- Email: [email protected]
Anemic Outcome Reporting Consistency: Evidence of Selective Reporting Bias in Leading Hematology Journals
Abstracts Clinical evaluation of these patients through genetic counseling and testing is high yield for identified at-risk families. Research based sequencing for novel mutations is indicated and ongoing. Grant acknowledgements: This work was supported in part by the MD Anderson Cancer Center Support Grant (CCSG) CA016672 and by the generous philanthropic contributions to MD Anderson’s MDS/AML Moon Shot Program. CDD is also supported by the Jeanne F. Shelby Scholarship Fund which has supported her R. Lee Clark Fellow award.
Anemic Outcome Reporting Consistency: Evidence of Selective Reporting Bias in Leading Hematology Journals 2
RES-191 Determining Treatment for Older Patients with High-Risk MDS and AML: which Factors Predict Optimal Quality of Life?
RES-184
1
result. Conclusion: We observed a high rate of discordance between pre-specified and published outcomes in high impact-factor hematology journals, suggesting need for continued improvement. Among outcomes which reported p-values, many outcome discrepancies favored publication of a statistically significant results, further suggesting selective outcome reporting bias.
Sara Tinsley ,1 Brent Small,2 Susan McMillan,2 Rami Komrokji,1 Jeffrey Lancet1 1
Moffitt Cancer Center, Dover, FL, United States; 2University of South
1
Cole Wayant, Caleb Scheckel, Chandler Hicks, Tim Nissen,1 Linda Leduc,1 Mousumi Som,1 Matt Vassar 1 1
Oklahoma State University Center for Health Sciences, Tulsa, United
States; 2Mayo Clinic, Scottsdale, United States
Context: Selective outcome reporting is a significant methodological concern leading to potential bias in clinical trial results. Comparisons between outcomes reported in trial registries and journal publications allow investigators to examine the frequency of selective reporting occurrences among trialists. Objective: We examined selective outcome reporting in randomized controlled trials (RCTs) published in high impact-factor hematology journals and whether outcome reporting favored statistically significant results. Design: A PubMed search identified RCTs published in five hematology journals from 2010-2015. After screening, primary and secondary outcomes were recorded for each trial and then compared with outcomes pre-specified in the trial’s registration. We compared whether pre-specified trial registry outcomes were changed in the publications, whether primary outcomes had been downgraded to secondary outcomes, secondary outcomes had been upgraded to primary outcomes, or primary outcomes had been omitted, added, or changed. Primary outcome: The prevalence of major discrepancies between clinical trial registries and publications. Results: A total of 93% (143/154) of included RCTs were registered prior to the completion of the trial and further analyzed. From these, 46.8% (72/154) were flagged as being discrepant due to the primary outcome not being defined in the registry or outcomes being specified in the registry but then unspecified in the publication. Of the remaining 82, 43.9% (36/82) demonstrated major outcome discrepancies between registry and publication. 46.3% (31/67) of included RCTs published in Blood, 43.9% (3/7) in ATVB, 20% (1/5) in Circulation Research, 0% (0/2) in Leukemia, and 100% (1/ 1) in Stem Cells included at least one major discrepancy. Only 11% (9/82) were found to be free of both major and non-major discrepancies. A total of 20 major discrepancies from 14 RCTs reported p-values and were evaluated for selective outcome reporting. 85% (17/20) of these major discrepancies favored a statistically significant
Florida, Tampa, FL, United States
Context: Evidence is lacking to guide patients diagnosed with acute myeloid leukemia and high-risk myelodysplastic syndrome in selecting the best treatment for improving quality of life. Objective: The aim of this study was to determine whether age, comorbidity, or fatigue had an impact on quality of life at two different intensities of therapy; intense or non-intense. Design: The study design was an exploratory, longitudinal design that compared quality of life between two treatment groups, intensive and non-intensive prior to start of new treatment and again one month post treatment. Enrollment of participants began in 12/2013, and ended 3/2015. Setting: The setting was the Moffitt Cancer Center, a National Cancer Institute designated comprehensive cancer center in Tampa, Florida. Patients or Other Participants: Recruitment of 85 patients with pathology confirmed high risk myelodysplastic syndrome and acute myeloid leukemia, 60 years of age and older were approached for participation. The sample was predominantly white, male, retired, and middle class. Main outcomes measures: The outcome of this study was quality of life, measured using the Functional Assessment of Cancer Therapy-Leukemia version measured twice with the second measurement used in the regression analysis. Comorbidities were calculated with the Charlson comorbidity index. Fatigue was measured with the Brief Fatigue Inventory. Age was measured in years. Results: Linear regression was performed to determine the moderating effect of treatment on age, comorbidity, and fatigue on changes in QOL scores. Type of treatment was significant (p¼.043) whereby the intensive treatment group had an improvement in QOL across the one-month followup. The main effect of fatigue was significant (p ¼ .014), with quality of life declining as fatigue increased. Conclusions: Intensive chemotherapy was associated with improvement in QOL score at one month for 60 years of age and older patients with high-risk MDS and AML. This study can help inform patients concerning the impact of treatment on QOL at one month, to assist with decision making. A larger sample size with more diverse demographic characteristics would enhance future studies. Supported by an American Cancer Society Doctoral Degree in Nursing Science Scholarship
Clinical Lymphoma, Myeloma & Leukemia September 2016
- S129
Anemic Outcome Reporting Consistency: Evidence of Selective Reporting Bias in Leading Hematology Journals 2
RES-191 Determining Treatment for Older Patients with High-Risk MDS and AML: which Factors Predict Optimal Quality of Life?
RES-184
1
result. Conclusion: We observed a high rate of discordance between pre-specified and published outcomes in high impact-factor hematology journals, suggesting need for continued improvement. Among outcomes which reported p-values, many outcome discrepancies favored publication of a statistically significant results, further suggesting selective outcome reporting bias.
Sara Tinsley ,1 Brent Small,2 Susan McMillan,2 Rami Komrokji,1 Jeffrey Lancet1 1
Moffitt Cancer Center, Dover, FL, United States; 2University of South
1
Cole Wayant, Caleb Scheckel, Chandler Hicks, Tim Nissen,1 Linda Leduc,1 Mousumi Som,1 Matt Vassar 1 1
Oklahoma State University Center for Health Sciences, Tulsa, United
States; 2Mayo Clinic, Scottsdale, United States
Context: Selective outcome reporting is a significant methodological concern leading to potential bias in clinical trial results. Comparisons between outcomes reported in trial registries and journal publications allow investigators to examine the frequency of selective reporting occurrences among trialists. Objective: We examined selective outcome reporting in randomized controlled trials (RCTs) published in high impact-factor hematology journals and whether outcome reporting favored statistically significant results. Design: A PubMed search identified RCTs published in five hematology journals from 2010-2015. After screening, primary and secondary outcomes were recorded for each trial and then compared with outcomes pre-specified in the trial’s registration. We compared whether pre-specified trial registry outcomes were changed in the publications, whether primary outcomes had been downgraded to secondary outcomes, secondary outcomes had been upgraded to primary outcomes, or primary outcomes had been omitted, added, or changed. Primary outcome: The prevalence of major discrepancies between clinical trial registries and publications. Results: A total of 93% (143/154) of included RCTs were registered prior to the completion of the trial and further analyzed. From these, 46.8% (72/154) were flagged as being discrepant due to the primary outcome not being defined in the registry or outcomes being specified in the registry but then unspecified in the publication. Of the remaining 82, 43.9% (36/82) demonstrated major outcome discrepancies between registry and publication. 46.3% (31/67) of included RCTs published in Blood, 43.9% (3/7) in ATVB, 20% (1/5) in Circulation Research, 0% (0/2) in Leukemia, and 100% (1/ 1) in Stem Cells included at least one major discrepancy. Only 11% (9/82) were found to be free of both major and non-major discrepancies. A total of 20 major discrepancies from 14 RCTs reported p-values and were evaluated for selective outcome reporting. 85% (17/20) of these major discrepancies favored a statistically significant
Florida, Tampa, FL, United States
Context: Evidence is lacking to guide patients diagnosed with acute myeloid leukemia and high-risk myelodysplastic syndrome in selecting the best treatment for improving quality of life. Objective: The aim of this study was to determine whether age, comorbidity, or fatigue had an impact on quality of life at two different intensities of therapy; intense or non-intense. Design: The study design was an exploratory, longitudinal design that compared quality of life between two treatment groups, intensive and non-intensive prior to start of new treatment and again one month post treatment. Enrollment of participants began in 12/2013, and ended 3/2015. Setting: The setting was the Moffitt Cancer Center, a National Cancer Institute designated comprehensive cancer center in Tampa, Florida. Patients or Other Participants: Recruitment of 85 patients with pathology confirmed high risk myelodysplastic syndrome and acute myeloid leukemia, 60 years of age and older were approached for participation. The sample was predominantly white, male, retired, and middle class. Main outcomes measures: The outcome of this study was quality of life, measured using the Functional Assessment of Cancer Therapy-Leukemia version measured twice with the second measurement used in the regression analysis. Comorbidities were calculated with the Charlson comorbidity index. Fatigue was measured with the Brief Fatigue Inventory. Age was measured in years. Results: Linear regression was performed to determine the moderating effect of treatment on age, comorbidity, and fatigue on changes in QOL scores. Type of treatment was significant (p¼.043) whereby the intensive treatment group had an improvement in QOL across the one-month followup. The main effect of fatigue was significant (p ¼ .014), with quality of life declining as fatigue increased. Conclusions: Intensive chemotherapy was associated with improvement in QOL score at one month for 60 years of age and older patients with high-risk MDS and AML. This study can help inform patients concerning the impact of treatment on QOL at one month, to assist with decision making. A larger sample size with more diverse demographic characteristics would enhance future studies. Supported by an American Cancer Society Doctoral Degree in Nursing Science Scholarship
Clinical Lymphoma, Myeloma & Leukemia September 2016
- S129