Washington DC Metropolitan Area Drug Study, 1992: Drug Use Among DC Women Delivering Live Births in DC Hospitals

Research Question: Is there a correlation among the marital status of mothers, the number of cigarettes they smoked per day during pregnancy, the number of alcoholic drinks this pregnancy, and how often the mother used marijuana, crack, or heroin during this pregnancy?

Sources of Data: Washington DC Metropolitan Area Drug Study, 1992: Drug Use Among DC Women Delivering Live Births in DC Hospitals

Summary of Statistical Procedure: A correlation matrix used to look at any association between the variables listed above. .

Findings:

SDA 2.1: Correl DC MADS, 1992: Live Birth Study - Infant File

Oct 22, 2006 (Sun 08:11 AM EDT)

Variables Role Name Label Range MD Dataset Correlate B_2 Current marital status 1-5 *--1 1 Correlate C_4 # of cigarettes smoked a day/ this pregn 1-8 91,*--1 1 Correlate C_9 # drinks usually had this pregnancy 1-23 91,*--1 1 Correlate C_12 How often used marijuana this pregnancy 1-8 91,*--1 1 Correlate C_16 How often used crack cocaine pregnancy 1-91 *--1 1 Correlate C_27 How often used heroin this pregnancy 1-91 *--1 1 Weight IB_ANWT Infant quex analysis weight variable .0000-67.5487 1

Correlation Matrix B_2 C_4 C_9 C_12 C_16 C_27 B_2 1.00 -.31 .37 -.32 -.42 .03 C_4 -.31 1.00 -.36 .14 .21 -.23 C_9 .37 -.36 1.00 -.17 .03 .24 C_12 -.32 .14 -.17 1.00 .02 .43 C_16 -.42 .21 .03 .02 1.00 -.08 C_27 .03 -.23 .24 .43 -.08 1.00 Missing data excluded: Listwise

Color coding: <-0.45 <-0.30 <-0.15 <0.00 >0.00 >0.15 >0.30 >0.45

Allocation of cases (unweighted) Valid cases 27 Cases with invalid codes on variables correlated 1,021 Total cases 1,048

Datasets 1 /SDA/SAMHDA/dcmads92-inf 2 /SDA/SAMHDA/dcmads92-inf

CSM, UC Berkeley

Discussion of Findings: The correlation matrix shows a moderate correlation between marital status and the number of alcoholic drinks during pregnancy, while there is almost a matching negative correlation between marital status and the number of cigarettes smoked per day. Another moderate correlation exists between the number of times mothers used heroin and the number of times mothers used marijuana during their pregnancy. A weak correlation exists between the number of times heroin was used and the number of alcoholic drinks during the pregnancy, but again, an almost equally weak but negative correlation exists between the number of times heroin was used and the number of cigarettes smoked per day. I think it is interesting that marital status might predict that a mother is likely to drink more alcoholic beverages during pregnancy, but fewer cigarettes. (I think I am interpreting the data correctly?)

Also interesting is the positive correlation between heroin and marijuana and the weaker positive and negative correlations between heroin use and alcohol and cigarette use. This might indicate a pattern based on how these chemicals affect the user, but I do not have any background for understanding this angle. These results might be used to indicate areas where further research could or should be conducted.