Contact Us | Location

Improving Informed Consent in Childhood Cancer Trials


by Eric Kodish MD Source: Fall 2003 CCCF Newsletter

Informed consent means different things to different people. In the most concrete sense, informed consent is a document that individual(s) can decide to sign or not sign. A more authentic definition of informed consent is that of a communication process related to a medical treatment or research decision. For families, health care professionals and children with cancer, informed consent can be a confusing and difficult challenge. One of the most perplexing challenges is the ambiguity that surrounds differences between consent for treatment of childhood cancer and consent for a child’s participation in a research clinical trial. For the past five years, funding from the National Cancer Institute has supported an effort at Rainbow Center for Pediatric Ethics, Cleveland Ohio, to better understand the informed consent process that takes place after a child is diagnosed with acute leukemia. Based on findings from the first four years of research, we have recently initiated a second study that attempts to improve informed consent.


This article summarizes results of the Project on Informed Consent (PIC), a limited institution CCG (Children’s Cancer Group) legacy study aimed at describing and understanding informed consent processes in pediatric leukemia randomized trials. Its key aims were to: 1) generate data to inform scientific understanding of the process of informed consent for clinical trials in childhood cancer; 2) identify potentially vulnerable sub-groups who may benefit from targeted interventions to improve informed consent; 3) work together with a group of parents to identify barriers to informed decision-making, and recommend measures to improve the consent process. This article highlights the critical role parents have played in making the PIC a success, and their ongoing role in the Multi-Site Intervention to Improve Consent (MUISIC) study.

In the PIC, 140 parents were recruited from six CCG institutions that routinely treat children with acute leukemia: Rainbow Babies & Children’s Hospital of University Hospitals of Cleveland in Cleveland, Ohio (21 parents); Children’s Hospital of Philadelphia in Philadelphia, Pennsylvania (42 parents); Children’s Hospital Medical Center in Cincinnati, Ohio (16 parents); Children’s Hospital of Los Angeles in Los Angeles, California (36 parents); MD Anderson Cancer Center in Houston, Texas (5 parents); and Children’s National Medical Center in Washington, D.C (20 parents). Data collection took place from July 1, 1999 until December 31, 2001.

Trained researchers observed and tape-recorded the Informed Consent Conference(s) (ICCs) that clinicians convened, for the purposes of discussing treatment options, including participation in a Randomized Clinical Trial (RCT). Each taped conference was later coded using the Observer Checklist, an instrument developed to code behaviors specific to clinical discussions related to cancer. Our version of the Observer Checklist includes codes for behaviors relating to discussion of the disease, its treatment, and participation in a clinical trial.

Parents were interviewed within 48 hours of their informed consent conference with their child’s clinician(s), and then interviewed again once they had made a decision about whether or not to participate in the RCT. To avoid conflicting responses from multiple family members, we limited our interviews to the parent or surrogate who was most active during the audiotaped informed consent discussions whenever possible. We designed the parent interviews, available in English and Spanish, to elicit information regarding parents – understanding of the informed consent information, as well as factors related to their decision to participate (or not) in the clinical trial for the treatment of their child’s leukemia. Follow up interviews were conducted over the telephone 6-8 months after diagnosis, and included items to assess long-term understanding of issues related to clinical trial participation.

Our methodology provides an accurate and comprehensive description of what actually transpires during the informed consent process. Carefully collected data on parental understanding complements this descriptive data, allowing us to link disclosure and understanding for each case we have studied. Clinicians who participated in an audiotaped informed consent conference were asked to complete a questionnaire relating to the information that was given to parents/ patients and the clinician’s recommendations regarding treatment options. In addition to this case-specific questionnaire, clinicians at all six institutions were asked to fill out a general questionnaire about their attitudes toward informed consent.


One hundred forty of the 164 parents (85%) approached for consent to our study agreed to participate. Forty-four percent of these 140 parents were minority subjects, including 36 Latinos, 13 African Americans, 6 Asians and 6 other minority subjects. The diagnoses of the children included 125 with acute lymphoblastic leukemia, 14 with acute myelogenous leukemia, and one with myelodysplastic syndrome. These children included 80 male and 60 female patients, with a mean age of 7.0 years (range 1-18). Eighty-four percent of the 140 parents in our study consented to participation in a randomized clinical trial for their child.

The mean length of the observed ICC was 78 minutes (range 25-183), with 6.5 participants present at the conference (range 2-15). The clinical trials offered to these families included CCG 1952 (n=24), CCG 1961 (n=65), CCG 1991 (n=36) and CCG 2961 (n=15). We observed presentation of the consent document during the ICC in 95% of cases, most frequently taking place at the end of the conference. An attending physician was most active during the ICC in 67% of cases, and a fellow was most active in 33%. A nurse was present in 47% of cases, and a clergy member attended only two of the 140 cases we studied. English was used in 113 ICCs, a Spanish translator in 26, and a Cantonese translator in 1 case.

Most of the conferences (76%) we studied took place without the childpatient being present. Of the 33 cases in which the child was present during the ICC, 16 involved children younger than 8 years of age. Among the 17 ICCs that older children attended, the child’s participation in discussion of the RCT was quite limited, and parents asked fewer questions than conferences without the child.

Our analysis of the Observer Checklist and parent interviews provides a wealth of information about how clinicians conduct informed consent conferences and how parents understand the information conveyed to them. Key findings include the observation of an explanation of randomization in 74% of cases, an explanation of the consent document in 73% of cases, and an explanation of treatment options outside of the clinical trial in 89% of the cases. The trial was described as voluntary in 94% of cases and the right to withdraw was discussed in 71% of cases.

Despite extensive prompting during structured interviews, 46 of 140 parent interviews analyzed (33%) show no evidence that the parent was able to distinguish off-study treatment from study participation for their child with leukemia. Even more striking is the finding that 51% of the parents we interviewed did not understand that their child would be randomly assigned to treatment in the RCT. Thus, although clinicians generally do explain parental choice and randomization during the ICC, we have found that many parents do not comprehend these aspects of trial participation.

Another critical finding from this study is that the number of parent questions was positively associated with better comprehension of choice about the clinical trial and understanding of randomization. These findings suggest that understanding is related to parental participation as measured by question asking during the ICC.


After observing 140 ICCs, we conducted Focus Groups with many of the parents who participated in our study. These focus groups were designed to elicit the views of parents regarding the informed consent process, including constructive suggestions for improving the process. The focus groups were then followed by the establishment of a Parent Advisory Group on Informed Consent (PAGIC), comprised of parent representatives nominated from each of the focus groups. The PAGIC met together in October of 2002, and this was a remarkable event. A partnership was forged, and the parents were able to assist the research team with interpretation of the data by providing their perspective as important stakeholders in the consent process.

The most significant achievement at the first PAGIC meeting was the development of a new model for the consent process that has the potential to revolutionize informed consent. The model provides a staged approach to informed consent and depends on a feedback loop to confirm effective communication prior to moving to the next stage. Most importantly, it allows families and health care professionals the opportunity to individualize the structure of the conference to meet the cognitive and emotional needs of parents facing the devastating news of a cancer diagnosis for their child.


Our research has demonstrated specific gaps between information disclosed by clinicians and parental understanding of key informed consent issues. The next challenge is to utilize these findings to test rationally-designed interventions that may improve the quality of the informed consent process in childhood leukemia trials. To this end, we have begun a multi-site study to develop, test, and implement two databased interventions: one for physicians and one for parents. Members of the PAGIC continue to provide input and expertise as we develop the interventions, and several will be involved in training physicians to conduct the ICC based on our new model. We will be testing the effects of each of these interventions on key outcomes measures. Along with other efforts from the COG (Children’s Oncology Group) Bioethics Committee, we are taking the lead in proactive measures to insure the integrity of our clinical research and working toward optimal communication under extremely difficult circumstances. Efforts to bolster parental understanding, and augment parental participation during the ICC are likely to provide a significant improvement in informed consent.

Eric Kodish, MD Director, Rainbow Center for Pediatric Ethics Rainbow Babies and Children’s Hospital, Room 680 11100 Euclid Avenue, MS 5016 Cleveland, OH 44106-6054