Samira Kohli v Dr. Prabha Manchanda (2008)

This is the consent case for the Indian syllabus. It answers the three questions a consent problem can ask: who may consent, how much information must be given, and what a consent covers. It settles that a competent adult alone can consent for herself, that a consent for a diagnostic procedure is not a consent for treatment, and that India follows the professional standard of disclosure rather than the prudent-patient test — for the present.

The appellant, an unmarried woman aged 44, consulted the respondent gynaecologist on 9 May 1995 with prolonged menstrual bleeding. She was advised a laparoscopy under general anaesthesia to confirm the diagnosis. On 10 May her signature was taken on an admission card, on a consent form for admission and treatment, and on a consent form for surgery which described the procedure as diagnostic and operative laparoscopy. Laparotomy may be needed.

While she was under anaesthesia, the respondent's assistant came out of the theatre and took the consent of the patient's mother, who was waiting outside, for a hysterectomy. The respondent then removed the uterus, both ovaries and the fallopian tubes. The patient said she had come only for a diagnostic test and, intending to marry and start a family, would have refused consent and chosen conservative treatment. The National Commission dismissed her complaint, holding that she suffered from endometriosis (Grade IV) and that the surgery had been done with care. She appealed.

There was no valid consent for the hysterectomy and the removal of the ovaries. The surgery was an unauthorized invasion and interference with appellant's body which amounted to a tortious act of assault and battery and therefore a deficiency in service. Because the surgeon had acted in good faith and for the patient's benefit, the Court moderated the relief: it denied the surgeon her entire fee and directed her to pay Rs. 25,000 as compensation.

The Court summarised the governing principles:

• (i) Real and valid consent. A doctor must seek and secure the patient's consent before commencing treatment, surgery included. The consent must be real and valid: the patient must have capacity and competence, the consent must be voluntary, and it must rest on adequate information about the nature of the procedure.
• (ii) Adequate information. The doctor should disclose the nature and procedure of the treatment and its purpose, benefits and effect; the alternatives, if any; an outline of the substantial risks; and the adverse consequences of refusing treatment. He need not explain remote or theoretical risks which may frighten or confuse the patient.
• (iii) Consent is procedure-specific. Consent given only for a diagnostic procedure, cannot be considered as consent for therapeutic treatment. That the unauthorised additional surgery is beneficial, saves time and expense or relieves future pain is no defence. The only exception is where the additional procedure is necessary to save life or preserve health and it would be unreasonable to wait until the patient regains consciousness.
• (iv) Common consent may be taken for diagnostic and operative procedures where both are contemplated, and for a surgical procedure together with a further procedure that may become necessary.
• (v) The standard of disclosure need not be of the stringent and high degree laid down in Canterbury, but of the extent accepted as normal and proper by a body of medical men skilled and experienced in the particular field.

On the mother's consent the Court was categorical: when a patient is a competent adult, there is no question of someone else giving consent on her behalf. On emergency, it held that the surgeon ought to have waited until the appellant regained consciousness, discussed the result of the laparoscopic examination and then taken her consent for the removal of her uterus and ovaries.

The Court set two models side by side. In the United Kingdom consent is real when it is voluntary, given by a competent person and based on the minimum adequate information, the amount of information being judged by what a responsible body of doctors would disclose (Bolam, applied to warnings in Sidaway). In the United States, Canterbury v Spence requires disclosure of every risk material to a reasonably prudent patient's decision. Having regard to ground realities in Indian health care, the Court consciously preferred the real-consent concept — while warning that as practice becomes more commercialised and patients more aware of their rights, a day may come when India has to move towards Canterbury, but not for the present.

The finding of no consent was then made on the documents themselves: the consent form showed a consent to diagnostic and operative laparoscopy, not to a hysterectomy. The moderation of relief followed from the surgeon's good faith — the wrong was the absence of authority, not bad medicine.

It is the spine of the long answer on consent: definition → why consent is required → the types (implied, express, real, informed, common) → who may consent (BNS ss. 27 and 28 for children and persons of unsound mind; the statutory consents under the termination, transplant and sex-selection laws) → emergency (BNS s. 30) → Samira Kohli's five principles → the statutory disclosure duties.

In a problem where a surgeon extends an operation on a relative's consent, the four steps are: the consent was for procedure A, and consent is procedure-specific; a spouse or parent cannot consent for a competent adult; benefit to the patient is no defence; and the emergency exception fails unless the further step was necessary to save life or health and waiting was unreasonable. Conclude with assault and battery in tort and deficiency in service, and note that relief may be moderated where the doctor acted in good faith — as here, fee denied plus Rs. 25,000.

It also decides an informed-refusal point, through principle (ii)(d): a refusal is only as good as the information behind it, so the adverse consequences of refusing must be explained — without exaggerating remote risks to push a patient into unnecessary treatment.

What it does not decide: it does not adopt the prudent-patient test (stating a common error to avoid — India follows the real-consent standard "for the present"); it does not hold that good faith is a defence to the absence of consent, only that it affects relief; and it does not deal with consent for a patient who is permanently incapable, which belongs with the end-of-life cases.