Skip to main content

Randomized controlled trial of an intervention to improve nurses’ hazardous drug handling (Friese et al., 2019)

Review Guidelines

Absence of conflict of interest.


Friese, C. R., Yang, J., Mendelsohn-Victor, K., & Mccullagh, M. C. (2019). Randomized controlled trial of an intervention to improve nurses’ hazardous drug handling. Oncology Nursing Forum, 46(2), 248-256.


The study’s objective was to examine the impact of the intervention on the health and safety of oncology nurses through their use of personal protective equipment (PPE) during hazardous drug administration.

Features of the Intervention

Evidence from prior research indicates that nurses who administer antineoplastic drugs, which are used in chemotherapy treatment for cancer, are more likely to experience certain health issues and may lead to persistent issues through unintentional exposure in the workplace. 

The intervention involved three components: (1) a one-hour educational module discussing the principles of safely handling hazardous chemotherapy drugs, (2) short videos discussing barriers to using personal protective equipment (PPE) tailored to the specific barriers each participant identified, and (3) a request to report spills of chemotherapy drugs and to submit plasma samples quarterly, after which they viewed a brief video describing hazardous spills and plasma results across participants over the prior quarter.  

The study targeted nurses working an average of at least 16 hours per week in ambulatory chemotherapy clinics within academic medical centers. Most nurses in the study had a Bachelor of Science degree in nursing, and study participants had an average of 12 years of experience in oncology nursing and managed an average of 5.5 patients. 

Features of the Study

The study used a cluster randomized controlled trial design to compare a group of nurses that received the intervention designed to encourage use of personal protective equipment when handling hazardous chemotherapy drugs to a control group of nurses.  

The study randomly assigned 12 hospital sites to treatment or control conditions. The participating sites were academic medical centers with ambulatory chemotherapy departments. The hospitals were randomized to study conditions as clusters., and within the 12 sites, 439 nurses were eligible to participate in the study. Four hundred fifteen (95%) of the eligible nurses enrolled in the study. One-hundred eighty-nine of these nurses were from sites assigned to the treatment group, and 226 were from practice sites assigned to the control group. 

The treatment condition involved educational modules, tailored information, and reporting on hazardous spills. The control condition could view an hour-long educational module that the intervention group also had access to.  

The study administered surveys through a study website. The surveys were administered at the start of the study and 12 months later.  

The authors used a statistical model to compare the outcomes of treatment and control group members. The study’s primary outcome was nurse-reported PPE use during hazardous drug administration.

Study Sites

Twelve sites were randomized via cluster randomized control design to either treatment (n=6) or control condition (n=6).  


The intervention did not have significant effects on the use of personal protective equipment among nurses delivering chemotherapy treatment. 

Considerations for Interpreting the Findings

The study has high attrition at the cluster level because one cluster in the control condition dropped out of the study. In addition, there was high attrition within clusters as well. 

The study used a linear mixed model with random intercepts. The study did not report a specification test or justification for their choice to use random effects rather than fixed effects. However, the authors did adjust for repeated measures from each nurse and hospital. 

Causal Evidence Rating

This study receives a low evidence rating. This means we are not confident that any estimated effects would be attributable to the intervention and not to other factors. The study did not find statistically significant effects. 

Reviewed by CLEAR

March 2022