No data may be collected before the Institutional Ethics Committee[5] has approved the study. The Indian Council of Medical Research's National Ethical Guidelines for Biomedical and Health Research Involving Human Participants (2017) is the reference Indian committees work from, and a dataset collected first is not rescued by a letter dated later. The practical consequence is that the ethics packet is where your thesis really begins, and a thesis that runs late has often lost the time to a packet that went round twice.
What goes in the packet
Your committee publishes its own checklist and states how many copies it wants, and that list governs everything below. Most Indian ethics committees ask for some version of the following.
- A covering letter and the committee's own application form, signed
- The full protocol: background, objectives, design, setting, inclusion and exclusion criteria, sampling technique, sample size with its formula and source, intervention or exposure, outcome definitions, the statistical analysis plan and the study period
- The participant information sheet, in plain language
- The informed consent form, in English and in the language your participants actually speak, with a translation certificate
- An assent form where minors are involved
- The case record form or proforma you will fill for each participant
- Curriculum vitae for the investigator, the guide and any co-investigators
- The investigator's undertaking, signed
- The guide's approval and the Head of Department's forwarding
- Good Clinical Practice or research ethics training certificates, where your committee requires them
- A declaration of funding and of conflicts of interest, including any kit, drug or device being supplied to you
- Any questionnaire or scale you intend to use, with the author's permission where it is not free to use
Read the protocol and the proforma side by side before you submit them, because the committee will. It checks whether every variable in the proforma is justified by an objective, and whether every outcome in the objectives is actually captured by the proforma. That mismatch is among the easiest objections for a member to raise and among the cheapest to prevent.
The information sheet and the consent form
These are two documents, not one, and committees treat them differently. The information sheet explains the study to the participant. The consent form records the decision. ICMR's 2017 guidelines set out what the information has to cover: the purpose and the procedures, how long participation lasts, what will be done with the data, the foreseeable risks, discomfort or inconvenience, any benefits, what happens if the participant is injured in the course of the research, how confidentiality will be maintained, that participation is voluntary and may be withdrawn without any effect on their treatment, and whom to contact with a question.
Write it to be read aloud. A sheet written at the reading level of your protocol is the document most reliably sent back. Short sentences, no abbreviation unexpanded, no word a patient would have to ask about, and a phone number that is answered. If the study offers the participant no benefit, say so plainly; a sheet that hints at better treatment for joining is worse than a blunt one.
The consent form has to exist in the language your participants speak, and your committee will normally want a certificate from the translator and sometimes a back-translation. Do not produce the translation yourself the night before submission. Prepare as many languages as your setting genuinely needs. For a participant who cannot read, a thumb impression with an impartial witness is the recognised route, and your committee will specify who may act as that witness.
Vulnerable participants
ICMR's guidelines treat some groups as needing additional safeguards because their ability to consent freely is constrained: children, people with cognitive impairment or mental illness, the unconscious and the critically ill, pregnant women, prisoners, and people whose economic or social position makes refusal feel costly. The committee will ask what your safeguards are, and will not accept their absence.
For children, consent is taken from a parent or guardian, and the guidelines also require the child's own assent, written or verbal, between the ages of seven and eighteen, in language the child understands. Assent is not a formality. A child's refusal means something, and the protocol should say what you will do when it is given.
Where your participants are your own patients, the committee will want to know who takes consent, because a treating doctor asking for consent carries an implicit pressure on the person asked. Naming a co-investigator who is not involved in that patient's care is a small change that answers a large objection.
Waiver of consent for record-based studies
A retrospective study of case records cannot obtain consent from patients discharged two years ago, and ICMR's guidelines allow an ethics committee to waive consent where the research carries less than minimal risk and the waiver does not adversely affect the rights and welfare of the participants. Record-based research is the usual example. The waiver, though, is the committee's decision and not yours: you request it, with reasons, and you wait for it.
Ask for it explicitly in the application. State why consent is impracticable, describe exactly which records you will access and through whom, and describe how identifiers will be stripped at the point of extraction. A waiver of consent is not a waiver of confidentiality, and it does not extend to contacting those patients afterwards. If you intend to telephone them for follow-up, that is a different study with consent inside it.
Confidentiality and data handling
Put a paragraph in the protocol that a committee member can tick off. Every participant receives a study number. Names, hospital registration numbers, addresses and phone numbers do not enter the master chart. Where you need a link between study number and patient, and in a prospective study you usually do, it lives in one separate file held by the investigator, and the protocol says where it is kept and for how long.
Say who has access, where the file is stored, whether it is password protected and what happens to it at the end. ICMR's 2017 guidelines set minimum periods for which study records are preserved after a study ends: at least three years for biomedical and health research, and at least five years for regulatory clinical trials. Institutions and sponsors often require longer, so confirm the period your own committee applies before you write a number into the protocol.
Photographs, operative images, radiographs and anything carrying a patient's face or name need separate explicit consent for use in the thesis or in any publication, and the consent form should carry that clause if you intend to use them.
What committees actually send back
- An information sheet written in technical language, or with no contact number on it
- No local-language consent form, or no translation certificate
- A proforma collecting variables no objective asks for, or collecting the patient's name
- Objectives that do not match the outcomes described in the methods
- A sample size with no formula, no assumed values and no cited source for them
- No statement of who will take consent, where the treating doctor is the investigator
- No safeguards stated where children or critically ill patients are to be included
- A waiver of consent assumed rather than requested
- Nothing on compensation or on management of research-related injury, in an interventional study
- No plan for incidental findings: what you will do when a test you ran for the study reveals something the patient needs treated
- A prospective interventional study with no mention of trial registration
Every one of those is fixable in an afternoon before submission and costs a full committee cycle afterwards. Read your own packet once as the member who would have to raise the objection.