Colagiuri, B., & Zachariae, R. (2010). Patient expectancy and post-chemotherapy nausea: A meta-analysis. Annals of Behavioral Medicine, 40, 3–14.

To evaluate the relationship between patient expectations of postchemotherapy nausea and subsequent reports of nausea

Databases searched were MEDLINE, PsycINFO, and CINAHL.

Search keywords were expectancy, expectations, placebo effects or placebo responses with chemotherapy and nausea or emesis.

Studies were included if they

• Measured patient expectation for post-chemotherapy nausea.
• Reported occurrence, severity, or frequency of post-chemotherapy nausea.

Studies were excluded if they focused only on anticipatory nausea.

The authors independently coded the sample characteristics, the independent and dependent variables, and whether the study met the inclusion criteria for each study. Differences were discussed, and a final assessment negotiated for each study. A global effect size was calculated for each study. No formal scoring was used to assign a quality score to each study. Publication bias was addressed using the Egger test and calculations of failsafe.

• The review consisted of 17 studies.
• The average study sample size was 141 participants, ranging from 29–67, and totaling 1,497 patients.
• Patients with breast cancer were most frequently studied.
• Ten studies involved highly emetogenic chemotherapy (HEC), and four involved therapy with low emetogenic potential.

• Patients who expected to experience nausea were almost twice as likely to experience postchemotherapy nausea (pooled effect size, Cohen’s d = 0.35, OR = 2.07, p < 0.0001).
• No significant relationship was found among nausea frequency, severity, duration, or peak nausea level.
• Findings differed based on the timing of measurement in the study. Larger effect sizes were seen in studies in which post-treatment nausea was assessed after multiple infusions compared to those in which nausea was assessed after the first infusion.
• A slightly higher effect size was seen in studies that controlled for nausea history.

This meta-analysis confirmed that nausea expectations and experience of post-chemotherapy nausea are positively correlated. The discussion also identified research methods, such as incorporating expectancy-based manipulations that consider the source of the patient’s expectations, actively investigating the possible moderators of the association between expectancy and post-chemotherapy nausea (e.g., age, anxiety), and measuring the expectancies before the first chemotherapy administration.

• Methods and timing of nausea measurement varied among studies, which influenced effect sizes seen here.
• No subgroup analysis was provided based on the level of emetogenic potential of the chemotherapy, which would be expected to influence actual symptom occurrence.
• Forms for abstraction of data (no documented protocol) were not used and, therefore, not available for replication or expansion of the meta-analysis.

Prechemotherapy assessment needs to include patient expectations regarding the experience of postchemotherapy nausea. If a patient expects that he or she will experience nausea, additional evidence-based interventions to minimize nausea should be used prior to and during chemotherapy administration.