On this page
Rotavirus is the most common cause of severe, dehydrating diarrhoea in young children in developing countries, according to the 2014 Lancet trial of Rotavac. India now gives its own vaccine against it to infants through the national programme. That vaccine started with an observation in a Delhi hospital nursery. This post covers M. K. Bhan, the research path, a worked example on how vaccine efficacy is calculated, and what students can learn from it. It is a history of science, not medical advice; questions about a child's vaccines belong with a doctor.
Who was M. K. Bhan?
The obituary of Maharaj Kishan Bhan in the National Medical Journal of India, by Vinod K. Paul, gives his dates as 9 November 1947 to 26 January 2020. It says he earned his MBBS at the Armed Forces Medical College, Pune, trained at Safdarjung Hospital, and worked in Kabul and Chandigarh before joining the All India Institute of Medical Sciences (AIIMS), New Delhi, in 1980.
At AIIMS he studied immunity in rotavirus infection and persistent diarrhoea. The obituary describes "painstaking 25-year research in rotavirus" that led to India's first indigenous rotavirus vaccine. From 2004 to 2012 he was Secretary of the Department of Biotechnology, where he set up the Biotechnology Industry Research Assistance Council (BIRAC) and biotechnology clusters in several cities. He received the Padma Bhushan and the Shanti Swarup Bhatnagar Award.
What did his team find in Delhi's newborns?
In 1993 Bhan and colleagues published Protection conferred by neonatal rotavirus infection against subsequent rotavirus diarrhea. They followed newborns in New Delhi who had picked up rotavirus in hospital in their first days of life. The abstract reports:
60% were infected by the fourth day of life, and the infection caused no symptoms.
It was caused mostly by an unusual strain, G9 P11, "not previously identified in humans".
The 148 infected children had 46% fewer attacks of rotavirus diarrhoea over 14 to 23 months than the 56 who were not infected: 0.23 against 0.42 episodes per child per year.
A strain that infects babies harmlessly and seems to protect them is a natural vaccine candidate. The authors were cautious, writing that its use as a vaccine would need further testing of its protection in the field.
How was Rotavac tested?
Rotavac is built on a G9P[11] strain called 116E, which the Lancet paper describes as a natural human-bovine reassortant. In the randomised, double-blind, placebo-controlled trial published in The Lancet in 2014, with Bhan as the last-named author, infants at sites in Delhi, Pune and Vellore received three doses at 6 to 7, 10 and 14 weeks. 4,532 were assigned the vaccine and 2,267 placebo. Efficacy against severe rotavirus gastroenteritis was 53.6% over the follow-up period and 56.4% in the first year of life. The trial recorded six cases of intussusception (a bowel blockage) in the vaccine group and two in the placebo group, all after the third dose.
The NMJI obituary says the Prime Minister launched Rotavac in March 2015. A study in Nature Medicine (2025) reports that India introduced Rotavac (G9P[11]) into its Universal Immunization Program in 2016. Across 31 hospitals in 9 states from 2016 to 2020, three doses were 54% effective against severe rotavirus gastroenteritis, and the share of hospitalised children testing positive for rotavirus fell from 40% to 20%.
Worked example: how is efficacy calculated?
The Lancet abstract gives the raw numbers for the main analysis: 71 severe cases in 4,752 person-years of follow-up in the vaccine group, and 76 cases in 2,360 person-years in the placebo group. A person-year is one child followed for one year.
Vaccine group rate: 71 / 4,752 = 0.0149, or about 1.5 per 100 person-years.
Placebo group rate: 76 / 2,360 = 0.0322, or about 3.2 per 100 person-years.
Rate ratio: 0.0149 / 0.0322 ≈ 0.46.
Efficacy ≈ 1 - 0.46 = 0.54, or 54%.
These match the abstract's figures of 1.5 and 3.2 per 100 person-years and a ratio of 0.46. The published 53.6% comes from the trial's full statistical model, so a hand calculation lands close but not exactly on it.
The same method works on the 1993 newborn study: 1 - 0.23/0.42 ≈ 0.45, close to the 46% reported from unrounded rates.
Common mistakes: comparing case counts instead of rates (the vaccine group was twice as large), and reading 54% efficacy as "54% of children are protected". It means the rate of severe disease was about halved. For how the immune system builds protection, see our post on human health and immunity.
What can a student take from how Bhan worked?
Notice the odd case. Healthy infected babies were the clue.
Test, then test again. A 1993 cohort study led, two decades later, to a large randomised trial and then to real-world checks.
Think about access. Building BIRAC and research clusters was about making Indian discoveries reach people.
If terms like "placebo-controlled" or "rate ratio" are unclear, the Fermi Concept Library covers every concept in physics, chemistry, biology, mathematics and computing, so you can look them up as you read. For another vaccine pioneer, read about Maurice Hilleman.
Next step: using the Nature Medicine figures, work out by what fraction rotavirus positivity among hospitalised children fell.
