Collecting and Interpreting Life Events Data in Clinical Trials
Collecting and Interpreting Life Events Data in Clinical Trials
If you want to live a happy life, tie it to a goal, not to people or things.
–Albert Einstein
The trouble with always trying to preserve the health of the body is that it is so difficult to do without destroying the health of the mind.
–G.K. Chesterton
BACKGROUND
Life events, such as a marriage, death, pregnancy, change in job, change in personal relationships, or moving to a new residence, are seldom recorded in clinical trials, apart from patient-reported outcome trials (quality of life). Nonetheless, there is a growing awareness that these events may influence the clinical interpretation of data or may confound the trial’s results, particularly in regard to quality-of-life issues.
A disproportionate number of either positive or negative life events that are not evenly distributed between treatment groups may lead to a false-positive or false-negative result in the trial. Ultimately, this could prevent a beneficial therapy from being marketed or an ineffective treatment from being marketed.
This chapter presents a number of approaches for systematically collecting, quantifying, and interpreting life events data.
Why Measure Life Events in a Clinical Trial?
In clinical trials that take place over several months or longer, external events in subjects’ lives may influence their responses to treatment or other interventions. To assess whether this has occurred and how life events may have influenced subjects, it is necessary to collect data on both events and their significance.
Collecting life events data should always be considered in long-term trials because the longer the trial, the more likely it is that collecting life events data will affect the interpretation of the collected data. It is particularly relevant to collect this information in quality-of-life trials because of the large potential for life events to affect quality of life independently of the effects due to the test treatment.
Can Life Events Occur as a Result of Drug Benefits, Adverse Events, or the Clinical Trial Process Itself?
Most life events that occur during a clinical trial (Table 64.1) occur independently of both the treatment received and trial procedures. Nonetheless, there may be a direct or indirect connection between the two. For example, a subject whose symptoms improve with therapy may, as a result, be able to obtain a job, earn a promotion, take an important trip, get married, or interact more effectively with important people in his or her life.
Table 64.1Examples of life events
A.
Family
1.
Pregnancy, abortion, miscarriage, birth
2.
Marriage, separation, divorce, reconciliation
3.
Quarrels or fights (increases or decreases in number or intensity)
4.
Death of friend, relative, or acquaintance
5.
Illness (self, child, relative, or close friend)
6.
Children have many important changes in their life that influence their parents.
B.
Work
1.
New job, new responsibilities
2.
Resigned, fired, or other major change
3.
New hours, conditions, or coworkers
4.
Quarrels with coworkers or superiors
5.
Longer hours
6.
Greater stress
C.
Social
1.
Personal habits are altered
2.
Social interactions change in number or frequency
3.
Vacation taken
4.
Eating, sleeping, or other regular patterns are altered
5.
Change in time spent on recreation, hobbies, sports
6.
Move to a new home
D.
Economic
1.
New loans taken out (or repaid)
2.
Bankruptcy
3.
Major change in status
4.
Make a major purchase
The opposite can also occur. The underlying cause of all adverse events should be assessed to determine whether life events might have had a role in their etiology. For example, a subject in a clinical trial may experience exacerbated or new symptoms (i.e., adverse events) that lead to a negative change in his life. Alternatively, if symptoms of the underlying disease worsen as a result of treatment, then that deterioration is classified as an adverse event causally related to treatment. Progression of disease independently of the trial may also occur. However, all new symptoms, clinical signs, or laboratory-measured abnormalities are classified as adverse events.
Positive (or negative) life experiences may result indirectly (or directly) from participation in a clinical trial and have nothing to do with the treatment received. For example, a subject may meet someone while attending a clinic visit and develop a romantic relationship, be physically attacked in the parking lot while entering or leaving a clinic, or have an automobile accident while driving to the clinic. Many other positive or negative life experiences may have a connection with a subject’s participation in a clinical trial and not with the treatment received.
COLLECTING LIFE EVENTS DATA
Collecting Life Events Data Prospectively
A separate data collection form can be used as a life events questionnaire. The data could be collected as written responses or as a checklist specifically designed to capture this information. Significant life events also can be documented by an open-ended question posed by an investigator, study coordinator, or other individual during the trial. The question could be phrased: “Have any events occurred during the past X weeks or months that have had a significant influence on your life?” This question may also be presented as a specific checklist or group of categories given to the subject to consider and complete.
In addition to identifying events, it is essential to capture the nature of the significance of each event for the subject. This may be accomplished with (a) a visual analogue scale using two or more anchors (e.g., extremely positive and extremely negative) to identify endpoints, (b) an open question, (c) a Likert scale, or (d) a numerical scale (e.g., see Tables 64.2 and 64.3). The preferred approach in most trials is to use a Likert scale or a numerical scale. Likert scales such as that shown in Table 64.4 provide appropriate labels that are relevant for subjects, physicians, or study coordinators. Measures of significance should be placed next to the descriptors of the life events so that significance data are collected at the same time.
If life events were collected without capturing data on the significance of the event for the individual subject, then the subject could be verbally asked for this assessment. If data on personal relevance of the life event were not collected during the trial, other approaches that could be considered include a telephone interview or an additional clinic visit.
Table 64.2Example of a form that may be used by a subject to identify and assess the relevance of life eventa
Subject:
Date: 01.02.09
Please identify all significant events that have affected your life since your last clinic visit and identify their relative importance in your life on the right as positive (+1, +2, +3, +4, or +5; +5 is the most positive) or as negative (0, −1, −2, −3, −4, or −5; −5 is the most negative).
1.
2.
3.
4.
5.
6.
7.
8.
Please circle the number that best describes the overall impact of all these events on your life (since your last visit). −5−4−3−2−1 0 +1 +2 +3 +4 +5
a Various formats can be used for the subject’s responses, such as ruled columns, boxes, and blank lines.
Table 64.3Example of a form that may be used by professional staff to collect data on life events in a clinical trial
Subject Number: _____
Visit: _____ (Date)
Interviewer: _____
Life events since last visit:
−5 to 0; 0 to +5 (a) (+5 is most positive, −5 is most negative)
(a) Based on your interview, rate the effect of each event on the subject.
Although these events may be in different directions, what was their overall effect on the subject (use the same scoring system)? (The subject must supply this integrated number.)
In most trials, life events data are a relatively minor parameter, so that a retrospective collection of life events data is possible if it is determined to be of value, even after a trial is completed, without a major compromise of its value. If life events are expected to affect or influence the primary endpoint of a clinical trial, then the failure to obtain data on their significance would be a major study flaw. For example, a treatment may be compared with placebo to evaluate whether socially withdrawn subjects increase the number and quality of their social interactions as a result of treatment. It is doubtful that any retrospective collection of data will convince readers and reviewers of the validity of the trial.
Only gold members can continue reading. Log In or Register to continue
