Research Letter
CLINICAL TRIALS
How well informed is the informed consent for cancer clinical trials?
Clinical Trials 2014, Vol. 11(6) 686–688 Ó The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1740774514548734 ctj.sagepub.com
Laeeq Malik1, James Kuo2, Desmond Yip2,3 and Alex Mejia1
Abstract Aims: The purpose of this study was to analyze the content of informed consent forms for clinical trials in medical oncology to assess readability, determine their completeness, and identify any shortcomings. Methods: Informed consent forms for Phase I–III studies that were conducted at two tertiary care cancer centers over a 3-year period were reviewed. Information pertaining to length of the informed consent form, research regimen/methods, treatment agent, potential risks, and benefits was extracted. The reading level was assessed by Flesch–Kincaid and Gunning-Fog index readability tests. Results: All of the 112 informed consent forms clearly stated the voluntary nature of participation. Nearly one half of the forms (51.8%) were of Phase I studies. The median length of informed consent form was 20 pages (range: 8–28). A detailed estimation of the frequency or intensity of risks (range: 3–8 pages) was provided. The average reading level of the informed consent forms was high (Flesch–Kincaid Grade Level of 9.8), which corresponds roughly to 10th-grade reading level. Less than 15% of all consent forms were written at the recommended eighth-grade reading level. A substantial number of forms did not report a potential risk to pregnant/lactating women (16.9%), mechanism of action of the investigational agent (34.8%), study schema (77.6%), a possibility of receiving sub-therapeutic dose (37%), or death (12.5%). Nearly one half of the forms (49.1%) stated clearly that individual participants may not benefit. Conclusion: Overall, these informed consent forms provided a detailed description of the trials in accordance to international guidelines. However, there remains room for improvement, particularly in areas of readability and document length. Keywords Clinical trial, consent, informed consent form
Clinical trials pave the way for the future development of novel anticancer agents and advances of the science. Written informed consent is an essential element and a legal requirement in the conduct of a clinical trial. The informed consent forms (ICFs) are written in accordance with Declaration of Helsinki and International Conference on Harmonization guidelines for Good Clinical Practice.1 It is recommended that the reading level of the informed consent document should be no higher than an eighth-grade level, and the page count not exceeding 6–9 pages.2 As the number of early phase clinical trials has increased steadily, an insight into patients’ understanding of these clinical trials is becoming increasingly relevant. Although the likelihood of direct benefit from clinical trials (particularly Phase I) is low, unrealistic optimism and therapeutic misconception are unfortunately common among cancer patients with advanced disease.3 It is unclear whether some of these misunderstandings stem from the ICFs.4 In order
to ensure that the basic ethical principle of respect for persons and the goals of informed consent are not undermined, it is critically important to determine whether ICFs contribute to any gap in study participants’ understanding of the clinical trials. The purpose of this study was to analyze the content of ICFs for clinical trials in medical oncology, to assess
1
Institute for Drug Development, Cancer Therapy & Research Center, The University of Texas Health Science Center, San Antonio, TX, USA 2 Department of Medical Oncology, The Canberra Hospital, Garran, ACT, Australia 3 ANU Medical School, Australian National University, Acton, ACT, Australia Corresponding author: Laeeq Malik, Institute for Drug Development, Cancer Therapy & Research Center, The University of Texas Health Science Center, San Antonio, TX 78229, USA. Email: [email protected]
Downloaded from ctj.sagepub.com at SUNY BINGHAMTON on March 8, 2015
Malik et al.
687
readability, to determine their completeness, and to identify any shortcomings. We reviewed ICFs of consecutive Phase I–III studies for metastatic cancer conducted between January 2011 and December 2013 at The University of Texas Health Science Center at San Antonio, Texas, and The Canberra Hospital in Australia. All the ICFs were previously approved by the local Institutional Review Board and Human Research Ethics Committee. Investigators at both sites extracted information pertaining to length of the document, research methods, treatment agent, potential risks, and benefits. The ICFs were analyzed for their respective level of readability by means of the Flesch– Kincaid Reading Ease, Flesch–Kincaid Grade Level, and Gunning-Fog index.5,6 The Flesch–Kincaid Reading Ease test evaluates comprehension difficulty of an English passage by using word length and sentence length on a 100-point scale. Higher reading scores reflect easier text, and lower scores indicate that the passage is difficult to read. Flesch–Kincaid Grade Level translates the difficulty score to United States grade level. It indicates an approximate number of years of education required to read a particular text. The Gunning-Fog index assesses text readability based upon difficult words and the sentence length. These formulas and related details are provided in the supplementary appendix. The readability tests were performed by a widely used online readability calculator.7 Data analysis was performed using Microsoft Excel 2003. Descriptive statistical analysis was conducted to summarize the characteristics of the ICFs. Of the 156 ICFs that were screened, 112 ICFs of therapeutic trials for metastatic disease were included in this analysis. The ICFs of non-therapeutic (n = 16), adjuvant (n = 18), neoadjuvant (n = 8), and radiation oncology (n = 2) trials were excluded. The baseline characteristics of the included trials are presented in Table 1. The median length of ICF was 20 pages (range: 8–28). A detailed estimation of the frequency or intensity of risks (range: 3–8 pages) was provided. The complex study designs and dose escalation strategies were explained over several pages. The average reading level of the ICFs was high (Flesch–Kincaid Grade Level was calculated to be 9.8), which corresponds roughly to a 10th-grade reading level. The Flesch– Kincaid Reading Ease and Gunning-Fog index scores (median of 55.5 and 11.2, respectively) indicate that the ICFs text was fairly difficult for most people to read. Only less than 15% of all consent forms were written at an eighth-grade reading level or below. All of the ICFs clearly stated the voluntary nature of participation. A substantial number of forms did not report a potential risk to pregnant/lactating women (16.9%), mechanism of action of the investigational agent (34.8%), study schema (77.6%), a possibility of receiving subtherapeutic dose (37%), or death (12.5%). Nearly one half of the forms, primarily Phase I (49.1%), stated
Table 1. Baseline characteristics of the included clinical trials. Trial characteristics
Number (%)
All trials Phase I Phase II Phase III Others Research funding Investigator initiated Cooperative group Trials sponsored by pharmaceutical industry Regulatory status of the investigational agent FDA/TGA approved Non FDA/TGA approved Nature of investigational agent Single biological agent Combination of biological agents Combination of chemotherapy and biological agent Chemotherapy agent Others Description of Phase I study purpose Dose finding Safety monitoring Others Description of Phase II study purpose Safety monitoring Efficacy evaluation Description of Phase III study purpose Safety monitoring Efficacy evaluation
112 57 (51.8) 33 (29.4) 19 (16.9) 3 (1.9) 6 (5.3) 14 (12.5) 92 (82.1) 34 (30.3) 78 (69.6) 64 (57.1) 16 (14.2) 19 (16.9) 7 (6.2) 6 (5.3) 22 (38.0) 33 (57.8) 2 (4.2) 12 (36.3) 21 (63.6) 4 (21) 15 (78.9)
FDA: Food and Drug Administration; TGA: Therapeutic Goods Administration (Australian equivalent to FDA).
clearly that individual participants may not benefit. The main goal of research was explicitly stated as safety testing, dose determination, or efficacy evaluation. A detailed analysis of all the recommended ICF variables is provided in the supplementary online appendix. Our findings of poor readability of the ICFs are in agreement with previously published reviews from other institutions.8,9 This is worrisome considering about 19% of adults in the state of Texas are classified as having low literacy skills.10 Similarly, Australian literacy surveys estimate that 46% of adults have the lowest measured level of literacy.11 This may undermine the doctrine of informed consent and patient autonomy. Simplification of the ICF document from conceptual, ethical, and methodological standpoints is required. Involvement of a patient advocate in the development of ICFs could help in simplifying the forms and selection of information that participants find necessary to know. Collaboration with cognitive and social psychologists to better design risk communication and presentation strategies may also be beneficial. The length of the consent document poses a challenge for cancer patients in finding out necessary information and may well exceed patient’s memory capacity. The National Cancer Institute has developed a template to simplify
Downloaded from ctj.sagepub.com at SUNY BINGHAMTON on March 8, 2015
688
Clinical Trials 11(6)
ICF so as to enhance the research participant’s understanding of the consent form.2 In a recent randomized controlled study, a short consent form (of only 5 pages) compared favorably with the lengthier standard form in all outcome measures, including patients’ understanding, levels of recall, concerns, trust, and voluntariness.12 A short consent form containing all the necessary information on the proposed relationship between the patient and the clinical research may better assist patients in making a deliberate choice to undergo or forego the investigational treatment.
Conclusion Within limitations of the bi-institutional nature of this study, retrospective methodology, and some inherent weaknesses in using these readability formulas, our work should be seen as critical avenues for future research. Further efforts are needed to reduce the length of main document, simplify the language, and incorporate the missing information. Declaration of conflicting interests None.
Funding None.
References 1. Karlberg J. Reviewing clinical trials: a guide for the Ethics Committee: Pfizer, http://www.pfizer.com/files/ research/research_clinical_trials/ethics_committee_guide.pdf 2. Comprehensive Working Group on Informed Consent in Cancer Clinical Trials, http://www.cancer.gov/clinicaltrials/conducting/simplification-of-informed-consent-docs/ page5#appendix6
3. Horng S and Grady C. Misunderstanding in clinical research: distinguishing therapeutic misconception, therapeutic misestimation, and therapeutic optimism. IRB 2003; 25: 11–16. 4. Daugherty CK. Impact of therapeutic research on informed consent and the ethics of clinical trials: a medical oncology perspective. J Clin Oncol 1999; 17: 1601–1617. 5. Kincaid JP, Fishburne RP, Rogers RL, et al. Derivation of new readability formulas (Automated Readability Index, Fog Count, and Flesch Reading Ease Formula) for Navy enlisted personnel. Research Branch Report no. 8-75, February 1975. Memphis, TN: Naval Air Station. 6. Flesch R. A new readability yardstick. J Appl Psychol 1948; 32: 221–233. 7. Readability calculator, http://www.readabilityformulas. com/free-readability-formula-tests.php (accessed March 2014). 8. Beardsley E, Jefford M and Mileshkin L. Longer consent forms for clinical trials compromise patient understanding: so why are they lengthening? J Clin Oncol 2007; 25: e13–e14. 9. Paasche-Orlow MK, Taylor HA and Brancati FL. Readability standards for informed-consent forms as compared with actual readability. N Engl J Med 2003; 348: 721–726. 10. National Center for Education Statistics, https://nces.ed. gov/naal/estimates/StateEstimates.aspx (2003, accessed July 2014). 11. Adult literacy and life skills survey, summary results, Australia, 2006, http://www.abs.gov.au/AUSSTATS/ [email protected]/Previousproducts/4228.0Main%20Features22006% 20(Reissue)?opendocument&tabname=Summary&prodno= 4228.0&issue=2006%20(Reissue)&num=&view= (accessed July 2014). 12. Matsui K, Lie RK, Turin TC, et al. A randomized controlled trial of short and standard-length consent forms for a genetic cohort study: is longer better? J Epidemiol 2012; 22: 308–316.
Downloaded from ctj.sagepub.com at SUNY BINGHAMTON on March 8, 2015
CLINICAL TRIALS
How well informed is the informed consent for cancer clinical trials?
Clinical Trials 2014, Vol. 11(6) 686–688 Ó The Author(s) 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1740774514548734 ctj.sagepub.com
Laeeq Malik1, James Kuo2, Desmond Yip2,3 and Alex Mejia1
Abstract Aims: The purpose of this study was to analyze the content of informed consent forms for clinical trials in medical oncology to assess readability, determine their completeness, and identify any shortcomings. Methods: Informed consent forms for Phase I–III studies that were conducted at two tertiary care cancer centers over a 3-year period were reviewed. Information pertaining to length of the informed consent form, research regimen/methods, treatment agent, potential risks, and benefits was extracted. The reading level was assessed by Flesch–Kincaid and Gunning-Fog index readability tests. Results: All of the 112 informed consent forms clearly stated the voluntary nature of participation. Nearly one half of the forms (51.8%) were of Phase I studies. The median length of informed consent form was 20 pages (range: 8–28). A detailed estimation of the frequency or intensity of risks (range: 3–8 pages) was provided. The average reading level of the informed consent forms was high (Flesch–Kincaid Grade Level of 9.8), which corresponds roughly to 10th-grade reading level. Less than 15% of all consent forms were written at the recommended eighth-grade reading level. A substantial number of forms did not report a potential risk to pregnant/lactating women (16.9%), mechanism of action of the investigational agent (34.8%), study schema (77.6%), a possibility of receiving sub-therapeutic dose (37%), or death (12.5%). Nearly one half of the forms (49.1%) stated clearly that individual participants may not benefit. Conclusion: Overall, these informed consent forms provided a detailed description of the trials in accordance to international guidelines. However, there remains room for improvement, particularly in areas of readability and document length. Keywords Clinical trial, consent, informed consent form
Clinical trials pave the way for the future development of novel anticancer agents and advances of the science. Written informed consent is an essential element and a legal requirement in the conduct of a clinical trial. The informed consent forms (ICFs) are written in accordance with Declaration of Helsinki and International Conference on Harmonization guidelines for Good Clinical Practice.1 It is recommended that the reading level of the informed consent document should be no higher than an eighth-grade level, and the page count not exceeding 6–9 pages.2 As the number of early phase clinical trials has increased steadily, an insight into patients’ understanding of these clinical trials is becoming increasingly relevant. Although the likelihood of direct benefit from clinical trials (particularly Phase I) is low, unrealistic optimism and therapeutic misconception are unfortunately common among cancer patients with advanced disease.3 It is unclear whether some of these misunderstandings stem from the ICFs.4 In order
to ensure that the basic ethical principle of respect for persons and the goals of informed consent are not undermined, it is critically important to determine whether ICFs contribute to any gap in study participants’ understanding of the clinical trials. The purpose of this study was to analyze the content of ICFs for clinical trials in medical oncology, to assess
1
Institute for Drug Development, Cancer Therapy & Research Center, The University of Texas Health Science Center, San Antonio, TX, USA 2 Department of Medical Oncology, The Canberra Hospital, Garran, ACT, Australia 3 ANU Medical School, Australian National University, Acton, ACT, Australia Corresponding author: Laeeq Malik, Institute for Drug Development, Cancer Therapy & Research Center, The University of Texas Health Science Center, San Antonio, TX 78229, USA. Email: [email protected]
Downloaded from ctj.sagepub.com at SUNY BINGHAMTON on March 8, 2015
Malik et al.
687
readability, to determine their completeness, and to identify any shortcomings. We reviewed ICFs of consecutive Phase I–III studies for metastatic cancer conducted between January 2011 and December 2013 at The University of Texas Health Science Center at San Antonio, Texas, and The Canberra Hospital in Australia. All the ICFs were previously approved by the local Institutional Review Board and Human Research Ethics Committee. Investigators at both sites extracted information pertaining to length of the document, research methods, treatment agent, potential risks, and benefits. The ICFs were analyzed for their respective level of readability by means of the Flesch– Kincaid Reading Ease, Flesch–Kincaid Grade Level, and Gunning-Fog index.5,6 The Flesch–Kincaid Reading Ease test evaluates comprehension difficulty of an English passage by using word length and sentence length on a 100-point scale. Higher reading scores reflect easier text, and lower scores indicate that the passage is difficult to read. Flesch–Kincaid Grade Level translates the difficulty score to United States grade level. It indicates an approximate number of years of education required to read a particular text. The Gunning-Fog index assesses text readability based upon difficult words and the sentence length. These formulas and related details are provided in the supplementary appendix. The readability tests were performed by a widely used online readability calculator.7 Data analysis was performed using Microsoft Excel 2003. Descriptive statistical analysis was conducted to summarize the characteristics of the ICFs. Of the 156 ICFs that were screened, 112 ICFs of therapeutic trials for metastatic disease were included in this analysis. The ICFs of non-therapeutic (n = 16), adjuvant (n = 18), neoadjuvant (n = 8), and radiation oncology (n = 2) trials were excluded. The baseline characteristics of the included trials are presented in Table 1. The median length of ICF was 20 pages (range: 8–28). A detailed estimation of the frequency or intensity of risks (range: 3–8 pages) was provided. The complex study designs and dose escalation strategies were explained over several pages. The average reading level of the ICFs was high (Flesch–Kincaid Grade Level was calculated to be 9.8), which corresponds roughly to a 10th-grade reading level. The Flesch– Kincaid Reading Ease and Gunning-Fog index scores (median of 55.5 and 11.2, respectively) indicate that the ICFs text was fairly difficult for most people to read. Only less than 15% of all consent forms were written at an eighth-grade reading level or below. All of the ICFs clearly stated the voluntary nature of participation. A substantial number of forms did not report a potential risk to pregnant/lactating women (16.9%), mechanism of action of the investigational agent (34.8%), study schema (77.6%), a possibility of receiving subtherapeutic dose (37%), or death (12.5%). Nearly one half of the forms, primarily Phase I (49.1%), stated
Table 1. Baseline characteristics of the included clinical trials. Trial characteristics
Number (%)
All trials Phase I Phase II Phase III Others Research funding Investigator initiated Cooperative group Trials sponsored by pharmaceutical industry Regulatory status of the investigational agent FDA/TGA approved Non FDA/TGA approved Nature of investigational agent Single biological agent Combination of biological agents Combination of chemotherapy and biological agent Chemotherapy agent Others Description of Phase I study purpose Dose finding Safety monitoring Others Description of Phase II study purpose Safety monitoring Efficacy evaluation Description of Phase III study purpose Safety monitoring Efficacy evaluation
112 57 (51.8) 33 (29.4) 19 (16.9) 3 (1.9) 6 (5.3) 14 (12.5) 92 (82.1) 34 (30.3) 78 (69.6) 64 (57.1) 16 (14.2) 19 (16.9) 7 (6.2) 6 (5.3) 22 (38.0) 33 (57.8) 2 (4.2) 12 (36.3) 21 (63.6) 4 (21) 15 (78.9)
FDA: Food and Drug Administration; TGA: Therapeutic Goods Administration (Australian equivalent to FDA).
clearly that individual participants may not benefit. The main goal of research was explicitly stated as safety testing, dose determination, or efficacy evaluation. A detailed analysis of all the recommended ICF variables is provided in the supplementary online appendix. Our findings of poor readability of the ICFs are in agreement with previously published reviews from other institutions.8,9 This is worrisome considering about 19% of adults in the state of Texas are classified as having low literacy skills.10 Similarly, Australian literacy surveys estimate that 46% of adults have the lowest measured level of literacy.11 This may undermine the doctrine of informed consent and patient autonomy. Simplification of the ICF document from conceptual, ethical, and methodological standpoints is required. Involvement of a patient advocate in the development of ICFs could help in simplifying the forms and selection of information that participants find necessary to know. Collaboration with cognitive and social psychologists to better design risk communication and presentation strategies may also be beneficial. The length of the consent document poses a challenge for cancer patients in finding out necessary information and may well exceed patient’s memory capacity. The National Cancer Institute has developed a template to simplify
Downloaded from ctj.sagepub.com at SUNY BINGHAMTON on March 8, 2015
688
Clinical Trials 11(6)
ICF so as to enhance the research participant’s understanding of the consent form.2 In a recent randomized controlled study, a short consent form (of only 5 pages) compared favorably with the lengthier standard form in all outcome measures, including patients’ understanding, levels of recall, concerns, trust, and voluntariness.12 A short consent form containing all the necessary information on the proposed relationship between the patient and the clinical research may better assist patients in making a deliberate choice to undergo or forego the investigational treatment.
Conclusion Within limitations of the bi-institutional nature of this study, retrospective methodology, and some inherent weaknesses in using these readability formulas, our work should be seen as critical avenues for future research. Further efforts are needed to reduce the length of main document, simplify the language, and incorporate the missing information. Declaration of conflicting interests None.
Funding None.
References 1. Karlberg J. Reviewing clinical trials: a guide for the Ethics Committee: Pfizer, http://www.pfizer.com/files/ research/research_clinical_trials/ethics_committee_guide.pdf 2. Comprehensive Working Group on Informed Consent in Cancer Clinical Trials, http://www.cancer.gov/clinicaltrials/conducting/simplification-of-informed-consent-docs/ page5#appendix6
3. Horng S and Grady C. Misunderstanding in clinical research: distinguishing therapeutic misconception, therapeutic misestimation, and therapeutic optimism. IRB 2003; 25: 11–16. 4. Daugherty CK. Impact of therapeutic research on informed consent and the ethics of clinical trials: a medical oncology perspective. J Clin Oncol 1999; 17: 1601–1617. 5. Kincaid JP, Fishburne RP, Rogers RL, et al. Derivation of new readability formulas (Automated Readability Index, Fog Count, and Flesch Reading Ease Formula) for Navy enlisted personnel. Research Branch Report no. 8-75, February 1975. Memphis, TN: Naval Air Station. 6. Flesch R. A new readability yardstick. J Appl Psychol 1948; 32: 221–233. 7. Readability calculator, http://www.readabilityformulas. com/free-readability-formula-tests.php (accessed March 2014). 8. Beardsley E, Jefford M and Mileshkin L. Longer consent forms for clinical trials compromise patient understanding: so why are they lengthening? J Clin Oncol 2007; 25: e13–e14. 9. Paasche-Orlow MK, Taylor HA and Brancati FL. Readability standards for informed-consent forms as compared with actual readability. N Engl J Med 2003; 348: 721–726. 10. National Center for Education Statistics, https://nces.ed. gov/naal/estimates/StateEstimates.aspx (2003, accessed July 2014). 11. Adult literacy and life skills survey, summary results, Australia, 2006, http://www.abs.gov.au/AUSSTATS/ [email protected]/Previousproducts/4228.0Main%20Features22006% 20(Reissue)?opendocument&tabname=Summary&prodno= 4228.0&issue=2006%20(Reissue)&num=&view= (accessed July 2014). 12. Matsui K, Lie RK, Turin TC, et al. A randomized controlled trial of short and standard-length consent forms for a genetic cohort study: is longer better? J Epidemiol 2012; 22: 308–316.
Downloaded from ctj.sagepub.com at SUNY BINGHAMTON on March 8, 2015
