This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Greenley, R. N. Articles by Kodish, E. Search for Related Content PubMed PubMed Citation Articles by Greenley, R. N. Articles by Kodish, E. Social Bookmarking                            What's this?

Stability of Parental Understanding of Random Assignment in Childhood Leukemia Trials: An Empirical Examination of Informed Consent

From the Department of Bioethics, Cleveland Clinic Foundation and the Lerner College of Medicine; and the Department of Pediatrics, Rainbow Babies and Children's Hospital of University Hospitals of Cleveland, Case Western Reserve University School of Medicine, Cleveland, OH

Address reprint requests to Rachel Neff Greenley, PhD, Department of Pediatrics, Division of Behavioral Pediatrics and Psychology, Rainbow Babies and Children's Hospital of University Hospitals of Cleveland, 11100 Euclid Ave, Cleveland, OH 44106-6038; e-mail: Rachel.Greenley{at}uhhs.com

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
PURPOSE: To examine stability versus change in parental understanding of random assignment in randomized clinical trials (RCTs) for pediatric leukemia and to identify factors associated with changes in understanding.

METHODS: Eighty-four parents of children diagnosed with acute lymphoblastic leukemia or acute myeloid leukemia who were enrolled onto a pediatric leukemia RCT at one of six US children's hospitals participated. Parents were interviewed twice, once within 48 hours after the Informed Consent Conference (ICC; time 1 [T1]) and again 6 months later (time 2 [T2]). Interviews focused on parental understanding of key components of the RCT, including random assignment. Interviews were audiotaped, transcribed, and later analyzed.

RESULTS: Changes in understanding of random assignment occurred in 19% of parents, with 17% of parents deteriorating in understanding from T1 to T2. Forty-nine percent of parents failed to understand random assignment at both times. Factors associated with understanding at both times included majority ethnicity, high socioeconomic status, parental reading of consent document, and presence of a nurse during the ICC. Physician discussion of specific components of the RCT was also associated with understanding at both times. Female caregivers and parents of low socioeconomic status were overrepresented among those who showed decay in understanding from T1 to T2.

CONCLUSION: Parents showed little gain in understanding over time. Factors that predicted understanding at diagnosis as well as sustained understanding over time may be important intervention targets. Attention to both modifiable and nonmodifiable barriers is important for clinical practice.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
Advances in the treatment of childhood leukemia during past decades have been closely tied to the use of randomized clinical trials (RCTs).¹ Federal mandates require that the legal guardian give informed consent for child participation before a RCT is initiated, and in RCTs for pediatric leukemia, this involves, in part, facilitating parental understanding of the concept of random assignment (ie, that their child will be randomly assigned to one of multiple treatment groups). Our previous work suggests that, although the vast majority of physicians explain random assignment (83%) and present a consent document (95%), many parents (50%) still fail to understand random assignment.² Although deficits in understanding of random assignment have been documented previously,^2-5 few studies have examined changes in understanding of random assignment over time, despite theoretical attention to the importance of studying this construct.⁶ Previous longitudinal studies conducted with a range of adult populations (eg, cancer patients,⁷ HIV patients,⁸ elderly,⁹ and pregnant women¹⁰) have yielded mixed results concerning the trajectory of understanding and are limited by a focus on only short-term changes in understanding.

Parental understanding of random assignment may be inhibited at diagnosis by emotional distress and by pressure for a decision about treatment course to be made quickly.^11,12 Understanding may improve as the shock of the diagnosis dissipates and as the family's routine is re-established. Thus, an examination of parental understanding later in treatment may yield results different from those documented at diagnosis. Moreover, attention to factors that predict changes in understanding over time may identify subgroups for whom ongoing education about the RCT is especially important. For example, Kodish et al² documented that high socioeconomic status (SES), ethnic majority status, nurse presence during the Informed Consent Conference (ICC), and physician discussion of specific RCT details were associated with higher parental understanding at diagnosis.

The present follow-up study was guided by the following two aims: (1) to document stability versus change in parental understanding of random assignment over a 6-month follow-up period after diagnosis, and (2) to identify factors (eg, demographic, parental, physician, relational, and contextual) associated with change in understanding over the follow-up period among families of children enrolled onto an RCT.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
Procedure
Study data were drawn from Informed Consent in the Children's Cancer Group,² a multisite longitudinal investigation of families of children with recently diagnosed leukemia focused on the informed consent process in childhood RCTs. Participants were recruited from a pool of consecutive patients with acute lymphoblastic leukemia (ALL) or acute myeloid leukemia (AML) between July 1, 1999 and December 31, 2001 at six US children's hospitals. Institutional review board approval was obtained from each site before data collection. Shortly after diagnosis, a research assistant obtained parental consent and, when appropriate, child assent from eligible families (ie, child assent was obtained when children were involved in the ICC). The treating physician also provided verbal consent. One hundred sixty-four families were approached, and 140 (85%) agreed to participate. After obtaining consent, research assistants observed and audiotaped the ICC, which is the physician-family meeting during which the RCT is explained. The ICC audiotape was later transcribed and coded by research assistants using the Observer Checklist,¹³ which is a system designed for use with clinical cancer-related discussions. Within 48 hours of ICC completion, the dominant parent (ie, the most active parent during the ICC) was interviewed by a research assistant (time 1 [T1]); and approximately 6 months after the ICC, the dominant parent completed a follow-up telephone interview (time 2 [T2]).

Participants
Participants included families of children diagnosed with ALL or AML who completed the T1 assessment and the 6-month follow-up phone interview (T2). Of the 110 families who met these criteria, the sample was limited to the 84 families in which a child was enrolled onto an RCT, given the lack of relevance of the construct of random assignment to those not involved in a RCT and because the follow-up interview administered to non-RCT families did not assess understanding of random assignment. Children's ages ranged from 1 to 18 years (mean, 7.0 years; standard deviation [SD], 4.9 years). Fifty-four percent of children were male, and the majority had ALL (n = 75; 87%). Participating parents were primarily white (59.3%) and female (n = 51), with a mean T1 age of 34.4 years (SD, 7.7 years). On average, participating parents had completed some post–high school training. Mean family SES fell below average on the Hollingshead Index of Social Position¹⁴ (ISP; mean ISP, 3.3; SD, 1.2).

Measures
Demographic predictors. Parents provided information on their ethnicity (coded as majority or minority), sex, education level (coded on a 7-point scale ranging from 1 [grade school] to 7 [postgraduate]), and occupational status via interview. The Hollingshead ISP, a 5-point rating of SES, was computed based on parental education level and occupation. ISP scores of 1 to 2 represent high SES, whereas scores of 3 to 5 denote low SES.^2,14

Parent-focused predictors. At T1, parents reported whether they read the informed consent document during the ICC. In addition, a parent question-asking index was computed by summing the number of questions asked during the ICC and dividing by the ICC's length to obtain the number of questions asked per minute.

Physician-focused predictors. Physician-related information was obtained from a physician-completed questionnaire and through information from the Observer Checklist. Physicians reported their number of years of experience (less [1 to 10 years] v more experienced [11 years or more]), training status (attending v fellow), and sex. Additionally, information about physician discussion of the following content areas during the ICC was recorded: (1) the different arms of the RCT, (2) how the RCT differs from standard treatment, (3) the right to withdraw from the RCT at any time, (4) that one arm of the RCT is standard treatment, and (5) that random assignment is used to assign the child to treatment group.

Relational predictors. Measures of the family-physician relationship included physician partnership building, physician rapport building, and family-physician communicative activity. Partnership (ie, statements encouraging family participation and understanding) and rapport (ie, statements of empathy or reassurance directed toward the family) building were coded from ICC transcripts using the Roter Interaction Analysis System,^15,16 which is a validated system for clinician-patient communication during medical appointments. Frequency ratings were summed to obtain total partnership and rapport building scores for each participant. Family-physician communicative activity was operationalized as the ratio of family member words spoken to physician words spoken during the ICC. Transcripts from the ICC were analyzed to obtain this ratio, with higher ratios indicating greater parent participation.

Contextual predictors. The Observer Checklist included an assessment of nurse presence during the ICC.

Outcome: Understanding of random assignment. Parental understanding of random assignment was evaluated via review of T1 and T2 transcripts. Questions that directly assessed parental understanding were evaluated first (eg, In this clinical research study, how will it be decided which treatment your child will receive? If your child is enrolled onto this clinical research study, will you be able to choose the treatment option you want?). When responses to these items did not yield evidence of understanding, a global search of all responses was performed. In all cases, parents were coded as understanding random assignment if they satisfied the following two-part definitional criterion: they described that their child would be assigned to the treatment group in a random fashion and that the RCT consisted of multiple treatment arms. Ratings of understanding of random assignment were dichotomous in nature, such that parents earned 1 point for articulating both of the above criteria and 0 points for articulating none or one of the criteria. Coding was performed by research assistants who evidenced 92% inter-rater agreement.

Analytic Plan
To describe changes in parental understanding of random assignment across the 6-month follow-up period, descriptive analyses were conducted to determine percentages of parents who fell into one of the following four mutually exclusive groups: (1) understood random assignment at T1 and T2, (2) did not understand random assignment at T1 or T2, (3) understood random assignment at T1 but not at T2, and (4) did not understand random assignment at T1 but understood at T2.

To identify factors associated with stability in understanding over the follow-up period, ² analyses and one-way analyses of variance or t tests were used with dichotomous and continuous outcomes, respectively. Analyses examined differences between stability in understanding (ie, participants who understood at T1 and T2), stability in lack of understanding (ie, participants who did not understand at T1 or T2), and deterioration in understanding (ie, participants who understood at T1 but not at T2). Given the small number of participants whose understanding increased from T1 to T2 (n = 2), this subgroup was excluded from all subsequent analyses. Effect size (ES) estimates were computed in accord with Cohen¹⁷ and are interpreted as an index of clinical significance. ES estimates were included because of the low power to detect statistically significant differences in analyses with small sample sizes.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
Preliminary Analyses
Comparison of the 84 participating families with participants who were excluded because of lack of RCT enrollment (n = 26) or because of missing data at T2 (n = 30; reasons for missing data included family not reachable via telephone, n = 19; patient died, n = 9; patient no longer in the care of T1 caregiver, n = 1; and parent declined participation at T2, n = 1) suggested no differences in parent race, sex, and age; family SES; or child sex and age. A significant between-group difference was documented for understanding of random assignment at T1 (² = 9.49; P < .01) such that the majority (approximately two thirds) of individuals off study and the majority (approximately two thirds) with missing T2 data did not understand random assignment at T1, whereas approximately half of parents included in the present investigation did not understand random assignment at T1. Thus, our sample consisted of more individuals who understood random assignment at T1 and, as such, may overestimate the proportion of decay in understanding from T1 to T2.

Changes in Understanding of Random Assignment
Analyses revealed stability in understanding for the majority of participants, whereas changes in parental understanding were documented for a sizable minority. Specifically, 32.1% of participants (n = 27) understood random assignment at T1 and T2, whereas 48.8% (n = 41) understood at neither T1 nor T2. In contrast, 19.1% (n = 16) of parents evidenced change in understanding of random assignment over time, with 16.7% (n = 14) of the total sample showing knowledge decay from T1 to T2 (ie, understood random assignment at T1 but not at T2) and 2.4% (n = 2) of the total sample showing an increase in understanding from T1 to T2 (ie, they did not understand at T1 but understood at T2). See Figure 1 for examples of common correct and incorrect explanations of random assignment (Drotar¹⁸ provides qualitative information about the content of parental verbalizations during the ICC).

View larger version (25K): [in this window] [in a new window]   Fig 1. Examples of random assignment explanations. AML, acute myeloid leukemia.

Factors Associated With Decay in Understanding From T1 to T2
To ascertain factors associated with decay in knowledge, analyses compared participants who sustained understanding over time (ie, participants who understood random assignment at T1 and T2) with participants who deteriorated in understanding (ie, participants who understood random assignment at T1 but not at T2; Tables 3 and 4). Given the small sample size (n = 42), findings did not achieve statistical significance. However, clinically meaningful factors (as evidenced by medium to large ES estimates) associated with deterioration in understanding of random assignment included parent sex (h = 0.50), family SES (h = 0.55), and physician partnership building (d = 0.70; P < .05). Male parents, high SES parents, and parents whose physicians engaged in more partnership building during the ICC were less likely to fall into the decay group.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

Our results suggest that ethnic minority parents, lower SES parents, and parents who did not read the consent document were over-represented among parents who did not understand random assignment at either time point, as were parents whose ICC failed to include a discussion of the multiple RCT treatment arms or a discussion that one arm of the RCT is standard treatment. In contrast, parents who asked more questions during the ICC were more likely to understand random assignment at both time points, as were parents who had a nurse present during the ICC. Regarding decay in understanding, current findings revealed that loss of understanding from T1 to T2 was more common among female parents, parents of low SES backgrounds, and parents with lower levels of physician partnership building during the ICC.

Attention to factors that predict understanding at diagnosis² as well as sustained understanding may guide intervention. This investigation revealed several relevant factors including majority ethnicity, high SES, parental reading of consent document, discussion of the multiple RCT arms, discussion that one arm of RCT is standard treatment, and nurse presence during the ICC. Interventions that reinforce the importance of reading the consent document and encourage parent question asking may be useful in promoting a collaborative parent-physician relationship, thereby enhancing parental understanding. In addition, interventions that encourage physician partnership building and discussion of key RCT components with parents during the ICC (eg, the multiple arms) may be of value, particularly among ethnic minority and lower SES families. Although nurse presence during the ICC related to improved understanding in the short term and over time, the mechanism by which they have an effect is unknown. The emotional support a nurse provides may create an environment in which parents feel more comfortable asking questions or seeking clarification from physicians. Greater question asking, may, in turn, predict understanding.² Our results suggest that parents who understood random assignment at T1 and T2 asked more questions during the ICC than parents who failed to understand at either point. However, the role of nurses in facilitating understanding of specific RCT components over the course of treatment has not been studied. Our data suggest that 66.7% (n = 56) of parents believed their nurse contributed to improved understanding of RCT aspects between diagnosis and the 6-month follow-up, although the mechanism by which nurses exert an influence is unclear.

Although the current findings extend our knowledge of changes in understanding of random assignment over time, several limitations should be acknowledged. First, current findings are generalizable only to parents of children involved in leukemia RCTs within major urban academic medical centers. These results may not generalize to other populations or settings. Second, given our sample size, conclusions about site differences in physician approaches to explaining random assignment during the ICC cannot be clearly ascertained. Third, findings provide information about only one aspect of informed consent, which is random assignment. Finally, the use of two data points provides only a preliminary longitudinal look at informed consent. Future research should examine changes in understanding over the entire treatment course by including multiple time points. Moreover, future research is recommended to extend our findings to other domains of informed consent understanding. Future work should also investigate whether subsequent physician-family communication about the RCT helps sustain understanding over time, as well as whether or not physician approaches to explaining random assignment vary by site or geographic region. Finally, attention to modifiable barriers to understanding (eg, parental reading of consent document, parent question asking, and physician education about key RCT components) is particularly important to inform intervention. Future interventions should focus on modifying barriers through work with both parents and health professionals. Increased health professional awareness of the subgroups of individuals at risk for lower levels of understanding (eg, those from low SES or ethnic minority backgrounds) is important because multiple discussions about the RCT throughout treatment may be necessary with these subgroups to enhance or maintain understanding over time.

	Authors' Disclosures of Potential Conflicts of Interest

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
The authors indicated no potential conflicts of interest.

	Author Contributions

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 

Conception and design: Rachel Neff Greenley, Dennis Drotar, Eric Kodish Provision of study materials or patients: Eric Kodish Data analysis and interpretation: Rachel Neff Greenley, Stephen J. Zyzanski Manuscript writing: Rachel Neff Greenley Final approval of manuscript: Rachel Neff Greenley, Dennis Drotar, Stephen J. Zyzanski, Eric Kodish

 

	Acknowledgment

 
We thank Michelle Eder for assistance with transcript coding and feedback on the various drafts of the article, Fatoumata Traore for assistance with data analysis and feedback on various drafts of the article, and Amy Yamokoski for feedback on various drafts of the article.

	NOTES

 
Supported by the National Cancer Institute Grant No. RO1 CA83267.

Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION Authors' Disclosures of... Author Contributions REFERENCES

 
1. Pui C: Childhood Leukemias. Cambridge, United Kingdom, Cambridge University Press, 1999

2. Kodish E, Eder M, Noll RB, et al: Communication of randomization in childhood leukemia trials. JAMA 291:470-475, 2004[Abstract/Free Full Text]

3. Kupst MJ, Patenaude AF, Walco GA, et al: Clinical trials in pediatric cancer: Parental perspectives on informed consent. J Pediatr Hematol Oncol 25:787-790, 2003[CrossRef][Medline]

4. Ruccione K, Kramer RF, Moore IK, et al: Informed consent for treatment of childhood cancer: Factors affecting parents' decision making. J Pediatr Oncol Nurs 8:112-121, 1991[Medline]

5. Wiley FM, Ruccione K, Moore IM, et al: Parents' perceptions of randomization in pediatric clinical trials. Cancer Pract 7:248-256, 1999[CrossRef][Medline]

6. Berg JW, Appelbaum PS, Lidz CW, et al: (eds): Informed Consent: Legal Theory and Clinical Practice (ed 2). Oxford, United Kingdom, Oxford University Press, 2001

7. Jensen AB, Madsen B, Andersen P, et al: Information for cancer patients entering a clinical trial: An evaluation of an information strategy. Eur J Cancer 29A:2235-2238, 1993[CrossRef]

8. Tindall B, Forde S, Ross MW, et al: Effects of two formats of informed consent on knowledge amongst persons with advanced HIV disease in a clinical trial of didanosine. Patient Educ Couns 24:261-266, 1994[CrossRef][Medline]

9. Rikkert MGMO, van der Bercken JHL, ten Have HAMJ, et al: Experienced consent in geriatrics research: A new method to optimize the capacity to consent in frail elderly subjects. J Med Ethics 23:271-276, 1997[Abstract/Free Full Text]

10. Weston J, Hannah M, Downes J: Evaluating the benefits of a patient information video during the informed consent process. Patient Educ Couns 30:239-245, 1997[CrossRef][Medline]

11. Eden OB: Consent difficulties in pediatric oncology. BMJ 308:272-275, 1994[Free Full Text]

12. Ferguson PR: Patients' perceptions of information provided in clinical trials. J Med Ethics 28:45-48, 2002[Abstract/Free Full Text]

13. Siminoff LA, Fetting JH: Factors affecting treatment decisions for a life-threatening illness: The case of medical treatment of breast cancer. Soc Sci Med 32:813-818, 1991[CrossRef][Medline]

14. Hollingshead A: Two Factor Index of Social Position. New Haven, CT, Yale University; 1957

15. Roter D: Patient participation in the patient-provider interaction: The effects of patient question asking on the quality of interaction, satisfaction and compliance. Health Educ Monogr 5:281-315, 1977[Medline]

16. Roter D, Hall J, Katz N: Patient-physician communication: A descriptive summary of the literature. Patient Educ Couns 12:99-119, 1988[CrossRef]

17. Cohen J: Statistical Power Analysis for the behavioral Sciences: Revised Edition. New York, NY, Academic Press, 1977

18. Drotar D: Correlates of parental participation during informed consent for randomized clinical trials in the treatment of childhood leukemia. Ethics Behav 14:1-15, 2004[Medline]

Submitted May 25, 2005; accepted November 30, 2005.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This article has been cited by other articles:

				Y. Freer, N. McIntosh, S. Teunisse, K. J. S. Anand, and E. M. Boyle More Information, Less Understanding: A Randomized Study on Consent Issues in Neonatal Research Pediatrics, May 1, 2009; 123(5): 1301 - 1305. [Abstract] [Full Text] [PDF]

				S. B. Johnson, A. E. Baughcum, K. Hood, L. E. Rafkin-Mervis, D. A. Schatz, and for the DPT-1 Study Group Participant and Parent Experiences in the Parenteral Insulin Arm of the Diabetes Prevention Trial for Type 1 Diabetes Diabetes Care, September 1, 2007; 30(9): 2193 - 2198. [Abstract] [Full Text] [PDF]

				P. A. Marshall, C. A. Adebamowo, A. A. Adeyemo, T. O. Ogundiran, M. Vekich, T. Strenski, J. Zhou, T. E. Prewitt, R. S. Cooper, and C. N. Rotimi Voluntary Participation and Informed Consent to International Genetic Research Am J Public Health, November 1, 2006; 96(11): 1989 - 1995. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Greenley, R. N. Articles by Kodish, E. Search for Related Content PubMed PubMed Citation Articles by Greenley, R. N. Articles by Kodish, E. Social Bookmarking                            What's this?