Table 1. Characteristics of four cohort studies and the collection of prenatal care visits data.

No	Cohort name	Location	Baseline survey and initial sample size (N)	Year of follow-up used in analysis and number examined (N)	Initial cohort
1	Pelotas 1982 Cohort Studies	Pelotas, Brazil	1982, 5914	2005, 4297	Children born in the city’s maternity hospitals (>99% of all births) during 1982. All social classes included.
2	The Institute of Nutrition of Central America and Panama (INCAP) Longitudinal Study	Rural Guatemala	1969–77, 2392	2004, 1571	Intervention trial of a high-energy and protein supplement in women, and children <7 years in 1969 and born during 1969–1977 in 4 villages
3	Cebu Longitudinal Health and Nutrition Survey (CLHNS)	Cebu (the Philippines)	1983, 3080	2005, 2032	Pregnant women living in 33 randomly selected neighborhoods; 75% urban. All social classes included.
4	1990 Birth to Twenty (BT20)	Soweto-Johannesburg (South Africa)	Apr 23-Jun 8 1990, 3273	2009, 2225	Babies born to pregnant women living in a defined urban geographical area. Predominantly poor, black sample.

Note: Each COHORTS data set varies in the way that prenatal care data was collected: 1) Mothers were asked at birth about number of antenatal clinic visits and the pregnancy month initiating the visits in the Pelotas Cohort 1982; 2) The pregnant month at each trimester prenatal exam was recorded by researchers in the Guatemalan sample. No visit if the information is missing; 3) At the 6-7th month pregnancy pregnant women were asked the number of prenatal care visits and the pregnant month of first visit to each of public/private doctor/nurse/midwife and traditional midwife/healer in CLHNS. The final definition would be the total number of visits to all public/private types of care provider and the timing of the first visit to any public/private care provider; 4) Number of prenatal care visits and timing of the first visit was asked three times in the survey: prenatal, 6 months and 3 years old of baby in BT20. However not all individuals have three answers. Data from 6 months survey were used preferentially, and then 3 years old followed by prenatal one.