Opioid Use Among HIV-Positive Pregnant Women and the Risk for Maternal–Fetal Complications

Ngoc H. Nguyen, PharmD; Erika N. Le, PharmD; Vanessa O. Mbah, PharmD; Emily B. Welsh, PharmD; Rana Daas, BS; Kiara K. Spooner, DrPH, MPH; Jason L. Salemi, PhD, MPH; Omonike A. Olaleye, PhD, MPH; Hamisu M. Salihu, MD, PhD


South Med J. 2020;113(6):292-297. 

In This Article


The analysis used the records of 57,331,147 pregnant women, 73,237 of whom were diagnosed as having HIV, which is equivalent to a prevalence of 127.7/100,000. In Table 1, we summarized the sociodemographic characteristics of the HIV-positive and -negative pregnant women in the study. Compared with HIV-negative pregnant women, those who were HIV positive tended to be slightly older, with 17.2% being of advanced maternal age (35 years old and older). There was a remarkable preponderance of pregnant black women who were HIV positive, accounting for 54.8% of all HIV-positive pregnant women (even though black women made up <12% of all pregnant women in the United States). Women in the lowest household income category and receiving public health insurance composed the majority of HIV-positive pregnant women. The southern part of the United States disproportionately accounted for the majority of HIV-positive pregnant women (56.8% of all HIV-positive cases), despite pregnant women from the South making up only <38% of all pregnant women. The Northeast comprised one-fourth of all HIV-positive pregnant women in the United States. This implies that >80% of all cases of HIV among pregnant women came from either the southern or northeastern part of the United States.

Of the 57,331,147 pregnant women in the entire study population, opioid use was diagnosed among 257,891 pregnant women, yielding a prevalence rate of 4.5/1000. The prevalence of opioid use among HIV-positive pregnant women was sixfold as high compared with their HIV-negative counterparts (27.8/1000 [N =2033] vs 4.5/1000 [N =255,850]). The Figure displays temporal trends in the use of opioids across the decades by HIV status. Among HIV-positive pregnant women, the prevalence of opioid use dropped from 41.7/1000 at the beginning of the study in 2002 to 32.5/1000 by the end of the study. Using joinpoint regression, this represented an annual decrease of 1.1%, which was not statistically significant (95% confidence interval 5.5–3.6). By contrast, the temporal trajectory of opioid use among HIV-negative pregnant women displayed an upward trend. Opioid use among HIV-negative women rose from a prevalence of 2.2/1000 at the commencement of the study to approximately 8.8/1000 by the end of the study period. This was equivalent to a fourfold increase in opioid use among HIV-negative pregnant women. The annual percentage increment in opioid prevalence among these women was 13.2%, which was substantial and significant (95% confidence interval 12.1–14.3).


Temporal trends in the prevalence of opioid use (per 1000 pregnancy-related hospitalizations), by human immunodeficiency virus (HIV) status, in the United States, 2002–2014. The x axis represents the year of discharge and the y axis represents the rate of opioid use (per 1000 pregnancy-related hospitalizations). Lines represent the trend estimated by joinpoint regression. Values represent the annual percentage change (APC) and point estimate (95% confidence interval).

Table 2 displays the association between HIV-positive status and opioid use, as well as selected maternal–fetal conditions. The adjusted odds ratios took into account the confounding effects of age group, race/ethnicity, household income, insurance status, admission day, hospital census region, hospital size, and teaching status. The adjusted estimates were relatively lower than the crude estimates, and the difference was more pronounced for opioid use, alcohol use, and sepsis. HIV-positive pregnant women exhibited a fourfold adjusted likelihood for opioid and alcohol use. The adjusted odds for tobacco use and depression among HIV-positive pregnant women were lower, approximately twofold when compared with HIV-negative pregnant women. Similar to maternal conditions, the fetuses of HIV-positive pregnant women also experienced elevated odds for adverse outcomes. The adjusted odds for spontaneous abortion, any abortive pregnancy, and early onset of delivery were significantly amplified among HIV-positive pregnant women; however, poor fetal growth was only 11% elevated among HIV-positive pregnant women and was not statistically significant.

Table 3 shows the association between HIV and opioid use status as the exposure and maternal–infant outcomes as endpoints. Among HIV-positive and HIV-negative pregnant women, the concomitant use of opioid medications worsened maternal outcomes severalfold. For HIV-positive pregnant women, the use of opioid drugs worsened maternal outcomes more than it did among HIV-negative pregnant women, except for tobacco use, in which risk estimates were comparable in magnitude. Among HIV-positive pregnant women who abstained from opioid use, risk estimates for maternal complications were substantially lower than for opioid users, regardless of the latter's HIV status. For fetal outcomes, the incremental risk caused by opioid use among HIV-positive pregnant women was much less pronounced and only observed for any abortive pregnancy and early onset of delivery.