The cholera hospital in Matlab, Bangladesh, has patients in the corridors and every nook and cranny, but as we walk through the mood is calm. Most beds have two people, usually a mother and child. The mood may be calm because acute, watery diarrhoea is part of life in Bangladesh, and the parents and patients know that when they reach the hospital they will be treated effectively for free and that almost nobody dies. The only deaths are those who arrive dead and in those with diarrhoea complicated by malnutrition or other infections—and even among them deaths are rare. Most people are discharged within 24 hours. Both wards and corridors are full because one of the two peaks of diarrhoea a year comes in April.

The services provided by the hospital in Matlab are part of an unwritten contract between ICDDR,B (once the International Centre for Diarrhoeal Disease Research, Bangladesh but now known simply by its acronym) and the people who live around Matlab. Because these people have been intensely studied for 50 years: every birth, marriage, divorce, death, and much more is recorded. A community researcher is responsible for around 7000 people and visits every household once every two months (it used to be every month).

Every last detail is known about these people, and every death is followed by a verbal autopsy. One finding is that 80% of deaths are now from chronic disease: the epidemiological transition has been remarkably fast in this poor, rural community. Many important trials have been conducted among the people of Matlab—including of oral rehydration therapy, family planning, and zinc for childhood diarrhoea. It’s the intensive study and the trial results that have made Matlab internationally famous, particularly among those interested in global and public health.

I was visiting for the second time, and I find it a peculiarly beautiful place. We arrived by speedboat, skimming down the wide river with its banks of maize, rice fields, palm trees, and buffaloes wading in the water. Children splashing in the water waved to us, and we waved back. Wooden boats with fisherman were rocked as we sped by—but seemed wholly unconcerned. And we passed the wooden boats with covered tops that are the homes of nomadic families living from fishing and selling small goods in the villages.

Our first stop was one of the subcentres where the medical assistants provide primary care to mothers and children and try to persuade women to give birth. Because the people in Matlab have faith in the services provided by ICDDR,B about 70% of women give birth in the centres, whereas in Bangladesh as a whole 70% of women give birth at home. If the subcentre can’t manage a problem, then patients may be transferred to Matlab hospital—by rickshaw ambulance. It takes about an hour. The hospital can’t at the moment perform a Caesarean section, and women who need a section have to be transferred to a government hospital. There are about 5000 births a year, and last year there were two maternal deaths. Some years there are five or six.

We enter the clinic where women are sat close together with their children waiting to be seen. Privacy is not a big issue in Bangladesh, and the women can see and hear the consultations. Nor is anybody bothered by us, although they do cover their heads as we enter. We seem to be a harmless diversion from the waiting. Charts on the wall tell us that immunization rates are all over 95% and contraceptive prevalence is 75%. (I wonder what the rates are back home in Clapham, where we are not so intensively studied. The immunization rates, I’m sure, are not so high.)

Then we head to a village to meet with a community researcher. The road is narrow and raised high and keeps turning at right angles to make its way through the paddy fields. The intense green of the paddy fields, all tiny but divided only by narrow ditches, stretches away into the distance. The road is raised because when the monsoon comes at the end of June, the paddy fields will become a lake. Even now approaching the end of the dry season there are plenty of ponds, and drowning is now one of the commonest causes of death in children.

The villages are also raised up, and we walk along a narrow path to reach a village of about 10 houses. There is no electricity, running water, or sewerage in many of these villages, but each house is classified into rich to poor quintiles by ICDDR,B. The “rich” houses have tin roofs, the poor ones are simply bamboo and other vegetation; most have earth floors, and people cook over fires, often burning cow dung. We are standing outside a third quintile house. It doesn’t have a tin roof and looks as if the monsoon rains will easily get through.

We are talking to the community researcher, who tells us that she knows everyone of her 7000 people. I ask her what she likes most about her job: “Being able to give women a pregnancy test when they think they may be pregnant and then putting them in contact with the clinic.” She’s been doing the job a long time and must have seen mothers die though childbirth. I never have, although I know that globally there are about half a million deaths a year. She shows us the big hardback book in which she records all the details, which are later entered into the computer. We are also shown PDAs (personal digital assistants), which are already used for some studies and will soon be used for all data collection.

As we talk more and more people gather. The children are keen to be photographed and are excited to see the digital images. As we walk from the village the children walk with us, I feel like the Pied Piper of Hamelin.

We drive again through the paddy fields to another village, where we visit the home of one of the community health workers. Her house is the grandest we’ve seen, two story and brick. Her husband died last year, but as she has done for decades she holds a clinic once a week where she vaccinates the children. Many of the babies are tiny, and a quarter of babies are low birth weight. Many of the mothers look young, although the legal age for marriage is 18 in Bangladesh. But many women don’t know how old they are, and some will pretend to be older than they are.

From the clinic we drive to the hospital and visit the large barge that was the beginning of the Matlab surveillance. The researchers slept upstairs and conducted their clinics and studies downstairs. As well as visiting the diarrhoea wards of the hospital we see to the laboratories and the ice cold rooms where the data are entered into the computers. A researcher shows us how one woman’s life has been recorded—her birth, marriage, divorce, and second marriage. Divorce seems to carry little stigma and is often “socially” rather than legally arranged. The researchers have 16 choices for recording the reason for divorce, including “Male debauches” and “Non-delivery of dowry.” We then visit the new ward where the hospital provides kangaroo care for babies born prematurely (the baby is skin to skin with the mother).

Finally it’s time to leave. We wave to the crowd as our speedboats putter up the narrow river. After passing another cluster of the nomads’ boats we reach the broad river and accelerate. The sun begins to go down, and I try to spot the village we visited. I imagine the people preparing for the night. Many will presumably go to bed as dark descends at about 7 pm, sleeping perhaps six to a room, and will rise with the dawn. This is a life far from the life I live. The place is so beautiful it’s possible to imagine the life as idyllic, but much of it must be hard. But, as one of the researchers said to us, people have little and are content with little. Is it true, I wonder? And is it also true that the whole place may be permanently under water by 2040? It may be.