Central site monitoring: Results from a test of accuracy in identifying trials and sites failing Food and Drug Administration inspection

Lindblad, Anne S.; Manukyan, Zorayr; Purohit-Sheth, Tejashri; Okwesili, Paul; Meeker-O’Connell, Ann; Ball, L.; Marler, John R.

doi:10.1177/1740774513508028

Cited by 39 publications

(50 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…16 Consistent with previous authors, 4,5 we caution using unqueried data for statistical assessment. To Statistical monitoring should be performed iteratively throughout a study starting as soon as there is adequate data volume and at a frequency determined by the pace of data collection or aligned with cohort cleaning or DSMB submission schedules.…”

Section: Conclusion and Recommendationssupporting

confidence: 63%

“…3,5 Inliers may be more indicative of fraud, whereas outliers may indicate sloppiness in data collection.…”

Section: Discussionmentioning

confidence: 99%

“…Building on prior research, [3][4][5][6][8][9][10] TransCelerate tested statistical-monitoring methods on a biopharmaceutical industry-sponsored clinical study database under conditions partially mimicking those found during real-world study conduct. The objective of this experiment was to detect data anomalies suggestive of noncompliance, but it primarily focused on intentionally fabricated data because of the potential difficulty in detecting this type of noncompliance.…”

Section: Introductionmentioning

confidence: 99%

“…1,2 One strategy being used is statistical monitoring, or the use of various statistical methods during study conduct to detect data anomalies suggestive of fabrication or noncompliance. 1,[3][4][5] The multicentric nature of most clinical trials offers an opportunity to check the plausibility of data from one site against data from all other sites. 3 Statistical monitoring relies on the highly structured nature of data because each protocol is expected to be implemented consistently at all sites.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Statistical Monitoring in Clinical Trials: Best Practices for Detecting Data Anomalies Suggestive of Fabrication or Misconduct

Knepper¹,

Lindblad

Sharma

et al. 2016

Ther Innov Regul Sci

Self Cite

View full text Add to dashboard Cite

Background: Traditional site-monitoring techniques are not optimal in finding data fabrication and other nonrandom data distributions with the greatest potential for jeopardizing the validity of study results. TransCelerate BioPharma conducted an experiment testing the utility of statistical methods for detecting implanted fabricated data and other signals of noncompliance. Methods: TransCelerate tested statistical monitoring on a data set from a chronic obstructive pulmonary disease (COPD) clinical study with 178 sites and 1554 subjects. Fabricated data were selectively implanted in 7 sites and 43 subjects by expert clinicians in COPD. The data set was partitioned to simulate studies of different sizes. Analyses of vital signs, spirometry, visit dates, and adverse events included distributions of standard deviations, correlations, repeated values, digit preference, and outlier/inlier detection. An interpretation team, including clinicians, statisticians, site monitoring, and data management, reviewed the results and created an algorithm to flag sites for fabricated data. Results: The algorithm identified 11 sites (19%), 19 sites (31%), 28 sites (16%), and 45 sites (25%) as having potentially fabricated data for studies 2A, 2, 1A, and 1, respectively. For study 2A, 3 of 7 sites with fabricated data were detected, 5 of 7 were detected for studies 2 and 1A, and 6 of 7 for study 1. Except for study 2A, the algorithm had good sensitivity and specificity (>70%) for identifying sites with fabricated data. Conclusions: We recommend a crossfunctional, collaborative approach to statistical monitoring that can adapt to study design and data source and use a combination of statistical screening techniques and confirmatory graphics.

show abstract

Section: Conclusion and Recommendationssupporting

confidence: 63%

“…3,5 Inliers may be more indicative of fraud, whereas outliers may indicate sloppiness in data collection.…”

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Statistical Monitoring in Clinical Trials: Best Practices for Detecting Data Anomalies Suggestive of Fabrication or Misconduct

Knepper¹,

Lindblad

Sharma

et al. 2016

Ther Innov Regul Sci

Self Cite

View full text Add to dashboard Cite

show abstract

“…48 Different implementations of CT monitoring have been proposed in the literature. 49,50 The most famous approach depends on "key hazard pointers", which are clinical information factors recognized as critical, and observed all through the trial against pre-determined thresholds. 51 One site that surpasses the limit for a key hazard marker is hailed for promote investigation.…”

Section: ) Use Of Technologymentioning

confidence: 99%

Qualitative features in clinical trials: coordinates for prevention of passive and active misconduct

et al. 2018

View full text Add to dashboard Cite

For many years, the quality concept in clinical trials has been discussed and recommended by Good Clinical Practice (GCP) guidelines. Regulatory Authorities and also the Public Involvement anticipate that the pharmaceutical industry will concentrate on creating quality frameworks amid the arranging and leading of conventions of controlled protocols. Nevertheless, many factors have been suggested as contributing to the occurrence of scientific misconduct within the research field, such as: personal and financial interests, site monitoring, available resources, workload, competition among investigators, and the implicit consent of sponsors. The negligence on data fraud represents not only omission but misconduct as well, in this case, a passive attitude intrinsically related to the act of transgression. A properly culture of research must be based on a fundamental ethos of integrity, openness and honest work of high quality in all parts of the research process. There is a need to change the focus from inspection-based quality improvement to planned systematic quality management within clinical trials. In search for a monitoring improvement, a full statistical way to deal with information recognition comprises of executing however many measurable tests as could be allowed on whatever number clinical information factors as could be expected under varied circumstances. Adoption of specific and preventive clinical trial monitoring procedures can identify potential misconduct and data fraud leading to improvement in overall data quality and scientific reports.

show abstract

Fraud in Clinical Trials

Buyse

Evans

2016

Wiley StatsRef: Statistics Reference Online

View full text Add to dashboard Cite

Scientific research has a long history of fraud. In recent decades, the lay press has extensively debated over some cases of fraud in clinical trials, thereby fueling suspicions that fraud is a major problem in clinical research. Furthermore, the boundary between fraud and simple carelessness is often fuzzy, although the former is characterized by a deliberate attempt to deceive. Fraud, in the limited sense of data fabrication or falsification by investigators, is, in all likelihood, a rare phenomenon in clinical trials. Fraud, however, is so much at variance with the ethics of scientific research that any amount of it is deemed utterly unacceptable. Some cases of fraud could actually be prevented by making randomized clinical trials simpler and closer to clinical routine. Even so, any potential fraud must be detected, and statistical methods are especially well suited for this task in the context of multicenter trials. The detection of fraud using statistical techniques rests on the difficulty of fabricating plausible data, particularly in high dimensions, and on the fact that clinical trials are prospective studies that use a unique protocol across a number of investigational centers. A comparison of each center with the others in terms of the distribution of some variables, either taken in isolation (univariate approach) or jointly (multivariate approach), can detect unusual patterns in the data. Central statistical monitoring makes intensive use of such comparisons for an overall assessment of the data consistency across all investigational centers, thereby supporting a risk‐based monitoring strategy of clinical trials.

show abstract

Central site monitoring: Results from a test of accuracy in identifying trials and sites failing Food and Drug Administration inspection

Cited by 39 publications

References 14 publications

Statistical Monitoring in Clinical Trials: Best Practices for Detecting Data Anomalies Suggestive of Fabrication or Misconduct

Statistical Monitoring in Clinical Trials: Best Practices for Detecting Data Anomalies Suggestive of Fabrication or Misconduct

Qualitative features in clinical trials: coordinates for prevention of passive and active misconduct

Fraud in Clinical Trials

Contact Info

Product

Resources

About