Chapter 3, Problem 66P

Precise quantification of smoking during pregnancy is difficult in retrospective studies. Routinely collected blood specimens from newborns for screening purposes may provide a low-cost method to objectively measure maternal smoking close to the time of delivery. Serum cotinine is an important biomarker of recent smoking. A study was performed comparing cotinine levels in dried blood spots in newborns with those in umbilical cord blood (the gold standard) among 428 newborns in the California Genetic Screening Program (Yang et al. [11]). The lowest detection limit for dried blood spot cotinine was 3.1 ng/mL. The data in Table 3.9 were presented relating dried blood spot cotinine determinations to umbilical cord blood cotinine determinations.

Suppose a cutoff of ≥ 5 ng/mL is proposed as a criterion for testing positive based on dried blood spot cotinine levels.

TABLE 3.9 Distribution of Cotinine Level in Dried Blood Spots from Newborns by Maternal Active Smoking Status* close to the time of delivery among 428 babies delivered in California, 2001–2003

^*Maternal active smoking at the time of delivery was defined as cord blood levels of ≥10 ng/mL.

Suppose it is estimated based on a large sample of births in California that 20% of mothers smoke at the time of delivery.

Suppose the screening test for detecting whether a mother smokes at the time of pregnancy is based on a cutoff of ≥ 5 ng/mL using dried blood specimens from the newborn.

What is the probability that a mother smokes at the time of delivery if the dried blood specimen is ≥ 5 ng/mL?