Faria, C., Li, X., Nagl, N., & McBride, A. (2014). Outcomes associated with 5-HT3-RA therapy selection in patients with chemotherapy-induced nausea and vomiting: A retrospective claims analysis. American Health and Drug Benefits, 7, 50–58.

To evaluate the clinical and economic impact of delayed chemotherapy-induced nausea and vomiting (CINV) on patients who received initial and maintenance therapy with the 5-hydroxytryptamine 3 (5HT3) receptor antagonist palonosetron compared to older agents

There was no intervention in this retrospective study; however, the procedure was outlined. Using the OptumInsight^® database, researchers evaluated the impact of 5HT3 on chemotherapy-naïve patients. This database provided information on all pharmacy and medical claims for each subject, and records were accessed six months prior to the initial chemotherapy treatment and six months after. Patients were not evaluated if the type of antiemetic changed or if chemotherapy was changed. ICD9 codes and pharmacy charges were monitored for primary or secondary diagnoses of nausea, vomiting, and dehydration. Economic outcomes also were evaluated and calculated.

• N = 26,974
• AVERAGE AGE = 55.7 years
• MALES: 30.9%, FEMALES: 69.1%
• KEY DISEASE CHARACTERISTICS: Patients who were chemotherapy-naïve; not receiving multiday chemotherapy; 53% breast; 20% lung; remainder split between ovarian, colon, and other cancers
• OTHER KEY SAMPLE CHARACTERISTICS: 72% of patients had a diagnosis for a single site.

• SITE: Not stated
• SETTING TYPE: Not specified  
•  LOCATION: OptumInsight database

• PHASE OF CARE: Active antitumor treatment
• APPLICATIONS: Elder care and palliative care

Retrospective database analysis

No instruments were used, but information extracted from the database included patient and treatment characteristics, nausea, vomiting, Charlson Comorbidity Index scores, antiemetic therapy, and specific 5HT3 use.

Preindex comorbidity scores were lowest in the palonosetron group and highest in the dolasetron group. The overall rate for delayed CINV at cycle 1 was 15.6% for all groups. When compared to palonosetron, patients who received ondansetron (p < 0.002), granisetron (p < 0.001), and dolasetron (p = 0.002) had higher rates of CINV in the second and subsequent cycles of chemotherapy. Dexamethasone was consistently used in the first cycle for all treatment groups. Aprepitant was used most often in the palonosetron group (10.7%) compared to ondansetron (3.6%), granisetron (2.3%), and dolasetron (2.5%).

5HT3 agents were effective in preventing CINV. There were differences in 5HT3 efficacy and cost. Delayed CINV rates increased with subsequent cycles of older 5HT3 agents. Palonosetron showed improvement over time. Similar trends were seen with healthcare resource use.

• Risk of bias (no control group)
• Risk of bias (no blinding)
• Risk of bias (no random assignment)
• Risk of bias(sample characteristics)
• Key sample group differences that could influence results
• Measurement/methods not well described 
• Other limitations/explanation: Comorbidities were lowest in the palonosetron group. The palonosetron group received aprepitant more often than the other groups. This may have influenced the outcome measures. This was a claims analysis, and cases of delayed CINV may not have been captured, especially when patients were given a supply of antiemetics to use at home. This study only included patients who were commercially insured. It did not capture any complementary methods that may have been used (i.e., acupressure, nutraceutical).

This study demonstrated cost effectiveness. Medical costs constituted the largest costs. Costs were higher if a patient experienced CINV. Nurses need to use guidelines and risk factors when starting patients on chemotherapy. Using a 5HT3 agent based on the emetogenic potential of chemotherapy is important.