By David Brown
Monday, February 1, 2010; A10
PORT-AU-PRINCE, HAITI -- How do you find out what the Haitian people really need now, nearly three weeks after the earthquake? You ask them.
Three-people teams, each including at least one member who speaks Creole and French, are fanning out to 152 sites across the country this week, asking a long series of questions to "key informants." When the teams are done, they hope to have a clearer picture of the food, shelter, water, sanitation and health-sector needs of the Haitian population.
That's the intent of a week-long project run by the United Nations' Office for the Coordination of Humanitarian Affairs, the World Health Organization and the U.S. Centers for Disease Control and Prevention.
It is an experiment in another way, too. The data are being collected on hand-held computers (PDAs) as well as on paper forms marked with pencils.
Much as truth is the first casualty of war, reliable information is one of the early casualties of natural disasters. Until fairly recently, disaster responders relied on their senses -- and their common sense -- to identify problems. The notion of measuring what you could see was viewed as an academic response to things such as earthquakes, hurricanes and tsunamis.
That view has changed.
The "evidence-based" paradigm slowly remaking medicine has crept into disaster medicine, too. So has the drive to make this field, like much of the rest of medical care, more democratic and responsive to the consumer and less paternalistic and responsive to the expert.
It is true, though, that time is of the essence, even a few weeks after a disaster such as this. Methodological rigor has to wrestle with the need to get things done.
The survey this week didn't ask questions of a random sample of Haitians in the way that a medical trial would. That would have been a huge and time-consuming undertaking. Instead, it sought out individuals expected to know what was happening to the people in their area: mayors, village directors, health officials. The places weren't chosen randomly either. The designers chose fairly evenly spaced sampling sites, with extra ones in the heavily damaged Port-au-Prince area.
There is also the matter of what to ask. Too little, and it's worthless. Too much, and it tries the patience of interviewer and informant. "This is a not-so-rapid rapid assessment," Carl Kinkade, a CDC epidemiologist, said as he fluttered the 10-page questionnaire while waiting in the mid-morning heat for interviewers at a dusty lot in a centrally located neighborhood.
One of the teams went to Kenscoff, an administrative district of six villages south of Port-au-Prince. Reaching it took an hour of steady uphill driving. Part of the way, the road traveled the edge of a gorge-like valley, with lush terraced fields on its slopes. The destination was a village called Nouvelle Touraine.
Relatively speaking, the village was spared. Of its 7,981 residents, one person died in the earthquake and one was injured. Two Catholic churches and 64 houses were destroyed; nine other houses were damaged.
The chief interviewer was Jules Figaro, a 31-year-old graduate of Haiti's state university, where he studied ethnology. He lives in Cite Soleil, the capital's famous slum, with his parents, five brothers and three sisters. He has never had a job. Assisting him were Alex Guerrier, 24, a fourth-year student of administrative sciences, and Anne-Marie Saint-Victor, a 21-year-old medical student. The team leader got $50 for the day, the assistants $30.
They spoke with the appointed director-general of Kenscoff, a 40-year-old man whose crowded desk included a Haitian flag on a dowel that he had put at half-staff. He answered the questions with confidence.
The temporary housing's protection from weather: poor. Privacy: acceptable. Security: poor. Sheltering households with water-purifying chemicals: less than 25 percent. Ones with necked-water containers that make water storage safer: 50 to 75 percent.
Then it was around the corner to the Sanitary Bureau, where a public health doctor and infectious diseases specialist, Dorothy Posy, provided the information.
Number of births in Nouvelle Touraine in the last week: 10. Number with trained attendants: nine. Cases of diarrheal disease: one.
Posy was happy to report that the 10 people on antiretroviral therapy for HIV infection and the four under treatment for tuberculosis had no interruption of treatment.
There were some problems with the PDA, and Guerrier couldn't keep up with all the answers. So he would copy down the ones he'd missed from the paper form later on. Consequently, the analysis of how similar the two sets of data are -- paper vs. digital -- will be muddied for this team. And then there was the problem of running into another survey team as they descended the stairs from the Sanitary Bureau.
Scott Dowell, head of CDC's efforts here, speaks Creole and spent much of his childhood at the Albert Schweitzer Hospital, about 50 miles northeast of the capital, where his father worked as a pediatrician. He said the problem of bumping into other teams was bound to happen occasionally. If a team finds no reliable informant at the place it is assigned, it is instructed to go the nearest place there is one.
"So two teams are going to end up in the same place some of the time," he said.
It turns out, though, this duplication was intentional, a random test of the "reproducibility" of the data.
It will be next week before the survey findings are compiled and considerably longer before the success of the PDA experiment is judged.
But there are some experts who think that the brief day of the PDA is over and that there's only one way to efficiently collect epidemiological data in the field. It's with a cellphone, the hand-held computer just about everyone in the world can afford.