INTRODUCTION
An early and sustaining responsibility of a researcher is adherence to the protocol throughout the duration of the study and beyond. Failure to do so invariably results in protocol deviations/violations (PD/PVs), which, however trivial, are likely to affect both ethical and scientific standards of research. Thus, it is the responsibility of researchers to actively identify, report and take corrective and timely measures to minimize harm. Indian Council of Medical Research’s (ICMR’s) National Ethical Guidelines for Biomedical and Health Research involving Human Participants (2017) recommend Ethics Committees (ECs) to review and take corrective actions on PD/PVs.[1]
Even though ICH-Good Clinical Practice (ICH-GCP) guidelines do not remarkably differentiate between PDs and violations, various attempts have been made to differentiate deviation from violation.[2–4] Bhatt has summarized most of the popular ones, based on whether the unplanned action is an administrative one or has an impact on the scientific validity or harm to participants.[5] In another dimension, what is a major infraction versus a minor one is still not clearly delineated. Ghooi et al. have classified PDs based on the impact on subject’s well-being and quality of data as Grade 1 (nil) to Grade 5 (death of participants).[6] The ICMR guidelines mention but do not define PDs/PVs or their related terms “minor and major.”[1] It is left to the EC to define, identify, grade, review, decide and make recommendations for PD/PVs. Therefore, identification of PD/PVs remains suboptimal.
Although PD/PVs may occur on the part of the sponsor, researcher or the participant, the focus of our article is to identify and assess PD/PVs by the postgraduate-researchers reviewed by our EC. We hope other ECs will benefit from our experience.
METHODOLOGY
Yenepoya EC -1 (YEC-1) has a robust postapproval review process including site monitoring visits focusing on clinical trials and high-risk studies. However, the bulk of our protocols are postgraduate dissertations and these remain usually “under the radar.” It was perceived that a cross sectional “ice-pick view” of PD/PVs in these studies would guide YEC-1 in revising its approach to postapproval monitoring processes to include postgraduate dissertations as well. YEC-1 conducted an audit from September 2018 to December 2018, of ongoing MD/MS dissertations. After permission from the Chairperson, the audit was conducted by designated YEC-1 members (including the authors), in two phases:
Self-reporting checklist
A checklist consisting of the following items was e-mailed to 80 postgraduate-researchers to enable self-reporting of PDs:
- Concordance with the approved protocol in terms of study team composition, participant recruitment, sample size, informed consent (IC), and data collection
- Compliance with the EC approval letter in terms of status of the validity of EC approval, version of the protocol approved, essential communications with the EC, adverse events (AEs), PD/PVs, and participant withdrawal/discontinuation.
Physical audit of research documents
The auditors reviewed the research documents-IC and data collection forms using a checklist which had the following items:
- Conformance to EC approval validity
- Sample size recruited vis-à-vis approved
- Protocol, data collection forms, and ICD versions followed vis-à-vis approved
- IC form-version, language, and signatures
- Data collection form-completeness
- Record of serious adverse events (AEs/SAEs).
Classification of protocol deviations
The PDs were classified as follows:
- Noncompliance: Administrative nonconformance to the EC communications with no/minimal impact on the scientific validity and participant safety
- Protocol deviation: Change in any component of the approved protocol which is likely to entail a minimal or less than minimal increase in the risk of harm to participants, with or without effect on the scientific validity of the study
- Protocol violations: Any PD likely to entail more than minimal increase in risk, or likely to cause more than minimal harm to the participants.
Data analysis
The data were tabulated and analyzed using the descriptive statistics. The variables studied were types of PD/PVs and they were expressed in percentages.
Corrective action
Based on the seriousness of the PDs, corrective actions were suggested to the student-researchers.
RESULTS
We audited 54 responses out of 80 dissertations and found PD/PVs in 23 protocols.
1. Profiles of the protocols:
a. Type of study [Figure 1]: Most dissertations were observational studies (73.75%). Interventional studies included drug studies (10%) and medical devices, lasers, and surgeries (7.5%). Other studies included epidemiological and interview-based studies (3.75% each) and in vitro studies (1.25%)
;)
Figure 1: Types of studies taken up for MD/MS dissertation projects
b. Type of review: Majority (57.5%) were approved through expedited review
c. Inclusion of vulnerable populations: Of the 80 studies, 20 studies involved potentially vulnerable populations (25%) of which 6 were on children, 4 on patients with cancers, 2 each on pregnant women and geriatric populations, and one each involving students, psychiatric illness, and renal transplant patients.
2. Protocol deviations identified: The types of PDs identified are listed in Table 1. The percentages mentioned refer to “protocols with PDs” and not “PD per participant.” Of the eight types of protocol transgressions, 2 were administrative (noncompliances) and the rest were minor PDs. There were no major transgressions (PVs).
Table 1: Plausible negative impact of protocol deviation/protocol violation on various stakeholders in research (n=54)
DISCUSSION
Types of studies and categories of risk
As expected, majority (82.5%) of the studies had no more than minimal risk as per the ICMR risk categories.[1] As per the National Medical Council (NMC), the aim of a thesis/dissertation is to contribute to the development of a spirit of inquiry, critical analysis, techniques of research, and literature study in the postgraduate student.[7] The NMC curriculum for postgraduate students includes training in research methodology, medical ethics and medico-legal aspects, but there is no specific mention of “research ethics principles and participant protection.” It is noteworthy that during the audit, we found dissertation topics were perceived to carry risk of potential harm to research participants (vulnerable populations -20% and drug interventions -17.5%). Figure 2 shows the distribution of postgraduate studies involving vulnerable populations.
Figure 2: Protocols involving vulnerable populations
As per NMC regulations, theses/dissertation studies are carried out under the guidance of a recognized postgraduate teacher. In a study by Fakhar et al., the amount of guidance/supervision varied from maximum in 31.42% studies, to moderate in 35.77% studies and minimum in 32.79% for the research and depended on the nature of the teacher, student, and choice of topic.[8] The student-researcher may not be well trained in GCP guidelines or research ethics, may or may not be supervised “maximally” by the guide and the topic of research may pose a more than minimal risk of harm to participants. Therefore, considering that nearly two-thirds of postgraduate dissertations across the country have less than optimal supervision, attendant risk of harm to research participants is to be expected. There is sufficient ground for a compromise of research ethics in postgraduate dissertations.
The NMC and the ICMR have placed no restrictions on postgraduates conducting more-than-minimal risk studies. It is often left to the discretion of the EC to approve the study as submitted or get it modified to minimize risk to the greatest extent possible. This is not a meaningful solution, as the EC can at best assess and suggest mitigations to the risk, on the basis of the protocol-related documents submitted by the researcher. These documents may not be an accurate account of what the researcher will actually do, postapproval.
The Board of Governors, which superseded the Medical Council of India, in September 2018, made it mandatory for postgraduate students to undergo online research methodology training-conducted by ICMR-NIE-by the end of the second semester applicable from the 2019 batch onwards.[9] Ethical issues covered in the modules are ethical framework in research, publication ethics and clinical trials. It is hoped that this may provide the students the requisite awareness and training to carry out research safely.
Types of protocol deviations
Nonreporting for audit
Twenty-six of 80 postgraduates did not respond to self-reporting checklist. This is not indicative of responsible conduct of research. The current guidelines make ECs responsible to maintain records of all correspondence with stakeholders in research.[2] None of them highlight the responsibility of the researcher to respond to any research-related queries/communication by the EC. ECs have their own set of challenges and rarely undertake monitoring and auditing of protocols.[10] One of the important responsibilities of the researchers is to report to EC for auditing as and when indicated. Post this audit, YEC-1 censured the errant postgraduate students and increased their awareness so as to promote responsible conduct of research.
On perusal of our standard operating procedure (SOP) manual, we found a similar lack of reference to any responsibility of the researcher towards communicating with the EC. Subsequently we have introduced the following amendments in our SOP manual:
- The revised YEC-1 approval letter to the PI now clearly mentions the responsibility of the PI to respond to all e-mail communications from YEC-1
- The revised definition of PD now includes “failure of the researcher to follow YEC-1 recommendations enlisted in the approval letter.”
Nonreporting of protocol deviations
A researcher is expected to possess intellectual honesty and accuracy in performing and reporting the research.[11] This should be read as including reporting of PDs. 16 out of 23 types of PDs were not self-reported nor declared when given an opportunity to do so, but detected at the time of auditing. According to a study the detection of deviations solely rests on the EC and if PDs are not self-reported by the researchers, it reflects poorly on the ethical conduct of research.[6] Expecting ECs to catch all PDs is unrealistic and unwelcome. Student-researchers conduct research under qualified guides. Some responsibility may be assigned to them to oversee student research, identify PDs, report them and take appropriate correction actions to uphold the scientific and ethical integrity of research.
Our findings are similar to those reported by Sweetman and Doig who collated reports from 80 clinical trials and found that one-third failed to report any PD, and all failed to report all PDs.[12] The findings reported by Jalgaonkar et al. from Mumbai are different from those of ours, probably because the sample they drew on was a mixed bag of postgraduate dissertations and clinical trials.[13] This is not to suggest that ECs can forgo the primary responsibility of carrying out site monitoring visits, but in addition, such internal audits (especially where there are large numbers of postgraduates) can be a useful, check mechanism to keep track of unreported PD/PVs.
Table 1 maps the various types of PDs against our assessment of the negative impact these might have on the scientific validity of the study, the risk to participants, the functioning of EC s and the credibility of the institutions.
The practice of not responding to audit requests, may possibly have a negative impact on the scientific validity of the study, since more often than not, EC recommendations help researchers strengthen their methodology. The practice of nonreporting of PDs likewise, may similarly negatively affect the study validity and add risk to participants. ECs can help researchers who report PDs by recommending changes in the protocol that might mitigate risks to participants. These two transgressions also negatively impact the functioning of an EC, since the latter has to invest more time and energy in monitoring.
Nonconformity to the ethics committee approval validity period
Guidelines are clear that research on human participants can be initiated only after obtaining EC approval. However, none of the guidelines mention mandatory adherence to the approval “validity period.” Res ipsa loquitur, with EC approval letter comes its validity. However, during the audit, we found that 10 (18.5%) student-researchers had not conformed to this condition. They had either collected data before the date of EC approval, or beyond the validity period, or in the interim period between lapse of validity and subsequent renewal. This is a PD, albeit minor. There are no studies in the literature which delve into the risk aspects of conducting studies beyond the validity period, but it may be likened to the contravention of any official document (such as a driving license). Hunter (2016) has argued that prerecruitment in a general sort of way-a practice commonly utilized by many clinical trial organizations is associated with ethical dilemmas (mainly revolving around inducement), and such practices should be curtailed by the enforcement of regulations or through due perusal of such material by ECs.[14] While we agree with the stand taken by Hunter, our problem is different. We are of the opinion that recruitment before the start or after the expiry of the EC approval validity period might possibly have a negative impact on the participants, compromising their autonomy. Post this audit, YEC-1 issues “protocol approval letter” with a specified validity period (usually 1 year) and researchers are required to restrict recruitment and data collection to within the validity period. If required, the researcher is required to seek 6-monthly extensions of the validity period. These requirements are clearly stated in the approval letter of YEC-1 and further reinforced by a box highlighting the important dates for the researcher to take note. As a value-added service, YEC-1 sends a reminder e-mail 1 month before the end of validity period.
Exceeding the approved sample size
The sample size determination is an essential component and a prerequisite to both scientific and ethical review of protocols. Sample size not only defines the study power, but is also a measure of ethical acceptability of the study through balancing of the average burden per participant.[15] Therefore, exceeding sample size may be considered a significant PD with potential risk. Four out of 54 researchers were found to have exceeded the approved sample size, indicating potential risk to the participants without essentially increasing the power of the study. However, three of the studies were observational with less than minimal risk and one was interventional (exceeding sample size by 1) where no AE/SAE occurred.” This reflects the views of Faber and Fonseca that recruiting more samples than required poses an ethical issue since it predisposes more participants to the risk of harm in research and also impacts the resources used in the research.[16] YEC-1 now specifies the approved sample size in the revised approval letter. Researchers are expected to submit protocol amendments prospectively, if they intend to increase sample size after providing justification for the same.
Modification in research team/methodology before or without ethics committee approval
A research protocol is the sum of all documents. Once approved, any modification requires further approval by EC. According to the ICMR guidelines, EC may conduct an expedited review of minor modifications which include administrative changes or correction of typographical errors and change in researcher (s) and a full review for modifications in the IC documents, investigator’s brochure, advertisements, recruitment methods, etc., which involve an altered risk.[1] This makes it clear that even minor modifications require review and approval from EC. In our audit, 4 out of 54 researchers (all observational studies) had modified the protocols without getting prior approval. The protocol modifications included change in guide (2), change in the data collection form (1) and change in the equipment used for data collection (1). Following an unapproved protocol may pose potential risk to participants. However, on further analysis, none of the deviations had led to major risk to participants and therefore the deviations were considered minor. As a corrective action, all researchers submitted protocol amendments which were reviewed and approved by YEC-1 for further data collection.
Informed consent and language incongruence
Although literacy rate is high in this part of the country, many patients visiting the medical college hospital speak diverse local languages and are unable to read or write English. To ensure understanding, comprehension and valid decision making in research, the IC process must be administered and documented in a language the participant is conversant with. One of the challenges in the IC process is the linguistic barrier. In multi-linguistic regions, when translated versions of the IC are employed, inaccurate and incomplete translations can lead to misunderstanding. This defeats the purpose of IC. Accurately translated versions of the IC document, therefore, are the essential components of a protocol. These are reviewed by the layperson, in particular, and approved by the EC. Besides being laid down in the ICH-GCP guidelines, IC form is also considered a legal document.[17] Two out of 54 student-researchers (both observational studies) claimed to have administered the IC in the participant’s native language, but documented the same in English form for all the research participants. YEC-1 recommended re-consent or exclusion of the participants from data analysis.
Informed consent and incomplete documentation
In research on human participants written IC is mandated by the ICMR unless other forms of consent like verbal or electronic consent are approved by the EC. IC is considered operational and valid only if it is signed by both-researcher and participant. The audit showed that almost one-third of the student-researchers had incomplete documentation of consent forms. These lacked signatures by participants (2) or researchers (16) or and one of which lacked both (1). Studies have recommended that IC should go beyond just signing of the document.[18] Unsigned IC is unacceptable both ethically and legally. IC is therefore an important process in order to ensure voluntary, informed and understood willingness of the participants to be part of the study. Every guideline for research on human participants emphasizes on proper documentation of IC and nonconformity is a PD. YEC-1 recommended re-consent where signatures of participants were missing or exclusion of such participants from data analysis. For the others, researchers were educated to follow all the essential steps of IC.
Insecure data storage
Case record forms are considered crucial for the collection of consistent and valid data, timely submission of documents to the concerned authorities and for meaningful analysis of the study.[19] Two out of 54 researchers reported that some of the data collection forms were misplaced during the research period. It is the duty of the researcher to maintain the privacy, confidentiality, and safety of the data collected during the period of the study and beyond. Misplacing files containing research data amounts to deviation from responsible conduct of research. Following the audit, the students were advised to store all research-related documents in a secure manner with adequate safeguards. There are no studies in medical literature (PubMed as on July 15, 2021) which speak about misplaced data collection forms and its impact on research or research participants.
CONCLUSION
The findings of the audit indicate that PDs are common in postgraduate research projects and are under-reported. A proactive EC is required to detect them and help minimize risk of harm to participants.
Limitation
Twenty-six postgraduates did not report for auditing. It is possible that many more PD/PVs would have occurred. The audit was limited only to document-check and not real time monitoring of the IC and data collection processes involved in research. The PDs detected may therefore be only the tip of the iceberg.
Epilogue
This article is not to bring to the fore any major PDs that affected research participants, but to create awareness about the occurrence of minor deviations which often remain unnoticed, unreported and uncorrected in the garb of the ignorance of the novice researchers. The crux of the publication is to question the integration of the practice of research ethics right from the beginning of a research career. The stress is on the responsibilities of the research guides, departments, institutions and the ECs to be more educative and vigilant about the postgraduate studies in terms of the protocol adherence and research integrity. Dissertation lays the foundation for research methodology and research ethics among the postgraduate students and should therefore be strong and deep-rooted.
There is a need for proactive EC strategies for PDs. EC should strive to promote research integrity among PGs who will be mainstream researchers in future.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
REFERENCES
1. Indian Council of Medical Research. National Ethical Guidelines for Biomedical and Health Research involving Human Participants 2017. Available from:
https://main.icmr.nic.in/sites/default/files/guidelines/ICMR_Ethical_Guidelines_2017.pdf. [Last accessed on 2021 Dec 18; 1345 hours]
2. International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH). ICH Harmonised Guideline Integrated Addendum to ICH E6 (R1):Guideline for Good Clinical Practice E6(R2) 2016. Available from:
https://database.ich.org/sites/default/files/E6_R2_Addendum.pdf. [Last accessed on 2021 Dec 18; 1400 hours]
3. EFGCP Geriatric Medicines Working Party (GMWP). Medical Research for and with Older People in Europe. London, UK: European Forum for Good Clinical Practice; 2013. Available from:
https://efgcp.eu/documents/EFGCPGMWPResearchGuidelinesFinaledited2013052770.pdf. [Last accessed on 2021 Dec 18; 1235 hours]
4. Food and Drug Administration Compliance Program Guidance Manual. Bioresearch Monitoring Ch. 48. Maryland: Silver Springs; 2018. Available from:
https://www.fda.gov/media/112538/download. [Last accessed on 2021 Dec 18; 1345 hours]
5. Bhatt A. Protocol deviation and violation. Perspect Clin Res 2012;3:117
6. Ghooi RB, Bhosale N, Wadhwani R, Divate P, Divate U. Assessment and classification of protocol deviations. Perspect Clin Res 2016;7:132-6
7. Medical Council of India (National Medical Council). Postgraduate medical education Regulations 2000. Available from:
https://www.nmc.org.in/wp-content/uploads/2019/12/Postgraduate-Medical-Education-Regulations-2000.pdf. [Last accessed on 2021 Dec 22; 1600 hours]
8. Fakhar S, Saeed G, Ashraf M, Nisar N, Akhter T. Factors leading to topic selection for dissertation writing by post graduate medical residents. Ann Pak Inst Med Sci 2015;11:119-23
9. Indian Council of Medical Research School of Public Health. Basic course in Biomedical Research. Chennai: Indian Council of Medical Research School of Public Health; 2021. [Last accessed on 2020 Sept 28]
10. Kadam R, Karandikar S. Ethics committees in India:Facing the challenges!. Perspect Clin Res 2012;3:50-6
11. National Research Council (US) and Institute of Medicine (US) Committee on Assessing Integrity in Research Environments. Integrity in Scientific Research:Creating an Environment That Promotes Responsible Conduct. Washington (DC): National Academies Press (US); 2002; 2. Available from:
https://www.ncbi.nlm.nih.gov/books/NBK20∊/. [Last accessed on 2021 Dec 21]
12. Sweetman EA, Doig GS. Failure to report protocol violations in clinical trials:A threat to internal validity?. Trials 2011;12:214
13. Jalgaonkar SV, Bhide SS, Tripathi RK, Shetty YC, Marathe PA, Katkar J, et al. An audit of protocol deviations submitted to an institutional ethics committee of a tertiary care hospital. PLoS One 2016;11:e0146334
14. Hunter D. Response to: We could be heroes:Ethical issues with the pre-recruitment of research participants by D. Hunter. J Med Ethics 2016;42:206
15. Bacchetti P, Wolf LE, Segal MR, McCulloch CE. Ethics and sample size. Am J Epidemiol 2005;161:105-10
16. Faber J, Fonseca LM. How sample size influences research outcomes. Dental Press J Orthod 2014;19:27-9
17. Nijhawan LP, Janodia MD, Muddukrishna BS, Bhat KM, Bairy KL, Udupa N, et al. Informed consent:Issues and challenges. J Adv Pharm Technol Res 2013;4:134-40
18. Kadam RA. Informed consent process:A step further towards making it meaningful!. Perspect Clin Res 2017;8:107-12
19. Bellary S, Krishnankutty B, Latha MS. Basics of case report form designing in clinical research. Perspect Clin Res 2014;5:159-66
