measure heart rate, electrolytes, breath volume, and the like. But what about the patients? What is happening to them? How do they feel? What are they thinking?

As the art of the clinical trial evolves, measurements of patient experience are becoming a larger and more important part of the process.But how this experience should be measured, and even who should measure it, remains controversial.

Ultimately, the purpose of patient experience files is to determine the efficacy of a drug or treatment.In “In the mind’s eye”, Cullen T. Vogelson discusses the role of such observations from the perspective of the trial clinician and the regulatory concerns faced by the researcher.

It is equally important, though, that we not dismiss the human side of the equation. The personal experiences of clinical trial participants are hard to define and quantify by a series of questions and answers.This article addresses the problem of measuring and evaluating patient experiences.

According to psychologist Saul Shiffman of the University of Pittsburgh, most measures of patient experience fall into three categories. In the first, patients report their own symptoms. For certain disease states, such as allergic reactions, bowel dysfunction, and eating disorders,the patient is almost certainly in the best position to track symptoms and report them. In the second category, patients may be asked to report symptoms that would be difficult, if not impossible, for an independent observer to measure, such as pain, fatigue, or anxiety levels. Shiffman distinguishes between these two categories on the basis of the ability of a third-party observer to make the same measurements.

A third type of measurement focuses on the patient’s quality of life.Many believe that quality-of-life assessments should be the aim of patient experience measurements. After all, when a person is told that his or her blood pressure is high, it is the potential impact of this measurement on the individual’s quality of life (and not the measurement itself) that provides the motivation to seek medical treatment.

Information on quality of life is sometimes self-reported, but it may also be assessed by a clinician. Insurance companies and regulating agencies, which frequently insist that new medical treatments improve patient quality of life,are increasingly asking for quantitative measurements of this variable.

Quality of life is often evaluated in terms of a patient’s sense of well-being and abilityto function physically, socially, and occupationally; however, it is far from an unambiguous term. Not surprisingly, some researchers have begun asking themselves whether they can ever justify second-guessing a patient’s experience in a clinical trial.

According to Shiffman, when an individual has a new experience, he or she encodes it in memory for later recall. When the experience is recalled, a sequence of events is reconstructed. But when an individual attempts to reflect on the experience, current events, such as a family quarrel that morning, may influence the way the experience is recalled.Because each processing state may be subject to biases influenced by present circumstances, Shiffman believes that retrospective reporting by a patient may not be the best way of learning about a patient’s experience.

Others have pointed out, however, that patient experience measured by a clinician may have limited relevance to a patient’s life. Psychometrician Sonja M. Huntof the University of Edinburgh wrote, “It is not the presence of symptoms or limitations of function which affect quality of life, but rather the meaning and significance of them for individual patients. Thus, patients with an identical health status may experience a range of existential states from despair to happiness. In other words, ordinary people are much less obsessed by their health than health researchers tend to believe.” Hunt co-developed a widely used health status measurement tool known as the Nottingham Health Profile.

A conventional method of evaluating patient experience is to ask the patient to respond to a series of questions that can be answered with yes-or-no responses or selections from multiple-choice options. Examples of these questions are shown in the box, “Quality-of-life questionnaires”.

Most questionnaires are not designed to evaluate subjective experience, however,so respondents may find themselves giving misleading or incomplete accounts of their experience for the sake of statistical efficiency.To address this problem, some questionnaires are designed to capture individual rather than group experience. For example, an individualized patient questionnaire focusing on quality of life might ask a patient to list areas of life that have been most affected by a health condition. This may be particularly appropriate when the spectrum of experience represented in a given patient sample is very wide.

A few researchers prefer to capture patient experience by using what the patient says or writes about his or her own experience. The investigator then breaks the patient’s comments into clauses and notes any emerging themes. A health-versus-sickness evaluation, for example, would look for statements by the patient that pertain to symptoms of illness or feelings of well-being. A significant drawback of this approach is that clinicians may interpret the patient’s comments differently because of differences in their own past experience or philosophical biases. Nevertheless, this type of content analysis has been used for more than 30 years.

Regardless of the method used, it is difficult to equate the feelings of a patient with (or separate them from) the objective goals of a clinical trial, if the goals are more complex than the elimination of a bacterium or the disappearance of a symptom.

From the clinician’s perspective, it might be beneficial to involve patients who are well informed about their condition. For patients faced with the option of participating in a clinical trial or undergoing conventional therapy, access to a patient experience database such as that offered by researchers at England’s Oxford University (discussed in the next section) would almost certainly be a godsend. As Lawrence B. Afrin of the Medical University of South Carolina, Charleston, says, “The more information one has about an illness, the less likely it is that one will make poor decisions about how to manage that illness.”

Georgia R. Sadler, professor of surgery at the University of California, San Diego, has noted that many clinical trial participants complain that they are never told what the research study concluded. As a result, many participants never learn how their own experience stacked up against those of others in the trial. Access to a patient experience database could go a long way in helping patients to learn how well or poorly their experiences were interpreted.

When people receive an unexpected diagnosis of a life-threatening illness, their first reaction may well be fright and bewilderment, especially if they know little about the disease. More than anything else at this time, they need good information about the disease, its treatment, and how the illness might affect the rest of their life.

At Oxford University, researchers are building a database of patient experience that will give patients, caregivers, health professionals, students, policy makers, and medical researchers immediate access over the Internet to just this kind of information.

The Oxford project, called the Database of Individual Patients’ Experience (DIPEx), will consist of a systematic collection and analysis of interviews with people about their experience of illness, evidence of the effects of treatment, and information about support groups and other resource materials. Plans call for the Web site to eventually contain information on approximately 3000 medical conditions.

According to Andrew Herxheimer, one of the principal investigators in the project, patient experience will be captured in the patients’ own words, and supplemental questions will be asked in such a way that there is no question that it is the respondent’s experience, and not the interviewer’s agenda, that is being recorded.

The DIPEx team learns of prospective interviewees primarily from family physicians, medical specialists, and self-help groups. Although it is expected that the DIPEx database will eventually includeinterviews with patients taking part in clinical trials, the project will initially focus only on the illness experiences of patients outside clinical trials, that is, in ordinary clinical practice.

But Sadler remains skeptical that such a database could ever represent the full range of experiences of patients with a particular illness. “Any follow-up over time will be self-selective. People who do very well may be more likely to report their experience. They will continue to tell you how lucky they feel to be alive. But people who are really debilitated don’t have that option. They don’t have the energy,” says Sadler.

Herxheimer is careful to point out that this type of sampling is not intended to be statistical. “The method does not try to get a representative sample of the population of people with the illness,” he says. “Such a sample would be inappropriate for our purpose.”

The database might also be of use to clinicians who are considering a future trial such that, although not directed to their drug of interest but to a related drug, they will be made aware of what to expect from their future patients.This information would also help them prepare potential participants for the trial.

The DIPEx team currently maintains a prototype Web site (www.dipex.org).Functioning databases on hypertension and prostate cancer are planned for later this year.

• Stone, A. A.; Shiffman, S. Ecological Momentary Assessment (EMA) in Behavioral Medicine. Ann. Behav. Med. 1994, 16, 200–202.
• Herxheimer, A.; et al. Database of patients’ experiences (DIPEx): A multimedia approach to sharing experiences and information. Lancet 2000, 355, 1540–1543.
• Yaphe, J.; et al. The use ofpatients’ stories by self-help groups: A survey of voluntary organizations in the UK on the register of the College of Health. Health Expectations 2000, 3, 176–181.
• Individual Quality of Life: Approaches to Conceptualization and Assessment; Joyce, C. R. B., O’Boyle, C. A., McGee, H., Eds.; Harwood Academic Publishers: Toronto, 1999.