Migrant women living with HIV in Europe: are they facing inequalities in the prevention of mother-to-child-transmission of HIV?: The European Pregnancy and Paediatric HIV Cohort Collaboration (EPPICC) study group in EuroCoord*

Migrant women living with HIV in Europe: are they facing inequalities in the prevention of... Abstract Background In pregnancy early interventions are recommended for prevention of mother-to-child-transmission (PMTCT) of HIV. We examined whether pregnant women who live with HIV in Europe and are migrants encounter barriers in accessing HIV testing and care. Methods Four cohorts within the European Pregnancy and Paediatric HIV Cohort Collaboration provided data for pooled analysis of 11 795 pregnant women who delivered in 2002–12 across ten European countries. We defined a migrant as a woman delivering in a country different from her country of birth and grouped the countries into seven world regions. We compared three suboptimal PMTCT interventions (HIV diagnosis in late pregnancy in women undiagnosed at conception, late anti-retroviral therapy (ART) start in women diagnosed but untreated at conception and detectable viral load (VL) at delivery in women on antenatal ART) in native and migrant women using multivariable logistic regression models. Results Data included 9421 (79.9%) migrant women, mainly from sub-Saharan Africa (SSA); 4134 migrant women were diagnosed in the current pregnancy, often (48.6%) presenting with CD4 count <350 cells/µl. Being a migrant was associated with HIV diagnosis in late pregnancy [OR for SSA vs. native women, 2.12 (95% CI 1.67, 2.69)] but not with late ART start if diagnosed but not on ART at conception, or with detectable VL at delivery once on ART. Conclusions Migrant women were more likely to be diagnosed in late pregnancy but once on ART virological response was good. Good access to antenatal care enables the implementation of PMTCT protocols and optimises both maternal and children health outcomes generally. Introduction In pregnant women with HIV, antenatal anti-retroviral therapy (ART) is crucial for viral suppression during pregnancy and at delivery and the prevention of mother-to-child-transmission (PMTCT).1 It is recommended that women conceiving on ART should continue ART during pregnancy and that those untreated at conception—i.e. women who are newly HIV-diagnosed in pregnancy and those already HIV-diagnosed but not treated—should start ART in early pregnancy.2 To ensure prompt antenatal ART start in women who are undiagnosed at conception it is recommended that HIV screening should be performed in early pregnancy. Migrant populations living in Europe are disproportionally affected by HIV3 but may face barriers in accessing HIV testing and care,4–6 including pregnant women. Migrant women may be less likely to receive optimum antenatal care than native women,7,8 potentially leading to delayed HIV diagnosis and delayed ART start with negative implications for mother-to-child-transmission (MTCT) risk and maternal health. We aimed to assess whether migrant women with HIV encounter barriers in accessing HIV testing and care during pregnancy using contemporary data across several European countries. To do so, we compared three markers of suboptimal PMTCT interventions in migrant and native women: HIV diagnosis in late pregnancy in women undiagnosed at conception, late ART start in women diagnosed but untreated at conception and detectable viral load (VL) at delivery in women on antenatal ART. Late ART start in women diagnosed but not treated at conception may reflect challenges in retention in HIV care and suboptimal adherence,9–12 whilst a detectable VL at delivery is a surrogate indicator of MTCT risk and delayed ART start.13 We addressed these objectives using data from 2002 to 2012 in the European Pregnancy and Paediatric Cohort Collaboration (EPPICC) in EuroCoord, a large collaboration of HIV observational studies in pregnant women and children.14 Methods Study design and participants Four cohorts across 10 European countries participating in EPPICC provided data on pregnancies of women diagnosed with HIV before or during pregnancy. The western arm (which included Belgium, Denmark, Germany, Italy, Netherlands, Poland and Sweden) of the European Collaborative Study, the UK and Ireland National Study of HIV in Pregnancy and Childhood (NSHPC), NENEXP (Spain) and the Madrid Cohort of HIV-Infected Mother–Infant Pairs (Spain) submitted data on eligible pregnancies using a standardised format (modified HIV Collaboration Data Exchange Protocol, http://www.hicdep.org) to the coordinating centre. Each participating study centre was responsible for ensuring that ethics approval for the data merger and analysis was in place and for compliance with local and national data protection requirements. We included women who delivered a singleton live birth in the cohort countries between 2002 and 2012; if a woman had repeated pregnancies during the study period, we included only the first one. We excluded women whose country of birth was unknown (445). We defined a migrant as a woman who delivered in a country different from her country of birth and a native (NAT) as a woman who delivered in her country of birth. We grouped countries of birth into seven world regions based on UN grouping categories:15 Western Europe and other Western countries, including North America, Australia and New Zealand (WEWC), Eastern Europe (EE), North Africa and The Middle East (NAME), Sub-Saharan Africa (SSA), Latin America (LA), The Caribbean (CRB), Asia and Oceania (excluding Australia and New Zealand) (ASIA/OC). Statistical analysis We compared three markers of suboptimal PMTCT interventions in migrant and native women: (i) diagnosis in late pregnancy in women who were undiagnosed at conception; (ii) late ART start in women who were diagnosed and untreated at conception; (iii) detectable VL at delivery (up to 28 days before and 7 days after delivery) in women who were on ART in pregnancy and delivered at ≥37 gestation weeks (GW) to exclude pregnancies with short ART duration because of pre-term delivery. We used 200 copies/ml as the cut-off for VL detectability due to changing use of assays with different detection limits over the study period. For similar reasons (i.e. changing guidelines over time and across European countries9), we defined a diagnosis in late pregnancy as a diagnosis at ≥20 GW and a late ART start as initiation after 28 GW. We defined the first CD4 count measured in pregnancy as the baseline CD4 count and assumed a baseline CD4 count < 350 cells/µl at antenatal diagnosis to be an indicator of late presentation for HIV care and a CD4 count < 200 cells/µl an indicator of advanced HIV disease.16 We fitted logistic regression models to investigate whether maternal region of birth was a factor associated with the above-mentioned markers of suboptimal PMTCT intervention. We adjusted the analyses for other factors known to be associated with late antenatal care presentation:1 year of delivery (2002–06 vs. 2007–12), maternal age (in tertiles <28, 28–33, >33 years), parity at enrolment (no/yes), HIV transmission mode [injecting drug use (IDU), heterosexual, other/unknown] and country of delivery (Spain, UK/Ireland and other European countries). In addition, the analysis on VL at delivery included a priori the following factors associated with VL suppression:17–19 time of ART initiation in pregnancy, baseline CD4 count and combination ART (≥three ART drugs) vs. a mono-dual ART (<3 ART drugs) regimen. Chi-square tests were performed to assess whether differences in proportions between native and migrant women were statistically significant (P < 0.05). Likelihood ratio test was used to determine if transmission category modified the association between diagnosis in late pregnancy and being a migrant woman. Statistical analyses were carried out using STATA v13.1 software (Stata Corp, College Station, TX, USA). Results Characteristics of participants There were 11 795 women who delivered a singleton live birth between 2002 and 2012. Of these women, 1188 (10.1%) delivered in Spain, 9317 (79.0%) in UK/Ireland and 1290 (10.9%) in other European countries; 9421 (79.9%) were migrants. The proportion of migrant women increased from 76.3% in 2002–06 to 83.8% in 2007–12. The most prevalent region of birth was SSA (8151, 69.1%); other regions of birth represented <3% each of all women but these proportions varied across European countries. In Spain, 20.7% of women were from SSA, 9.8% from LA and 62.9% were native. Maternal characteristics by world region of birth are presented in table 1. Table 1 Maternal characteristics of all women by world region of birtha   NAT  WEWC  EE  NAME  SSA  LA  CRB  ASIA/OC  Total    n (%)  n (%)  n (%)  n (%)  n (%)  n (%)  n (%)  n (%)  n (%)    2374 (20.1)  274 (2.3)  188 (1.6)  79 (0.7)  8151 (69.1)  198 (1.7)  252 (2.1)  279 (2.7)  11 795 (100.0)  Country of delivery                      Spain  747 (31.5)  5 (1.8)  27 (14.4)  21 (26.6)  246 (3.0)  116 (58.6)  22 (8.7)  4 (1.4)  1188 (10.1)      UK/Ireland  1235 (52.0)  248 (90.5)  109 (58.0)  31 (39.2)  7194 (88.3)  39 (19.7)  226 (89.7)  235 (84.2)  9317 (79.0)      Otherb  392 (16.5)  21 (7.7)  52 (27.7)  27 (34.2)  711 (8.7)  43 (21.7)  4 (1.6)  40 (14.3)  1290 (10.9)  Year of delivery                      2002-2006  1466 (61.8)  116 (42.3)  70 (37.2)  49 (62.0)  4091 (50.2)  110 (55.6)  134 (53.2)  141 (50.5)  6177 (52.4)      2007-2012  908 (38.3)  158 (57.7)  118 (62.8)  30 (38.0)  4060 (49.8)  88 (44.4)  118 (46.8)  138 (49.5)  5618 (47.6)  Parity (n = 11 652)                      Yes  1205 (52.2)  114 (41.8)  59 (33.0)  40 (52.6)  4430 (54.7)  110 (59.8)  158 (63.5)  131 (47.5)  6247 (53.6)  Number of ART drugs in pregnancy                  <3 ART  280 (11.8)  23 (8.4)  19 (10.1)  6 (7.6)  594 (7.3)  16 (8.1)  23 (9.1)  16 (5.7)  977 (8.3)      ≥3 ART  1939 (81.7)  242 (88.3)  154 (81.9)  67 (84.8)  7224 (88.6)  167 (84.3)  220 (87.3)  251 (90.0)  10 264 (87.0)      Unknown  155 (6.5)  9 (3.3)  15 (8.0)  6 (7.6)  333 (4.1)  15 (7.6)  9 (3.6)  12 (4.3)  554 (4.7)  ART at conception (n =11 245)                  No  1469 (68.1  193 (72.3)  136 (75.6)  56 (73.7)  5616 (71.3)  123 (69.1)  197 (80.1)  193 (71.8)  7983 (71.0)      Yes  688 (31.9)  74 (27.7)  44 (24.4)  20 (26.3)  2256 (28.7)  55 (30.9)  49 (19.9)  76 (28.3)  3262 (29.0)  CD4 (cells/µl) at baseline (n = 10 121)              <350  552 (31.7)  65 (26.2)  60 (37.3)  23 (34.9)  3324 (45.6)  51 (39.5)  67 (28.3)  113 (44.8)  4255 (42.0)  <200  165 (9.5)  21 (8.5)  13 (8.1)  10 (15.2)  1179 (16.2)  14 (10.9)  17 (7.2)  39 (15.5)  1458 (14.4)    NAT  WEWC  EE  NAME  SSA  LA  CRB  ASIA/OC  Total    n (%)  n (%)  n (%)  n (%)  n (%)  n (%)  n (%)  n (%)  n (%)    2374 (20.1)  274 (2.3)  188 (1.6)  79 (0.7)  8151 (69.1)  198 (1.7)  252 (2.1)  279 (2.7)  11 795 (100.0)  Country of delivery                      Spain  747 (31.5)  5 (1.8)  27 (14.4)  21 (26.6)  246 (3.0)  116 (58.6)  22 (8.7)  4 (1.4)  1188 (10.1)      UK/Ireland  1235 (52.0)  248 (90.5)  109 (58.0)  31 (39.2)  7194 (88.3)  39 (19.7)  226 (89.7)  235 (84.2)  9317 (79.0)      Otherb  392 (16.5)  21 (7.7)  52 (27.7)  27 (34.2)  711 (8.7)  43 (21.7)  4 (1.6)  40 (14.3)  1290 (10.9)  Year of delivery                      2002-2006  1466 (61.8)  116 (42.3)  70 (37.2)  49 (62.0)  4091 (50.2)  110 (55.6)  134 (53.2)  141 (50.5)  6177 (52.4)      2007-2012  908 (38.3)  158 (57.7)  118 (62.8)  30 (38.0)  4060 (49.8)  88 (44.4)  118 (46.8)  138 (49.5)  5618 (47.6)  Parity (n = 11 652)                      Yes  1205 (52.2)  114 (41.8)  59 (33.0)  40 (52.6)  4430 (54.7)  110 (59.8)  158 (63.5)  131 (47.5)  6247 (53.6)  Number of ART drugs in pregnancy                  <3 ART  280 (11.8)  23 (8.4)  19 (10.1)  6 (7.6)  594 (7.3)  16 (8.1)  23 (9.1)  16 (5.7)  977 (8.3)      ≥3 ART  1939 (81.7)  242 (88.3)  154 (81.9)  67 (84.8)  7224 (88.6)  167 (84.3)  220 (87.3)  251 (90.0)  10 264 (87.0)      Unknown  155 (6.5)  9 (3.3)  15 (8.0)  6 (7.6)  333 (4.1)  15 (7.6)  9 (3.6)  12 (4.3)  554 (4.7)  ART at conception (n =11 245)                  No  1469 (68.1  193 (72.3)  136 (75.6)  56 (73.7)  5616 (71.3)  123 (69.1)  197 (80.1)  193 (71.8)  7983 (71.0)      Yes  688 (31.9)  74 (27.7)  44 (24.4)  20 (26.3)  2256 (28.7)  55 (30.9)  49 (19.9)  76 (28.3)  3262 (29.0)  CD4 (cells/µl) at baseline (n = 10 121)              <350  552 (31.7)  65 (26.2)  60 (37.3)  23 (34.9)  3324 (45.6)  51 (39.5)  67 (28.3)  113 (44.8)  4255 (42.0)  <200  165 (9.5)  21 (8.5)  13 (8.1)  10 (15.2)  1179 (16.2)  14 (10.9)  17 (7.2)  39 (15.5)  1458 (14.4)  a NAT, native population; WEWC, Western Europe and similar countries; EE, Eastern Europe; NAME, North Africa and the Middle East SSA-Sub-Saharan Africa; LA, Latin America; CRB, Caribbean; ASIA/OC, Asia and Oceania. b Belgium, Denmark, Germany, Italy, Netherlands, Poland, Sweden. Most women (9817, 83.2%) had acquired HIV heterosexually; among native women and women from WEWC, IDU acquisition was relatively common (21.1 and 16.8%, respectively). Overall, 152 women (1.3% of the whole dataset) were reported as being vertically infected, mainly native women (85) and women from SSA (55). Gestational age at delivery was available for 11 506 pregnancies, of which 14.8% (1704) were pre-term. Native women were more likely to deliver pre-term than migrant women (19.5 vs. 13.6%, P < 0.001). Between 2002 and 2012, the MTCT rate declined from 1.78% (95% CI 1.07, 2.78) to 0.70% (95% CI 0.23, 1.63). There was no difference between the unadjusted MTCT rates in migrant and native women [0.96% (95% CI 0.77, 1.17) vs. 1.22% (0.82, 1.75) respectively, P = 0.25]. Markers of suboptimal PMTCT interventions Of the 11 795 women, we identified 4797 women who were diagnosed in pregnancy; 31.9% (1529) of them were diagnosed at ≥20 GW, with this proportion varying by region of birth (table 2). We further identified 2253 women who were diagnosed and untreated at conception and who started ART during their pregnancy; 13.1% (295) of them started ART at >28 GW. Table 2 Markers of suboptimal PMTCT interventions in women with HIV by world region of birth   NAT  WEWC  EE  NAME  SSA  LA  CRB  ASIA/OC  Total  Women diagnosed in current pregnancy ( n = 4797)  Diagnosed at ≥ 20 GW  n = 663  n = 111  n = 89  n = 33  n = 3557  n = 74  n = 123  n = 147  n = 4797  162 (24.4%)  21 (18.9%)  31 (34.8%)  15 (45.5%)  1206 (33.9%)  24 (32.4%)  32 (26.0%)  38 (25.9%)  1529 (31.9%)  Women diagnosed but untreated at conception (n = 2253)  ART start >28 GW  n = 602  n = 60  n = 35  n = 16  n = 1423  n = 41  n = 50  n = 26  n = 2253  83 (13.8%)  6 (10.0%)  2 (5.7%)  2 (12.5%)  183 (12.9%)  7 (17.1%)  8 (16.0%)  4 (15.4%)  295 (13.1%)  Women on antenatal ART with VL at delivery ( n = 5323)a  >200 copies/ml  n = 863  n = 143  n = 84  n = 34  n = 3858  n = 81  n = 130  n = 130  n = 5323  133 (15.4%)  23 (16.1%)  13 (15.5%)  6 (17.7%)  579 (15.0%)  17 (21.0%)  17 (13.1%)  19 (14.6%)  807 (15.1%)    NAT  WEWC  EE  NAME  SSA  LA  CRB  ASIA/OC  Total  Women diagnosed in current pregnancy ( n = 4797)  Diagnosed at ≥ 20 GW  n = 663  n = 111  n = 89  n = 33  n = 3557  n = 74  n = 123  n = 147  n = 4797  162 (24.4%)  21 (18.9%)  31 (34.8%)  15 (45.5%)  1206 (33.9%)  24 (32.4%)  32 (26.0%)  38 (25.9%)  1529 (31.9%)  Women diagnosed but untreated at conception (n = 2253)  ART start >28 GW  n = 602  n = 60  n = 35  n = 16  n = 1423  n = 41  n = 50  n = 26  n = 2253  83 (13.8%)  6 (10.0%)  2 (5.7%)  2 (12.5%)  183 (12.9%)  7 (17.1%)  8 (16.0%)  4 (15.4%)  295 (13.1%)  Women on antenatal ART with VL at delivery ( n = 5323)a  >200 copies/ml  n = 863  n = 143  n = 84  n = 34  n = 3858  n = 81  n = 130  n = 130  n = 5323  133 (15.4%)  23 (16.1%)  13 (15.5%)  6 (17.7%)  579 (15.0%)  17 (21.0%)  17 (13.1%)  19 (14.6%)  807 (15.1%)  NAT, native; WEWC, Western Europe and similar countries; EE, Eastern Europe; NAME, North Africa and the Middle East; SSA, Sub-Saharan Africa; LA, Latin America; CRB, Caribbean; ASIA/OC, Asia and Oceania; GW, gestational weeks. a pre-term (<37 GW) deliveries excluded. Migrant women were more likely to be HIV-diagnosed during pregnancy than native women [4134/9421 (43.9%) vs. 663/2374 (27.9%), P < 0.001] (maternal characteristics in Supplementary table S1). At antenatal diagnosis migrant women were also more likely to have CD4 < 350/µl (48.6 vs. 29.5%, P < 0.001) and more likely to have CD4 < 200 cells/µl (20.0 vs. 9.3%, P < 0.001). The proportions of women with low CD4 count at antenatal HIV diagnosis by region of birth are given in figure 1. Figure 1 View largeDownload slide Percentage of women among those diagnosed in pregnancy having a baseline CD4 count below 350 cells/µl and below 200 cells/µl by world region of birth (n, total number of women diagnosed in pregnancy with available data for CD4 count; NAT, native; WEWC, Western Europe and similar countries; EE, Eastern Europe; NAME, North Africa and the Middle East; SSA, Sub-Saharan Africa; LA, Latin America; CRB, Caribbean; ASIA/OC, Asia and Oceania) Figure 1 View largeDownload slide Percentage of women among those diagnosed in pregnancy having a baseline CD4 count below 350 cells/µl and below 200 cells/µl by world region of birth (n, total number of women diagnosed in pregnancy with available data for CD4 count; NAT, native; WEWC, Western Europe and similar countries; EE, Eastern Europe; NAME, North Africa and the Middle East; SSA, Sub-Saharan Africa; LA, Latin America; CRB, Caribbean; ASIA/OC, Asia and Oceania) Native women were more likely to be diagnosed but untreated at conception than migrant women [602/2374 (25.4%) vs. 1651/9421 (17.5%); P < 0.001) (maternal characteristics in Supplementary table S2). HIV diagnosis in late pregnancy in women undiagnosed at conception Migrant women were more likely to be diagnosed in late pregnancy than native women [1367/4134 (33.1%) vs. 162/663 (24.4%), P < 0.001]. The proportion of those diagnosed in late pregnancy was highest among women from NAME (15/33, 45.5%) and lowest among women from WEWC (21/111, 18.9%) (table 2, Supplementary table S1). Analysis stratified by transmission mode (Pinteraction = 0.0087) and adjusted for calendar year, age at delivery, country of delivery and parity indicated that women from EE, NAME and SSA were more likely to be diagnosed in late pregnancy compared with native women (women who acquired HIV by heterosexual mode only—table 3). The risk increased if a woman was parous, younger than 28 years, delivered before 2007 and in Spain. The small number of migrant women who acquired HIV by IDU and were diagnosed in late pregnancy (n = 10) precluded further analysis. Table 3 Factors associated with being diagnosed at ≥ 20 GW rather than earlier in women diagnosed in pregnancy   Antenatal HIV diagnosis at < 20 GW  Antenatal HIV diagnosis at ≥ 20 GW  Crude OR (95% CI)  Adjusteda OR (95% CI) (Heterosexual only n = 4241)  Adjusteda OR (95% CI) (IDU only n = 81)    n (%)  n (%)          3268 (68.1)  1529 (31.9%)        World region of birth            NAT  501 (75.6)  162 (24.3)  1.00  1.00  1.00      WEW  90 (81.1)  21 (18.9)  0.72 (0.43,1.20)  1.22 (0.69,2.16)  0.31 (0.05, 2.00)      EE  58 (65.2)  31 (34.8)  1.65 (1.03,2.65)  2.00 (1.11,3.62)  1.39 (0.23, 8.52)      NAME  18 (54.6)  15 (45.5)  2.58 (1.27,5.23)  3.16 (1.43,6.99)  –      SSA  2351 (66.1)  1206 (33.9)  1.59 (1.31,1.92)  2.12 (1.67,2.69)  0.49 (0.10, 2.45)      LA  50 (67.6)  24 (32.4)  1.48 (0.88,2.49)  1.56 (0.86,2.84)  –      CRB  91 (74.0)  32 (26.0)  1.09 (0.70,1.69)  1.21 (0.75,1.95)  –      ASIA/OC  109 (74.2)  38 (25.9)  1.08 (0.72,1.62)  1.45 (0.90,2.33)  –  Country of delivery              UK/Ireland  2861 (68.9)  1294 (31.4)  1.00  1.00  1.00      Spain  143 (57.4)  106 (42.6)  1.64 (1.26,2.12)  1.76 (1.24,2.51)  1.50 (0.33, 6.81)      Other EU  264 (67.2)  129 (32.8)  1.08 (0.87,2.12)  0.89 (0.67,1.20)  1.13 (0.20, 6.41)  Year of delivery              2002–06  1722 (62.5)  1033 (37.5)  1.00  1.00  1.00      2007–12  1546 (75.7)  496 (24.3)  0.53 (0.47,0.61)  0.88 (0.85,0.90)  0.90 (0.72, 1.12)  Maternal age at delivery (years)            <28  1252 (65.3)  665 (34.7)  1.28 (1.10,1.49)  1.38 (1.18,1.62)  1.02 (0.35, 2.96)      28–32  1131 (69.9)  486 (30.1)  1.03 (0.88,1.49)  1.00  1.00      >32  867 (70.7)  360 (29.3)  1.00  1.01 (0.85,1.21)  1.63 (0.42, 6.38)  Parity                Nulliparous  1810 (70.7)  750 (29.3)  1.00  1.00  1.00      Parous  1442 (65.2)  770 (34.8)  1.29 (1.14,1.46)  1.37 (1.19,1.58)  2.27 (0.77, 6.71)    Antenatal HIV diagnosis at < 20 GW  Antenatal HIV diagnosis at ≥ 20 GW  Crude OR (95% CI)  Adjusteda OR (95% CI) (Heterosexual only n = 4241)  Adjusteda OR (95% CI) (IDU only n = 81)    n (%)  n (%)          3268 (68.1)  1529 (31.9%)        World region of birth            NAT  501 (75.6)  162 (24.3)  1.00  1.00  1.00      WEW  90 (81.1)  21 (18.9)  0.72 (0.43,1.20)  1.22 (0.69,2.16)  0.31 (0.05, 2.00)      EE  58 (65.2)  31 (34.8)  1.65 (1.03,2.65)  2.00 (1.11,3.62)  1.39 (0.23, 8.52)      NAME  18 (54.6)  15 (45.5)  2.58 (1.27,5.23)  3.16 (1.43,6.99)  –      SSA  2351 (66.1)  1206 (33.9)  1.59 (1.31,1.92)  2.12 (1.67,2.69)  0.49 (0.10, 2.45)      LA  50 (67.6)  24 (32.4)  1.48 (0.88,2.49)  1.56 (0.86,2.84)  –      CRB  91 (74.0)  32 (26.0)  1.09 (0.70,1.69)  1.21 (0.75,1.95)  –      ASIA/OC  109 (74.2)  38 (25.9)  1.08 (0.72,1.62)  1.45 (0.90,2.33)  –  Country of delivery              UK/Ireland  2861 (68.9)  1294 (31.4)  1.00  1.00  1.00      Spain  143 (57.4)  106 (42.6)  1.64 (1.26,2.12)  1.76 (1.24,2.51)  1.50 (0.33, 6.81)      Other EU  264 (67.2)  129 (32.8)  1.08 (0.87,2.12)  0.89 (0.67,1.20)  1.13 (0.20, 6.41)  Year of delivery              2002–06  1722 (62.5)  1033 (37.5)  1.00  1.00  1.00      2007–12  1546 (75.7)  496 (24.3)  0.53 (0.47,0.61)  0.88 (0.85,0.90)  0.90 (0.72, 1.12)  Maternal age at delivery (years)            <28  1252 (65.3)  665 (34.7)  1.28 (1.10,1.49)  1.38 (1.18,1.62)  1.02 (0.35, 2.96)      28–32  1131 (69.9)  486 (30.1)  1.03 (0.88,1.49)  1.00  1.00      >32  867 (70.7)  360 (29.3)  1.00  1.01 (0.85,1.21)  1.63 (0.42, 6.38)  Parity                Nulliparous  1810 (70.7)  750 (29.3)  1.00  1.00  1.00      Parous  1442 (65.2)  770 (34.8)  1.29 (1.14,1.46)  1.37 (1.19,1.58)  2.27 (0.77, 6.71)  GW, gestational weeks; IDU, injecting drug use; NAT, native population; WEWC, Western Europe and similar countries; EE, Eastern Europe; NAME, North Africa and the Middle East; SSA-Sub-Saharan Africa; LA, Latin America; CAR, Caribbean; ASIA/OC, Asia and Oceania. a Odds ratios adjusted for world region of birth, country of delivery, year of delivery, maternal age at delivery and parity and stratified by heterosexual mode and IDU (intravenous drug use) mode of HIV acquisition (Pinteraction = 0.0087). Late ART start in women diagnosed but untreated at conception ART was started in late pregnancy in 13.8% (83/602) of native and 12.8% (212/1651) of migrant women table 2. In an analysis of 2193 women with full-term deliveries, adjusted for country of delivery, year of delivery, maternal age at delivery, mode of HIV transmission and parity, delivering before 2007 (OR 2.58, 95% CI 1.94, 3.42) and being parous (OR 1.45, 95% CI 1.11, 1.90) were associated with late ART start but not being a migrant from SSA (OR 0.98, 95% CI 0.69, 1.37) or from other world regions (OR 0.90, 95% CI 0.54, 1.48) (Supplementary table S3). In a sensitivity analysis including only women who acquired HIV heterosexually, being a migrant was not associated with late treatment initiation [OR (women from SSA vs. native women) 1.11, 95% CI 0.75, 1.64]. VL at delivery Data on VL at delivery were available for 5323 (56.6%) of women who met our inclusion criteria (full-term delivery and on antenatal ART, n = 9406). Women who were native, delivered before 2007 and/or in Spain and/or started ART in early pregnancy were less likely to have data on VL at delivery. Of these 5323 women, delivery VL was above 200 copies/ml in 807 (15.1%) women, comprising 674/4460 (15.1%) migrant and 133/863 (15.4%) native (Supplementary table S4). As expected, multivariable analysis suggested that late ART start in pregnancy was the strongest factor associated with having a detectable VL at delivery. Other factors associated with increased risk of non-suppressed VL included delivering between 2002 and 2006 rather than 2007–12, having a baseline CD4 count < 350 cells/µl and use of mono/dual ART rather than cART (Supplementary table S4) but not being a migrant. Sensitivity analyses restricted to women starting treatment in pregnancy (n = 3705) confirmed the main analysis findings that being a migrant was not associated with a detectable VL at delivery (OR 1.06, 95% CI 0.79, 1.43) even after including only women who acquired HIV heterosexually (OR 1.20, 95% CI 0.86, 1.68). Among the sub-group of women starting ART late, 36% (271/754) of migrants delivered with a detectable VL vs. 39% (42/109) of native women (P = 0.60). Discussion Our study has identified a large proportion of HIV-positive women in Europe delivering a live birth in 2002–12 who were migrants, of whom 44% were diagnosed with HIV in their current pregnancy. At antenatal diagnosis 49% of migrant women had CD4 < 350 cells/µl compared with 29.5% of native women suggesting that pregnancy is an important opportunity for undiagnosed migrant women to learn their HIV status.20 Reassuringly, our analyses suggest that between 2002 and 2012 (our study period), PMTCT interventions have improved in both migrant and native women. Once on ART, migrant women had a good virological response and we did not observe any difference in the crude MTCT rate between migrant and native women. However, we found that migrant women, particularly those from SSA, NAME and EE were more likely to be diagnosed in late pregnancy but we did not identify delayed ART initiation among migrant women who were already diagnosed at conception but not on ART. The observed differences between migrant and native women with respect to late antenatal HIV diagnosis partly agree with previous studies conducted in France21 and Italy,22,23 which reported that migrant women were more likely to receive late HIV screening and/or inappropriate antenatal care compared with native women, with higher rates of MTCT in migrant women. Although we found that migrant women were less likely to know their HIV status at conception, more likely to be diagnosed late in pregnancy and more likely to be diagnosed with low CD4 counts, there was no difference in the proportion achieving undetectable VL by delivery between native and migrant women among those receiving antenatal ART and delivering at term. This suggests that duration of antenatal ART was sufficient to reduce MTCT risk in these migrant women, consistent with the earlier French study where there was no difference in rates of uncontrolled viremia at delivery between African- and French-born women in the ART era.21 Although outside the scope of this paper, it is also likely that among those mother–infant pairs with detectable maternal VL at delivery other PMTCT interventions were successfully applied, given the similar and low MTCT rates in migrant and native women. Antenatal HIV screening is the cornerstone of PMTCT, with timely identification of previously undiagnosed women allowing application of optimum interventions. We showed that, among women with unknown HIV status at conception, migrant women from EE, NAME and SSA were more likely to be diagnosed after 20 weeks gestation than native women after accounting for other factors associated with late antenatal care presentation.7 Although few Western European countries have routine repeat HIV testing for women with negative screens at antenatal booking, it is possible that some women in this ‘late screen’ group acquired HIV in pregnancy and were diagnosed based on a second HIV test. Another finding was a higher rate of HIV diagnosis with severe immunosuppression among migrant women implying missing opportunities for earlier diagnosis and treatment. A recent Italian study has also reported similar findings, with women from SSA more likely to present with advanced HIV disease in pregnancy.24 Previous studies have identified barriers to accessing HIV testing and care experienced by migrants, such as time or financial constraints, language and cultural barriers and living and working conditions.4,5,25–27 In Europe, a large and increasing proportion of pregnant women with HIV are already aware of their HIV status at conception.1,28,29 Retention of patients in HIV care can be challenging, particularly postnatally, and a subsequent pregnancy often provides the means to re-establish a woman in HIV care.17 We found that most diagnosed women not on treatment at conception started ART in the first or second trimester, and report no difference in risk of starting ART late between migrant and native women. Although we showed a nearly 6-fold increased risk of delivery with detectable VL in women who were not on ART until the third trimester, this was driven by women delivering in the early years (2002–07) reflecting changing guidelines. Another reassuring finding was a significant decline in risk of late ART start over calendar time. Western Europe has recorded low MTCT rates for many years, with continuing declines driven by high uptake of antenatal screening, a large and increasing proportion of women on suppressive ART at conception, and prompt start of ART in pregnancy for untreated women. However, socio-economic disparities in access to health care exist,30 and overall low MTCT rates do not preclude the existence of sub-groups at increased risk of poor outcomes, such as the one in six women here with non-suppressed VL at delivery or those who delivered pre-term. Given the benefits of early ART start for health and survival,31,32,34 the considerable number of migrant women in Europe who access antenatal and HIV care services late is of public health concern.6 Our study had some limitations. We excluded women whose country of birth was unknown. As there is no universally agreed definition of the term ‘migrant’ we used country of birth (as reported by the mother) to determine whether a woman was a migrant, but we did not attempt to distinguish between country of birth, nationality or ethnic group. We did not consider the legal status of a migrant, time of arrival in the country of delivery or language spoken, which may affect the ability to access the local health care system, as such information was unavailable. Data were driven by the NSHPC, the largest dataset, and by women from SSA, who made up nearly 70% of the whole study population. We merged all women from SSA into one group, not acknowledging the rich cultural, ethnic and language diversity within SSA.7,33 Although poor socioeconomic status may delay HIV diagnosis and start of ART30 and increase risk of virological non-suppression in treated adults,35 we were unable to adjust for socio-economic factors. As this study is of women diagnosed before or during delivery only, we could not explore migrant women’s risk of delivering with undiagnosed HIV infection (e.g. due to testing decline or incident infection in pregnancy). Women from countries with a high HIV prevalence may remain at elevated risk of HIV acquisition post migration36,37 (e.g. 30% of SSA women living with HIV in France are estimated to have acquired HIV while in France37); highlighting the importance of HIV testing current partners of pregnant women as well as HIV prevention measures for all women including those from high prevalence settings. In conclusion, we have shown that although some migrant women were more likely to be diagnosed in late pregnancy, there was no overall difference between migrant and native women with respect to achieving an undetectable VL by delivery. Good access to antenatal care will not only enable the implementation of PMTCT protocols but also optimize both maternal and child health outcomes generally. Similarly, tailored screening programmes for migrant communities, whether antenatal or for the general population, are needed to help improve the cascade of HIV care and broaden HIV prevention programmes. Wider access to HIV testing and care would give migrant women an opportunity to know their HIV-status before conceiving and either reinforce prevention measures or start ART before conception thus optimizing PMTCT interventions and women’s health. Our findings indicate that particular attention should be given to facilitate access to services for migrant women with children. Supplementary data Supplementary data are available at EURPUB online. Funding This work was funded by the EU Seventh Framework Programme (FP7/2007-2013) under EuroCoord grant agreement n° 260694. The EPPICC Study Group: C. Thorne, H. Bailey; C. Giaquinto, O. Rampon, A. Mazza and A. De Rossi; I. Grosch Wörner; H.J. Scherpbier, M. Kreyenbroek, M.H. Godfried, F.J.B. Nellen and L. van Leeuwen; L. Navér, A.B. Bohlin, S. Lindgren, A. Kaldma, E. Belfrage; J. Levy, P. Barlow, Y. Manigart, M. Hainaut and T. Goetghebuer; B. Vandercam, B. Brichard, D. Van der Linden, H. Waterloos; C. Viscoli; A. De Maria; G. Bentivoglio, S. Ferrero, C. Gotta; N.H. Valerius, V. Rosenfeldt; V. Savasi, S. Fiore, M Crivelli; A Viganò, V. Giacomet, C. Cerini, C. Raimondi and G. Zuccotti; S. Alberico, G Maso, M. Tropea, V. Barresi,. ssa W. Buffolano, R. Tiseo, P. Martinelli, M. Sansone, G. Maruotti, A. Agangi; C. Tibaldi, S. Marini, G. Masuelli, C. Benedetto; M. Marczynska, S. Dobosz, J. Popielska, A. Oldakowska (European Collaborative Study—Western sites); J.T. Ramos, M.I. González-Tomé, M.L. Navarro, M.J. Mellado, L. Escosa, C. Fernández-McPhee, M.A. Roa, J Beceiro, I. Olabarrieta, E. Muñoz (Madrid Cohort of HIV-Infected Mother–Infant Pairs); A. Noguera-Julian, C. Fortuny-Guasch, P. Soler-Palacín, M. Coll, V. Pineda, Ll Mayol, M. Mendez, N. Rovira, T. Vallmanya, A. Mur, L. Garcia, N. Rius, O. Calavia (NENEXP); G. Favarato, K. Francis, A. Horn, H. Peters, R Sconza, C. Thorne, P. Tookey (NSHPC). Conflicts of interest: H.B. has received funding from the International AIDS Society and Public Health England. F.B. has received honoraria and a project grant from Gilead Sciences Europe Ltd. C.T. has received grant funding from EU FP7, UK Medical Research Council, AbbVie, Public Health England, PENTA Foundation and has carried out paid consultancies for UNICEF. For the remaining authors none were declared. Key points In Europe a large proportion of pregnant women with HIV are migrants—mainly from sub-Saharan Africa, albeit with some differences between European countries. 3When compared with pregnant native-born women, migrant women are less likely to know their HIV status at conception, more likely to be diagnosed in late pregnancy and at antenatal diagnosis, more likely to have low CD4 counts. Although some migrant women are more likely to be diagnosed in late pregnancy, among those on ART there is no overall difference between migrant and native women with respect to achieving an undetectable viral load by delivery, a proxy indicator of MTCT risk. Tailored screening programmes for migrant communities, whether antenatal or for the general population, are needed to help improve the cascade of HIV care and broaden HIV prevention programmes. References 1 Townsend CL, Byrne L, Cortina-Borja M, et al.   Earlier initiation of ART and further decline in mother-to-child HIV transmission rates, 2000-2011. AIDS  2014; 28: 1049– 57. Google Scholar CrossRef Search ADS PubMed  2 Mussini C, Antinori A, Bhagani S, et al.   European AIDS Clinical Society Standard of Care meeting on HIV and related coinfections: The Rome Statements. HIV Med  2016; 17: 445– 52. Google Scholar CrossRef Search ADS PubMed  3 Del Amo J, Likatavicius G, Perez-Cachafeiro S, et al.   The epidemiology of HIV and AIDS reports in migrants in the 27 European Union countries, Norway and Iceland: 1999-2006. Eur J Public Health  2011; 21: 620– 6. Google Scholar CrossRef Search ADS PubMed  4 Deblonde J, De Koker P, Hamers FF, et al.   Barriers to HIV testing in Europe: a systematic review. Eur J Public Health  2010; 20: 422– 32. Google Scholar CrossRef Search ADS PubMed  5 Deblonde J, Sasse A, Del Amo J, et al.   Restricted access to antiretroviral treatment for undocumented migrants: a bottle neck to control the HIV epidemic in the EU/EEA. BMC Public Health  2015; 15: 1228. Google Scholar CrossRef Search ADS PubMed  6 Hernando V, Alvarez-del Arco D, Alejos B, et al.   HIV Infection in Migrant Populations in the European Union and European Economic Area in 2007-2012: An Epidemic on the Move. J Acquir Immune Defic Syndr  2015; 70: 204– 11. Google Scholar CrossRef Search ADS PubMed  7 French CE, Thorne C, Byrne L, Cortina-Borja M, Tookey PA. Presentation for care and antenatal management of HIV in the UK, 2009-2014. HIV Med  2017; 18: 161– 70. Google Scholar CrossRef Search ADS PubMed  8 Ng R, Macdonald EM, Loutfy MR, et al.   Adequacy of prenatal care among women living with human immunodeficiency virus: a population-based study. BMC Public Health  2015; 15: 514. Google Scholar CrossRef Search ADS PubMed  9 Aebi-Popp K, Mulcahy F, Rudin C, et al.   National Guidelines for the prevention of mother-to-child transmission of HIV across Europe–how do countries differ?. Eur J Public Health  2013; 23: 1053– 8. Google Scholar CrossRef Search ADS PubMed  10 French CE, Thorne C, Tariq S, et al.   Immunologic status and virologic outcomes in repeat pregnancies to HIV-positive women not on antiretroviral therapy at conception: a case for lifelong antiretroviral therapy? AIDS  2014; 28: 1369. Google Scholar CrossRef Search ADS PubMed  11 Phillips T, Thebus E, Bekker L-G, et al.   Disengagement of HIV-positive pregnant and postpartum women from antiretroviral therapy services: a cohort study. J Int AIDS Soc  2014; 17: 12 McNairy ML, Teasdale CA, El-Sadr WM, et al.   Mother and child both matter: reconceptualizing the prevention of mother-to-child transmission care continuum. Curr Opin HIV AIDS  2015; 10: 403– 10. Google Scholar CrossRef Search ADS PubMed  13 European Collaborative Study. Time to undetectable viral load after highly active antiretroviral therapy initiation among HIV-infected pregnant women. Clin Infect Dis  2007; 1647– 56. 14 Chiappini E, Galli L, Giaquinto C, et al.   Use of combination neonatal prophylaxis for the prevention of mother-to-child transmission of HIV infection in European high-risk infants. AIDS  2013; 27: 991– 1000. Google Scholar CrossRef Search ADS PubMed  15 Monge S, Jarrin I, Mocroft A, et al.   Mortality in migrants living with HIV in western Europe (1997-2013): a collaborative cohort study. Lancet HIV  2015; 2: E540– 9. Google Scholar CrossRef Search ADS PubMed  16 Antinori A, Coenen T, Costagiola D, et al.   Late presentation of HIV infection: a consensus definition. HIV Med  2011; 12: 61– 4. Google Scholar CrossRef Search ADS PubMed  17 Aebi-Popp K, Mulcahy F, Glass TR, et al.   Missed opportunities among HIV-positive women to control viral replication during pregnancy and to have a vaginal delivery. J Acquir Immune Defic Syndr  2013; 64: 58– 65. Google Scholar CrossRef Search ADS PubMed  18 Floridia M, Ravizza M, Pinnetti C, et al.   Treatment change in pregnancy is a significant risk factor for detectable HIV-1 RNA in plasma at end of pregnancy. HIV Clin Trials  2010; 11: 303– 11. Google Scholar CrossRef Search ADS PubMed  19 Katz IT, Leister E, Kacanek D, et al.   Factors associated with lack of viral suppression at delivery among highly active antiretroviral therapy-naive women with HIV: a cohort study. Ann Intern Med  2015; 162: 90– 9. Google Scholar CrossRef Search ADS PubMed  20 Soriano-Arandes A, Noguera-Julian A, Lopez-Lacort M, et al.   Pregnancy as an opportunity to diagnose human-immunodeficiency virus immigrant women in Catalonia. Enferm Infec Microbiol Clin  2016; doi:10.1016/j.eimc.2016.07.011 21 Jasseron C, Mandelbrot L, Tubiana R, et al.   Prevention of mother-to-child HIV transmission: similar access for sub-Sahara African immigrants and for French women?. AIDS  2008; 22: 1503– 11. Google Scholar CrossRef Search ADS PubMed  22 Chiappini E, Galli L, Lisi C, et al.   Risk of perinatal HIV infection in infants born in Italy to immigrant mothers. Clin Infect Dis  2011; 53: 310– 3. Google Scholar CrossRef Search ADS PubMed  23 Izzo I, Forleo MA, Casari S, et al.   Maternal characteristics during pregnancy and risk factors for positive HIV RNA at delivery: a single-cohort observational study (Brescia, Northern Italy). BMC Public Health  2011; 11: 124. Google Scholar CrossRef Search ADS PubMed  24 Floridia M, Tamburrini E, Masuelli G, et al.   Brief report: consequences of presentation with advanced HIV disease in pregnancy: data from a National Study in Italy. J Acquir Immune Defic Syndr  2015; 70: 452– 5. Google Scholar CrossRef Search ADS PubMed  25 Burns FM, Imrie JY, Nazroo J, et al.   Why the(y) wait? Key informant understandings of factors contributing to late presentation and poor utilization of HIV health and social care services by African migrants in Britain. AIDS Care  2007; 19: 102– 8. Google Scholar CrossRef Search ADS PubMed  26 Burns FM, Johnson AM, Nazroo J, et al.   Missed opportunities for earlier HIV diagnosis within primary and secondary healthcare settings in the UK. AIDS  2008; 22: 115– 22. Google Scholar CrossRef Search ADS PubMed  27 Fakoya I, Reynolds R, Caswell G, Shiripinda I. Barriers to HIV testing for migrant black Africans in Western Europe. HIV Med  2008; 9(Suppl 2): 23– 5. Google Scholar CrossRef Search ADS PubMed  28 Briand N, Warszawski J, Mandelbrot L, et al.   Is intrapartum intravenous zidovudine for prevention of mother-to-child HIV-1 transmission still useful in the combination antiretroviral therapy era? Clin Infect Dis  2013; 57: 903– 14. Google Scholar CrossRef Search ADS PubMed  29 Reitter A, Stücker A, Linde R, et al.   Pregnancy complications in HIV‐positive women: 11‐year data from the Frankfurt HIV Cohort. HIV Med  2014; 15: 525– 36. Google Scholar CrossRef Search ADS PubMed  30 COHERE. Delayed HIV diagnosis and initiation of antiretroviral therapy: inequalities by educational level, COHERE in EuroCoord, Socio-economic - Inequalities HIV Writing Group for Collaboration of Observational HIV Epidemiological Research in Europe in EuroCoord. AIDS  2014; 28: 2297– 306. PubMed  31 Mocroft A, Lundgren JD, Sabin ML, et al.   Risk factors and outcomes for late presentation for HIV-positive persons in Europe: results from the Collaboration of Observational HIV Epidemiological Research Europe Study (COHERE). PLoS Med  2013; 10: e1001510. Google Scholar CrossRef Search ADS PubMed  32 TEMPRANO ANRS 12136 Study Group. A Trial of Early Antiretrovirals and Isoniazid Preventive Therapy in Africa. New England Journal of Medicine  2015; 373: 808– 22. CrossRef Search ADS PubMed  33 Tariq S, Elford J, Cortina-Borja M, et al.   Childhood. The association between ethnicity and late presentation to antenatal care among pregnant women living with HIV in the UK and Ireland. AIDS Care  2012; 24: 978– 85. Google Scholar CrossRef Search ADS PubMed  34 World Health Organisation. Consolidated Guidelines on the Use of Antiretroviral Drugs for Treating and Preventing HIV Infection: Recommendations for a Public Health Approach. 2013. Available at: http://apps.who.int/iris/bitstream/10665/85321/1/9789241505727_eng.pdf 35 Saunders P, Goodman A, Smith C, et al.   Does gender or mode of HIV acquisition affect virological response to modern antiretroviral therapy (ART)?. HIV Med  2016; 17: 18– 27. Google Scholar CrossRef Search ADS PubMed  36 Alvarez-del Arco D, Fakoya I, Monge S, et al.   HIV Acquisition among migrants living in Europe: Results from aMASE - Advancing Migrant Access to Health Services in Europe. 15th European AIDS Conference. Barcelona, 21–24 October, 2015. Abstract PS3/5 37 Desgrées-du-Loû A, Pannetier J, Ravalihasy A, et al.   Sub-Saharan African migrants living with HIV acquired after migration, France, ANRS PARCOURS study, 2012 to 2013. Eurosurveillance  2015; 20: © The Author 2017. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The European Journal of Public Health Oxford University Press

Migrant women living with HIV in Europe: are they facing inequalities in the prevention of mother-to-child-transmission of HIV?: The European Pregnancy and Paediatric HIV Cohort Collaboration (EPPICC) study group in EuroCoord*

Loading next page...
 
/lp/ou_press/migrant-women-living-with-hiv-in-europe-are-they-facing-inequalities-fzs0Jigq3J
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.
ISSN
1101-1262
eISSN
1464-360X
D.O.I.
10.1093/eurpub/ckx048
Publisher site
See Article on Publisher Site

Abstract

Abstract Background In pregnancy early interventions are recommended for prevention of mother-to-child-transmission (PMTCT) of HIV. We examined whether pregnant women who live with HIV in Europe and are migrants encounter barriers in accessing HIV testing and care. Methods Four cohorts within the European Pregnancy and Paediatric HIV Cohort Collaboration provided data for pooled analysis of 11 795 pregnant women who delivered in 2002–12 across ten European countries. We defined a migrant as a woman delivering in a country different from her country of birth and grouped the countries into seven world regions. We compared three suboptimal PMTCT interventions (HIV diagnosis in late pregnancy in women undiagnosed at conception, late anti-retroviral therapy (ART) start in women diagnosed but untreated at conception and detectable viral load (VL) at delivery in women on antenatal ART) in native and migrant women using multivariable logistic regression models. Results Data included 9421 (79.9%) migrant women, mainly from sub-Saharan Africa (SSA); 4134 migrant women were diagnosed in the current pregnancy, often (48.6%) presenting with CD4 count <350 cells/µl. Being a migrant was associated with HIV diagnosis in late pregnancy [OR for SSA vs. native women, 2.12 (95% CI 1.67, 2.69)] but not with late ART start if diagnosed but not on ART at conception, or with detectable VL at delivery once on ART. Conclusions Migrant women were more likely to be diagnosed in late pregnancy but once on ART virological response was good. Good access to antenatal care enables the implementation of PMTCT protocols and optimises both maternal and children health outcomes generally. Introduction In pregnant women with HIV, antenatal anti-retroviral therapy (ART) is crucial for viral suppression during pregnancy and at delivery and the prevention of mother-to-child-transmission (PMTCT).1 It is recommended that women conceiving on ART should continue ART during pregnancy and that those untreated at conception—i.e. women who are newly HIV-diagnosed in pregnancy and those already HIV-diagnosed but not treated—should start ART in early pregnancy.2 To ensure prompt antenatal ART start in women who are undiagnosed at conception it is recommended that HIV screening should be performed in early pregnancy. Migrant populations living in Europe are disproportionally affected by HIV3 but may face barriers in accessing HIV testing and care,4–6 including pregnant women. Migrant women may be less likely to receive optimum antenatal care than native women,7,8 potentially leading to delayed HIV diagnosis and delayed ART start with negative implications for mother-to-child-transmission (MTCT) risk and maternal health. We aimed to assess whether migrant women with HIV encounter barriers in accessing HIV testing and care during pregnancy using contemporary data across several European countries. To do so, we compared three markers of suboptimal PMTCT interventions in migrant and native women: HIV diagnosis in late pregnancy in women undiagnosed at conception, late ART start in women diagnosed but untreated at conception and detectable viral load (VL) at delivery in women on antenatal ART. Late ART start in women diagnosed but not treated at conception may reflect challenges in retention in HIV care and suboptimal adherence,9–12 whilst a detectable VL at delivery is a surrogate indicator of MTCT risk and delayed ART start.13 We addressed these objectives using data from 2002 to 2012 in the European Pregnancy and Paediatric Cohort Collaboration (EPPICC) in EuroCoord, a large collaboration of HIV observational studies in pregnant women and children.14 Methods Study design and participants Four cohorts across 10 European countries participating in EPPICC provided data on pregnancies of women diagnosed with HIV before or during pregnancy. The western arm (which included Belgium, Denmark, Germany, Italy, Netherlands, Poland and Sweden) of the European Collaborative Study, the UK and Ireland National Study of HIV in Pregnancy and Childhood (NSHPC), NENEXP (Spain) and the Madrid Cohort of HIV-Infected Mother–Infant Pairs (Spain) submitted data on eligible pregnancies using a standardised format (modified HIV Collaboration Data Exchange Protocol, http://www.hicdep.org) to the coordinating centre. Each participating study centre was responsible for ensuring that ethics approval for the data merger and analysis was in place and for compliance with local and national data protection requirements. We included women who delivered a singleton live birth in the cohort countries between 2002 and 2012; if a woman had repeated pregnancies during the study period, we included only the first one. We excluded women whose country of birth was unknown (445). We defined a migrant as a woman who delivered in a country different from her country of birth and a native (NAT) as a woman who delivered in her country of birth. We grouped countries of birth into seven world regions based on UN grouping categories:15 Western Europe and other Western countries, including North America, Australia and New Zealand (WEWC), Eastern Europe (EE), North Africa and The Middle East (NAME), Sub-Saharan Africa (SSA), Latin America (LA), The Caribbean (CRB), Asia and Oceania (excluding Australia and New Zealand) (ASIA/OC). Statistical analysis We compared three markers of suboptimal PMTCT interventions in migrant and native women: (i) diagnosis in late pregnancy in women who were undiagnosed at conception; (ii) late ART start in women who were diagnosed and untreated at conception; (iii) detectable VL at delivery (up to 28 days before and 7 days after delivery) in women who were on ART in pregnancy and delivered at ≥37 gestation weeks (GW) to exclude pregnancies with short ART duration because of pre-term delivery. We used 200 copies/ml as the cut-off for VL detectability due to changing use of assays with different detection limits over the study period. For similar reasons (i.e. changing guidelines over time and across European countries9), we defined a diagnosis in late pregnancy as a diagnosis at ≥20 GW and a late ART start as initiation after 28 GW. We defined the first CD4 count measured in pregnancy as the baseline CD4 count and assumed a baseline CD4 count < 350 cells/µl at antenatal diagnosis to be an indicator of late presentation for HIV care and a CD4 count < 200 cells/µl an indicator of advanced HIV disease.16 We fitted logistic regression models to investigate whether maternal region of birth was a factor associated with the above-mentioned markers of suboptimal PMTCT intervention. We adjusted the analyses for other factors known to be associated with late antenatal care presentation:1 year of delivery (2002–06 vs. 2007–12), maternal age (in tertiles <28, 28–33, >33 years), parity at enrolment (no/yes), HIV transmission mode [injecting drug use (IDU), heterosexual, other/unknown] and country of delivery (Spain, UK/Ireland and other European countries). In addition, the analysis on VL at delivery included a priori the following factors associated with VL suppression:17–19 time of ART initiation in pregnancy, baseline CD4 count and combination ART (≥three ART drugs) vs. a mono-dual ART (<3 ART drugs) regimen. Chi-square tests were performed to assess whether differences in proportions between native and migrant women were statistically significant (P < 0.05). Likelihood ratio test was used to determine if transmission category modified the association between diagnosis in late pregnancy and being a migrant woman. Statistical analyses were carried out using STATA v13.1 software (Stata Corp, College Station, TX, USA). Results Characteristics of participants There were 11 795 women who delivered a singleton live birth between 2002 and 2012. Of these women, 1188 (10.1%) delivered in Spain, 9317 (79.0%) in UK/Ireland and 1290 (10.9%) in other European countries; 9421 (79.9%) were migrants. The proportion of migrant women increased from 76.3% in 2002–06 to 83.8% in 2007–12. The most prevalent region of birth was SSA (8151, 69.1%); other regions of birth represented <3% each of all women but these proportions varied across European countries. In Spain, 20.7% of women were from SSA, 9.8% from LA and 62.9% were native. Maternal characteristics by world region of birth are presented in table 1. Table 1 Maternal characteristics of all women by world region of birtha   NAT  WEWC  EE  NAME  SSA  LA  CRB  ASIA/OC  Total    n (%)  n (%)  n (%)  n (%)  n (%)  n (%)  n (%)  n (%)  n (%)    2374 (20.1)  274 (2.3)  188 (1.6)  79 (0.7)  8151 (69.1)  198 (1.7)  252 (2.1)  279 (2.7)  11 795 (100.0)  Country of delivery                      Spain  747 (31.5)  5 (1.8)  27 (14.4)  21 (26.6)  246 (3.0)  116 (58.6)  22 (8.7)  4 (1.4)  1188 (10.1)      UK/Ireland  1235 (52.0)  248 (90.5)  109 (58.0)  31 (39.2)  7194 (88.3)  39 (19.7)  226 (89.7)  235 (84.2)  9317 (79.0)      Otherb  392 (16.5)  21 (7.7)  52 (27.7)  27 (34.2)  711 (8.7)  43 (21.7)  4 (1.6)  40 (14.3)  1290 (10.9)  Year of delivery                      2002-2006  1466 (61.8)  116 (42.3)  70 (37.2)  49 (62.0)  4091 (50.2)  110 (55.6)  134 (53.2)  141 (50.5)  6177 (52.4)      2007-2012  908 (38.3)  158 (57.7)  118 (62.8)  30 (38.0)  4060 (49.8)  88 (44.4)  118 (46.8)  138 (49.5)  5618 (47.6)  Parity (n = 11 652)                      Yes  1205 (52.2)  114 (41.8)  59 (33.0)  40 (52.6)  4430 (54.7)  110 (59.8)  158 (63.5)  131 (47.5)  6247 (53.6)  Number of ART drugs in pregnancy                  <3 ART  280 (11.8)  23 (8.4)  19 (10.1)  6 (7.6)  594 (7.3)  16 (8.1)  23 (9.1)  16 (5.7)  977 (8.3)      ≥3 ART  1939 (81.7)  242 (88.3)  154 (81.9)  67 (84.8)  7224 (88.6)  167 (84.3)  220 (87.3)  251 (90.0)  10 264 (87.0)      Unknown  155 (6.5)  9 (3.3)  15 (8.0)  6 (7.6)  333 (4.1)  15 (7.6)  9 (3.6)  12 (4.3)  554 (4.7)  ART at conception (n =11 245)                  No  1469 (68.1  193 (72.3)  136 (75.6)  56 (73.7)  5616 (71.3)  123 (69.1)  197 (80.1)  193 (71.8)  7983 (71.0)      Yes  688 (31.9)  74 (27.7)  44 (24.4)  20 (26.3)  2256 (28.7)  55 (30.9)  49 (19.9)  76 (28.3)  3262 (29.0)  CD4 (cells/µl) at baseline (n = 10 121)              <350  552 (31.7)  65 (26.2)  60 (37.3)  23 (34.9)  3324 (45.6)  51 (39.5)  67 (28.3)  113 (44.8)  4255 (42.0)  <200  165 (9.5)  21 (8.5)  13 (8.1)  10 (15.2)  1179 (16.2)  14 (10.9)  17 (7.2)  39 (15.5)  1458 (14.4)    NAT  WEWC  EE  NAME  SSA  LA  CRB  ASIA/OC  Total    n (%)  n (%)  n (%)  n (%)  n (%)  n (%)  n (%)  n (%)  n (%)    2374 (20.1)  274 (2.3)  188 (1.6)  79 (0.7)  8151 (69.1)  198 (1.7)  252 (2.1)  279 (2.7)  11 795 (100.0)  Country of delivery                      Spain  747 (31.5)  5 (1.8)  27 (14.4)  21 (26.6)  246 (3.0)  116 (58.6)  22 (8.7)  4 (1.4)  1188 (10.1)      UK/Ireland  1235 (52.0)  248 (90.5)  109 (58.0)  31 (39.2)  7194 (88.3)  39 (19.7)  226 (89.7)  235 (84.2)  9317 (79.0)      Otherb  392 (16.5)  21 (7.7)  52 (27.7)  27 (34.2)  711 (8.7)  43 (21.7)  4 (1.6)  40 (14.3)  1290 (10.9)  Year of delivery                      2002-2006  1466 (61.8)  116 (42.3)  70 (37.2)  49 (62.0)  4091 (50.2)  110 (55.6)  134 (53.2)  141 (50.5)  6177 (52.4)      2007-2012  908 (38.3)  158 (57.7)  118 (62.8)  30 (38.0)  4060 (49.8)  88 (44.4)  118 (46.8)  138 (49.5)  5618 (47.6)  Parity (n = 11 652)                      Yes  1205 (52.2)  114 (41.8)  59 (33.0)  40 (52.6)  4430 (54.7)  110 (59.8)  158 (63.5)  131 (47.5)  6247 (53.6)  Number of ART drugs in pregnancy                  <3 ART  280 (11.8)  23 (8.4)  19 (10.1)  6 (7.6)  594 (7.3)  16 (8.1)  23 (9.1)  16 (5.7)  977 (8.3)      ≥3 ART  1939 (81.7)  242 (88.3)  154 (81.9)  67 (84.8)  7224 (88.6)  167 (84.3)  220 (87.3)  251 (90.0)  10 264 (87.0)      Unknown  155 (6.5)  9 (3.3)  15 (8.0)  6 (7.6)  333 (4.1)  15 (7.6)  9 (3.6)  12 (4.3)  554 (4.7)  ART at conception (n =11 245)                  No  1469 (68.1  193 (72.3)  136 (75.6)  56 (73.7)  5616 (71.3)  123 (69.1)  197 (80.1)  193 (71.8)  7983 (71.0)      Yes  688 (31.9)  74 (27.7)  44 (24.4)  20 (26.3)  2256 (28.7)  55 (30.9)  49 (19.9)  76 (28.3)  3262 (29.0)  CD4 (cells/µl) at baseline (n = 10 121)              <350  552 (31.7)  65 (26.2)  60 (37.3)  23 (34.9)  3324 (45.6)  51 (39.5)  67 (28.3)  113 (44.8)  4255 (42.0)  <200  165 (9.5)  21 (8.5)  13 (8.1)  10 (15.2)  1179 (16.2)  14 (10.9)  17 (7.2)  39 (15.5)  1458 (14.4)  a NAT, native population; WEWC, Western Europe and similar countries; EE, Eastern Europe; NAME, North Africa and the Middle East SSA-Sub-Saharan Africa; LA, Latin America; CRB, Caribbean; ASIA/OC, Asia and Oceania. b Belgium, Denmark, Germany, Italy, Netherlands, Poland, Sweden. Most women (9817, 83.2%) had acquired HIV heterosexually; among native women and women from WEWC, IDU acquisition was relatively common (21.1 and 16.8%, respectively). Overall, 152 women (1.3% of the whole dataset) were reported as being vertically infected, mainly native women (85) and women from SSA (55). Gestational age at delivery was available for 11 506 pregnancies, of which 14.8% (1704) were pre-term. Native women were more likely to deliver pre-term than migrant women (19.5 vs. 13.6%, P < 0.001). Between 2002 and 2012, the MTCT rate declined from 1.78% (95% CI 1.07, 2.78) to 0.70% (95% CI 0.23, 1.63). There was no difference between the unadjusted MTCT rates in migrant and native women [0.96% (95% CI 0.77, 1.17) vs. 1.22% (0.82, 1.75) respectively, P = 0.25]. Markers of suboptimal PMTCT interventions Of the 11 795 women, we identified 4797 women who were diagnosed in pregnancy; 31.9% (1529) of them were diagnosed at ≥20 GW, with this proportion varying by region of birth (table 2). We further identified 2253 women who were diagnosed and untreated at conception and who started ART during their pregnancy; 13.1% (295) of them started ART at >28 GW. Table 2 Markers of suboptimal PMTCT interventions in women with HIV by world region of birth   NAT  WEWC  EE  NAME  SSA  LA  CRB  ASIA/OC  Total  Women diagnosed in current pregnancy ( n = 4797)  Diagnosed at ≥ 20 GW  n = 663  n = 111  n = 89  n = 33  n = 3557  n = 74  n = 123  n = 147  n = 4797  162 (24.4%)  21 (18.9%)  31 (34.8%)  15 (45.5%)  1206 (33.9%)  24 (32.4%)  32 (26.0%)  38 (25.9%)  1529 (31.9%)  Women diagnosed but untreated at conception (n = 2253)  ART start >28 GW  n = 602  n = 60  n = 35  n = 16  n = 1423  n = 41  n = 50  n = 26  n = 2253  83 (13.8%)  6 (10.0%)  2 (5.7%)  2 (12.5%)  183 (12.9%)  7 (17.1%)  8 (16.0%)  4 (15.4%)  295 (13.1%)  Women on antenatal ART with VL at delivery ( n = 5323)a  >200 copies/ml  n = 863  n = 143  n = 84  n = 34  n = 3858  n = 81  n = 130  n = 130  n = 5323  133 (15.4%)  23 (16.1%)  13 (15.5%)  6 (17.7%)  579 (15.0%)  17 (21.0%)  17 (13.1%)  19 (14.6%)  807 (15.1%)    NAT  WEWC  EE  NAME  SSA  LA  CRB  ASIA/OC  Total  Women diagnosed in current pregnancy ( n = 4797)  Diagnosed at ≥ 20 GW  n = 663  n = 111  n = 89  n = 33  n = 3557  n = 74  n = 123  n = 147  n = 4797  162 (24.4%)  21 (18.9%)  31 (34.8%)  15 (45.5%)  1206 (33.9%)  24 (32.4%)  32 (26.0%)  38 (25.9%)  1529 (31.9%)  Women diagnosed but untreated at conception (n = 2253)  ART start >28 GW  n = 602  n = 60  n = 35  n = 16  n = 1423  n = 41  n = 50  n = 26  n = 2253  83 (13.8%)  6 (10.0%)  2 (5.7%)  2 (12.5%)  183 (12.9%)  7 (17.1%)  8 (16.0%)  4 (15.4%)  295 (13.1%)  Women on antenatal ART with VL at delivery ( n = 5323)a  >200 copies/ml  n = 863  n = 143  n = 84  n = 34  n = 3858  n = 81  n = 130  n = 130  n = 5323  133 (15.4%)  23 (16.1%)  13 (15.5%)  6 (17.7%)  579 (15.0%)  17 (21.0%)  17 (13.1%)  19 (14.6%)  807 (15.1%)  NAT, native; WEWC, Western Europe and similar countries; EE, Eastern Europe; NAME, North Africa and the Middle East; SSA, Sub-Saharan Africa; LA, Latin America; CRB, Caribbean; ASIA/OC, Asia and Oceania; GW, gestational weeks. a pre-term (<37 GW) deliveries excluded. Migrant women were more likely to be HIV-diagnosed during pregnancy than native women [4134/9421 (43.9%) vs. 663/2374 (27.9%), P < 0.001] (maternal characteristics in Supplementary table S1). At antenatal diagnosis migrant women were also more likely to have CD4 < 350/µl (48.6 vs. 29.5%, P < 0.001) and more likely to have CD4 < 200 cells/µl (20.0 vs. 9.3%, P < 0.001). The proportions of women with low CD4 count at antenatal HIV diagnosis by region of birth are given in figure 1. Figure 1 View largeDownload slide Percentage of women among those diagnosed in pregnancy having a baseline CD4 count below 350 cells/µl and below 200 cells/µl by world region of birth (n, total number of women diagnosed in pregnancy with available data for CD4 count; NAT, native; WEWC, Western Europe and similar countries; EE, Eastern Europe; NAME, North Africa and the Middle East; SSA, Sub-Saharan Africa; LA, Latin America; CRB, Caribbean; ASIA/OC, Asia and Oceania) Figure 1 View largeDownload slide Percentage of women among those diagnosed in pregnancy having a baseline CD4 count below 350 cells/µl and below 200 cells/µl by world region of birth (n, total number of women diagnosed in pregnancy with available data for CD4 count; NAT, native; WEWC, Western Europe and similar countries; EE, Eastern Europe; NAME, North Africa and the Middle East; SSA, Sub-Saharan Africa; LA, Latin America; CRB, Caribbean; ASIA/OC, Asia and Oceania) Native women were more likely to be diagnosed but untreated at conception than migrant women [602/2374 (25.4%) vs. 1651/9421 (17.5%); P < 0.001) (maternal characteristics in Supplementary table S2). HIV diagnosis in late pregnancy in women undiagnosed at conception Migrant women were more likely to be diagnosed in late pregnancy than native women [1367/4134 (33.1%) vs. 162/663 (24.4%), P < 0.001]. The proportion of those diagnosed in late pregnancy was highest among women from NAME (15/33, 45.5%) and lowest among women from WEWC (21/111, 18.9%) (table 2, Supplementary table S1). Analysis stratified by transmission mode (Pinteraction = 0.0087) and adjusted for calendar year, age at delivery, country of delivery and parity indicated that women from EE, NAME and SSA were more likely to be diagnosed in late pregnancy compared with native women (women who acquired HIV by heterosexual mode only—table 3). The risk increased if a woman was parous, younger than 28 years, delivered before 2007 and in Spain. The small number of migrant women who acquired HIV by IDU and were diagnosed in late pregnancy (n = 10) precluded further analysis. Table 3 Factors associated with being diagnosed at ≥ 20 GW rather than earlier in women diagnosed in pregnancy   Antenatal HIV diagnosis at < 20 GW  Antenatal HIV diagnosis at ≥ 20 GW  Crude OR (95% CI)  Adjusteda OR (95% CI) (Heterosexual only n = 4241)  Adjusteda OR (95% CI) (IDU only n = 81)    n (%)  n (%)          3268 (68.1)  1529 (31.9%)        World region of birth            NAT  501 (75.6)  162 (24.3)  1.00  1.00  1.00      WEW  90 (81.1)  21 (18.9)  0.72 (0.43,1.20)  1.22 (0.69,2.16)  0.31 (0.05, 2.00)      EE  58 (65.2)  31 (34.8)  1.65 (1.03,2.65)  2.00 (1.11,3.62)  1.39 (0.23, 8.52)      NAME  18 (54.6)  15 (45.5)  2.58 (1.27,5.23)  3.16 (1.43,6.99)  –      SSA  2351 (66.1)  1206 (33.9)  1.59 (1.31,1.92)  2.12 (1.67,2.69)  0.49 (0.10, 2.45)      LA  50 (67.6)  24 (32.4)  1.48 (0.88,2.49)  1.56 (0.86,2.84)  –      CRB  91 (74.0)  32 (26.0)  1.09 (0.70,1.69)  1.21 (0.75,1.95)  –      ASIA/OC  109 (74.2)  38 (25.9)  1.08 (0.72,1.62)  1.45 (0.90,2.33)  –  Country of delivery              UK/Ireland  2861 (68.9)  1294 (31.4)  1.00  1.00  1.00      Spain  143 (57.4)  106 (42.6)  1.64 (1.26,2.12)  1.76 (1.24,2.51)  1.50 (0.33, 6.81)      Other EU  264 (67.2)  129 (32.8)  1.08 (0.87,2.12)  0.89 (0.67,1.20)  1.13 (0.20, 6.41)  Year of delivery              2002–06  1722 (62.5)  1033 (37.5)  1.00  1.00  1.00      2007–12  1546 (75.7)  496 (24.3)  0.53 (0.47,0.61)  0.88 (0.85,0.90)  0.90 (0.72, 1.12)  Maternal age at delivery (years)            <28  1252 (65.3)  665 (34.7)  1.28 (1.10,1.49)  1.38 (1.18,1.62)  1.02 (0.35, 2.96)      28–32  1131 (69.9)  486 (30.1)  1.03 (0.88,1.49)  1.00  1.00      >32  867 (70.7)  360 (29.3)  1.00  1.01 (0.85,1.21)  1.63 (0.42, 6.38)  Parity                Nulliparous  1810 (70.7)  750 (29.3)  1.00  1.00  1.00      Parous  1442 (65.2)  770 (34.8)  1.29 (1.14,1.46)  1.37 (1.19,1.58)  2.27 (0.77, 6.71)    Antenatal HIV diagnosis at < 20 GW  Antenatal HIV diagnosis at ≥ 20 GW  Crude OR (95% CI)  Adjusteda OR (95% CI) (Heterosexual only n = 4241)  Adjusteda OR (95% CI) (IDU only n = 81)    n (%)  n (%)          3268 (68.1)  1529 (31.9%)        World region of birth            NAT  501 (75.6)  162 (24.3)  1.00  1.00  1.00      WEW  90 (81.1)  21 (18.9)  0.72 (0.43,1.20)  1.22 (0.69,2.16)  0.31 (0.05, 2.00)      EE  58 (65.2)  31 (34.8)  1.65 (1.03,2.65)  2.00 (1.11,3.62)  1.39 (0.23, 8.52)      NAME  18 (54.6)  15 (45.5)  2.58 (1.27,5.23)  3.16 (1.43,6.99)  –      SSA  2351 (66.1)  1206 (33.9)  1.59 (1.31,1.92)  2.12 (1.67,2.69)  0.49 (0.10, 2.45)      LA  50 (67.6)  24 (32.4)  1.48 (0.88,2.49)  1.56 (0.86,2.84)  –      CRB  91 (74.0)  32 (26.0)  1.09 (0.70,1.69)  1.21 (0.75,1.95)  –      ASIA/OC  109 (74.2)  38 (25.9)  1.08 (0.72,1.62)  1.45 (0.90,2.33)  –  Country of delivery              UK/Ireland  2861 (68.9)  1294 (31.4)  1.00  1.00  1.00      Spain  143 (57.4)  106 (42.6)  1.64 (1.26,2.12)  1.76 (1.24,2.51)  1.50 (0.33, 6.81)      Other EU  264 (67.2)  129 (32.8)  1.08 (0.87,2.12)  0.89 (0.67,1.20)  1.13 (0.20, 6.41)  Year of delivery              2002–06  1722 (62.5)  1033 (37.5)  1.00  1.00  1.00      2007–12  1546 (75.7)  496 (24.3)  0.53 (0.47,0.61)  0.88 (0.85,0.90)  0.90 (0.72, 1.12)  Maternal age at delivery (years)            <28  1252 (65.3)  665 (34.7)  1.28 (1.10,1.49)  1.38 (1.18,1.62)  1.02 (0.35, 2.96)      28–32  1131 (69.9)  486 (30.1)  1.03 (0.88,1.49)  1.00  1.00      >32  867 (70.7)  360 (29.3)  1.00  1.01 (0.85,1.21)  1.63 (0.42, 6.38)  Parity                Nulliparous  1810 (70.7)  750 (29.3)  1.00  1.00  1.00      Parous  1442 (65.2)  770 (34.8)  1.29 (1.14,1.46)  1.37 (1.19,1.58)  2.27 (0.77, 6.71)  GW, gestational weeks; IDU, injecting drug use; NAT, native population; WEWC, Western Europe and similar countries; EE, Eastern Europe; NAME, North Africa and the Middle East; SSA-Sub-Saharan Africa; LA, Latin America; CAR, Caribbean; ASIA/OC, Asia and Oceania. a Odds ratios adjusted for world region of birth, country of delivery, year of delivery, maternal age at delivery and parity and stratified by heterosexual mode and IDU (intravenous drug use) mode of HIV acquisition (Pinteraction = 0.0087). Late ART start in women diagnosed but untreated at conception ART was started in late pregnancy in 13.8% (83/602) of native and 12.8% (212/1651) of migrant women table 2. In an analysis of 2193 women with full-term deliveries, adjusted for country of delivery, year of delivery, maternal age at delivery, mode of HIV transmission and parity, delivering before 2007 (OR 2.58, 95% CI 1.94, 3.42) and being parous (OR 1.45, 95% CI 1.11, 1.90) were associated with late ART start but not being a migrant from SSA (OR 0.98, 95% CI 0.69, 1.37) or from other world regions (OR 0.90, 95% CI 0.54, 1.48) (Supplementary table S3). In a sensitivity analysis including only women who acquired HIV heterosexually, being a migrant was not associated with late treatment initiation [OR (women from SSA vs. native women) 1.11, 95% CI 0.75, 1.64]. VL at delivery Data on VL at delivery were available for 5323 (56.6%) of women who met our inclusion criteria (full-term delivery and on antenatal ART, n = 9406). Women who were native, delivered before 2007 and/or in Spain and/or started ART in early pregnancy were less likely to have data on VL at delivery. Of these 5323 women, delivery VL was above 200 copies/ml in 807 (15.1%) women, comprising 674/4460 (15.1%) migrant and 133/863 (15.4%) native (Supplementary table S4). As expected, multivariable analysis suggested that late ART start in pregnancy was the strongest factor associated with having a detectable VL at delivery. Other factors associated with increased risk of non-suppressed VL included delivering between 2002 and 2006 rather than 2007–12, having a baseline CD4 count < 350 cells/µl and use of mono/dual ART rather than cART (Supplementary table S4) but not being a migrant. Sensitivity analyses restricted to women starting treatment in pregnancy (n = 3705) confirmed the main analysis findings that being a migrant was not associated with a detectable VL at delivery (OR 1.06, 95% CI 0.79, 1.43) even after including only women who acquired HIV heterosexually (OR 1.20, 95% CI 0.86, 1.68). Among the sub-group of women starting ART late, 36% (271/754) of migrants delivered with a detectable VL vs. 39% (42/109) of native women (P = 0.60). Discussion Our study has identified a large proportion of HIV-positive women in Europe delivering a live birth in 2002–12 who were migrants, of whom 44% were diagnosed with HIV in their current pregnancy. At antenatal diagnosis 49% of migrant women had CD4 < 350 cells/µl compared with 29.5% of native women suggesting that pregnancy is an important opportunity for undiagnosed migrant women to learn their HIV status.20 Reassuringly, our analyses suggest that between 2002 and 2012 (our study period), PMTCT interventions have improved in both migrant and native women. Once on ART, migrant women had a good virological response and we did not observe any difference in the crude MTCT rate between migrant and native women. However, we found that migrant women, particularly those from SSA, NAME and EE were more likely to be diagnosed in late pregnancy but we did not identify delayed ART initiation among migrant women who were already diagnosed at conception but not on ART. The observed differences between migrant and native women with respect to late antenatal HIV diagnosis partly agree with previous studies conducted in France21 and Italy,22,23 which reported that migrant women were more likely to receive late HIV screening and/or inappropriate antenatal care compared with native women, with higher rates of MTCT in migrant women. Although we found that migrant women were less likely to know their HIV status at conception, more likely to be diagnosed late in pregnancy and more likely to be diagnosed with low CD4 counts, there was no difference in the proportion achieving undetectable VL by delivery between native and migrant women among those receiving antenatal ART and delivering at term. This suggests that duration of antenatal ART was sufficient to reduce MTCT risk in these migrant women, consistent with the earlier French study where there was no difference in rates of uncontrolled viremia at delivery between African- and French-born women in the ART era.21 Although outside the scope of this paper, it is also likely that among those mother–infant pairs with detectable maternal VL at delivery other PMTCT interventions were successfully applied, given the similar and low MTCT rates in migrant and native women. Antenatal HIV screening is the cornerstone of PMTCT, with timely identification of previously undiagnosed women allowing application of optimum interventions. We showed that, among women with unknown HIV status at conception, migrant women from EE, NAME and SSA were more likely to be diagnosed after 20 weeks gestation than native women after accounting for other factors associated with late antenatal care presentation.7 Although few Western European countries have routine repeat HIV testing for women with negative screens at antenatal booking, it is possible that some women in this ‘late screen’ group acquired HIV in pregnancy and were diagnosed based on a second HIV test. Another finding was a higher rate of HIV diagnosis with severe immunosuppression among migrant women implying missing opportunities for earlier diagnosis and treatment. A recent Italian study has also reported similar findings, with women from SSA more likely to present with advanced HIV disease in pregnancy.24 Previous studies have identified barriers to accessing HIV testing and care experienced by migrants, such as time or financial constraints, language and cultural barriers and living and working conditions.4,5,25–27 In Europe, a large and increasing proportion of pregnant women with HIV are already aware of their HIV status at conception.1,28,29 Retention of patients in HIV care can be challenging, particularly postnatally, and a subsequent pregnancy often provides the means to re-establish a woman in HIV care.17 We found that most diagnosed women not on treatment at conception started ART in the first or second trimester, and report no difference in risk of starting ART late between migrant and native women. Although we showed a nearly 6-fold increased risk of delivery with detectable VL in women who were not on ART until the third trimester, this was driven by women delivering in the early years (2002–07) reflecting changing guidelines. Another reassuring finding was a significant decline in risk of late ART start over calendar time. Western Europe has recorded low MTCT rates for many years, with continuing declines driven by high uptake of antenatal screening, a large and increasing proportion of women on suppressive ART at conception, and prompt start of ART in pregnancy for untreated women. However, socio-economic disparities in access to health care exist,30 and overall low MTCT rates do not preclude the existence of sub-groups at increased risk of poor outcomes, such as the one in six women here with non-suppressed VL at delivery or those who delivered pre-term. Given the benefits of early ART start for health and survival,31,32,34 the considerable number of migrant women in Europe who access antenatal and HIV care services late is of public health concern.6 Our study had some limitations. We excluded women whose country of birth was unknown. As there is no universally agreed definition of the term ‘migrant’ we used country of birth (as reported by the mother) to determine whether a woman was a migrant, but we did not attempt to distinguish between country of birth, nationality or ethnic group. We did not consider the legal status of a migrant, time of arrival in the country of delivery or language spoken, which may affect the ability to access the local health care system, as such information was unavailable. Data were driven by the NSHPC, the largest dataset, and by women from SSA, who made up nearly 70% of the whole study population. We merged all women from SSA into one group, not acknowledging the rich cultural, ethnic and language diversity within SSA.7,33 Although poor socioeconomic status may delay HIV diagnosis and start of ART30 and increase risk of virological non-suppression in treated adults,35 we were unable to adjust for socio-economic factors. As this study is of women diagnosed before or during delivery only, we could not explore migrant women’s risk of delivering with undiagnosed HIV infection (e.g. due to testing decline or incident infection in pregnancy). Women from countries with a high HIV prevalence may remain at elevated risk of HIV acquisition post migration36,37 (e.g. 30% of SSA women living with HIV in France are estimated to have acquired HIV while in France37); highlighting the importance of HIV testing current partners of pregnant women as well as HIV prevention measures for all women including those from high prevalence settings. In conclusion, we have shown that although some migrant women were more likely to be diagnosed in late pregnancy, there was no overall difference between migrant and native women with respect to achieving an undetectable VL by delivery. Good access to antenatal care will not only enable the implementation of PMTCT protocols but also optimize both maternal and child health outcomes generally. Similarly, tailored screening programmes for migrant communities, whether antenatal or for the general population, are needed to help improve the cascade of HIV care and broaden HIV prevention programmes. Wider access to HIV testing and care would give migrant women an opportunity to know their HIV-status before conceiving and either reinforce prevention measures or start ART before conception thus optimizing PMTCT interventions and women’s health. Our findings indicate that particular attention should be given to facilitate access to services for migrant women with children. Supplementary data Supplementary data are available at EURPUB online. Funding This work was funded by the EU Seventh Framework Programme (FP7/2007-2013) under EuroCoord grant agreement n° 260694. The EPPICC Study Group: C. Thorne, H. Bailey; C. Giaquinto, O. Rampon, A. Mazza and A. De Rossi; I. Grosch Wörner; H.J. Scherpbier, M. Kreyenbroek, M.H. Godfried, F.J.B. Nellen and L. van Leeuwen; L. Navér, A.B. Bohlin, S. Lindgren, A. Kaldma, E. Belfrage; J. Levy, P. Barlow, Y. Manigart, M. Hainaut and T. Goetghebuer; B. Vandercam, B. Brichard, D. Van der Linden, H. Waterloos; C. Viscoli; A. De Maria; G. Bentivoglio, S. Ferrero, C. Gotta; N.H. Valerius, V. Rosenfeldt; V. Savasi, S. Fiore, M Crivelli; A Viganò, V. Giacomet, C. Cerini, C. Raimondi and G. Zuccotti; S. Alberico, G Maso, M. Tropea, V. Barresi,. ssa W. Buffolano, R. Tiseo, P. Martinelli, M. Sansone, G. Maruotti, A. Agangi; C. Tibaldi, S. Marini, G. Masuelli, C. Benedetto; M. Marczynska, S. Dobosz, J. Popielska, A. Oldakowska (European Collaborative Study—Western sites); J.T. Ramos, M.I. González-Tomé, M.L. Navarro, M.J. Mellado, L. Escosa, C. Fernández-McPhee, M.A. Roa, J Beceiro, I. Olabarrieta, E. Muñoz (Madrid Cohort of HIV-Infected Mother–Infant Pairs); A. Noguera-Julian, C. Fortuny-Guasch, P. Soler-Palacín, M. Coll, V. Pineda, Ll Mayol, M. Mendez, N. Rovira, T. Vallmanya, A. Mur, L. Garcia, N. Rius, O. Calavia (NENEXP); G. Favarato, K. Francis, A. Horn, H. Peters, R Sconza, C. Thorne, P. Tookey (NSHPC). Conflicts of interest: H.B. has received funding from the International AIDS Society and Public Health England. F.B. has received honoraria and a project grant from Gilead Sciences Europe Ltd. C.T. has received grant funding from EU FP7, UK Medical Research Council, AbbVie, Public Health England, PENTA Foundation and has carried out paid consultancies for UNICEF. For the remaining authors none were declared. Key points In Europe a large proportion of pregnant women with HIV are migrants—mainly from sub-Saharan Africa, albeit with some differences between European countries. 3When compared with pregnant native-born women, migrant women are less likely to know their HIV status at conception, more likely to be diagnosed in late pregnancy and at antenatal diagnosis, more likely to have low CD4 counts. Although some migrant women are more likely to be diagnosed in late pregnancy, among those on ART there is no overall difference between migrant and native women with respect to achieving an undetectable viral load by delivery, a proxy indicator of MTCT risk. Tailored screening programmes for migrant communities, whether antenatal or for the general population, are needed to help improve the cascade of HIV care and broaden HIV prevention programmes. References 1 Townsend CL, Byrne L, Cortina-Borja M, et al.   Earlier initiation of ART and further decline in mother-to-child HIV transmission rates, 2000-2011. AIDS  2014; 28: 1049– 57. Google Scholar CrossRef Search ADS PubMed  2 Mussini C, Antinori A, Bhagani S, et al.   European AIDS Clinical Society Standard of Care meeting on HIV and related coinfections: The Rome Statements. HIV Med  2016; 17: 445– 52. Google Scholar CrossRef Search ADS PubMed  3 Del Amo J, Likatavicius G, Perez-Cachafeiro S, et al.   The epidemiology of HIV and AIDS reports in migrants in the 27 European Union countries, Norway and Iceland: 1999-2006. Eur J Public Health  2011; 21: 620– 6. Google Scholar CrossRef Search ADS PubMed  4 Deblonde J, De Koker P, Hamers FF, et al.   Barriers to HIV testing in Europe: a systematic review. Eur J Public Health  2010; 20: 422– 32. Google Scholar CrossRef Search ADS PubMed  5 Deblonde J, Sasse A, Del Amo J, et al.   Restricted access to antiretroviral treatment for undocumented migrants: a bottle neck to control the HIV epidemic in the EU/EEA. BMC Public Health  2015; 15: 1228. Google Scholar CrossRef Search ADS PubMed  6 Hernando V, Alvarez-del Arco D, Alejos B, et al.   HIV Infection in Migrant Populations in the European Union and European Economic Area in 2007-2012: An Epidemic on the Move. J Acquir Immune Defic Syndr  2015; 70: 204– 11. Google Scholar CrossRef Search ADS PubMed  7 French CE, Thorne C, Byrne L, Cortina-Borja M, Tookey PA. Presentation for care and antenatal management of HIV in the UK, 2009-2014. HIV Med  2017; 18: 161– 70. Google Scholar CrossRef Search ADS PubMed  8 Ng R, Macdonald EM, Loutfy MR, et al.   Adequacy of prenatal care among women living with human immunodeficiency virus: a population-based study. BMC Public Health  2015; 15: 514. Google Scholar CrossRef Search ADS PubMed  9 Aebi-Popp K, Mulcahy F, Rudin C, et al.   National Guidelines for the prevention of mother-to-child transmission of HIV across Europe–how do countries differ?. Eur J Public Health  2013; 23: 1053– 8. Google Scholar CrossRef Search ADS PubMed  10 French CE, Thorne C, Tariq S, et al.   Immunologic status and virologic outcomes in repeat pregnancies to HIV-positive women not on antiretroviral therapy at conception: a case for lifelong antiretroviral therapy? AIDS  2014; 28: 1369. Google Scholar CrossRef Search ADS PubMed  11 Phillips T, Thebus E, Bekker L-G, et al.   Disengagement of HIV-positive pregnant and postpartum women from antiretroviral therapy services: a cohort study. J Int AIDS Soc  2014; 17: 12 McNairy ML, Teasdale CA, El-Sadr WM, et al.   Mother and child both matter: reconceptualizing the prevention of mother-to-child transmission care continuum. Curr Opin HIV AIDS  2015; 10: 403– 10. Google Scholar CrossRef Search ADS PubMed  13 European Collaborative Study. Time to undetectable viral load after highly active antiretroviral therapy initiation among HIV-infected pregnant women. Clin Infect Dis  2007; 1647– 56. 14 Chiappini E, Galli L, Giaquinto C, et al.   Use of combination neonatal prophylaxis for the prevention of mother-to-child transmission of HIV infection in European high-risk infants. AIDS  2013; 27: 991– 1000. Google Scholar CrossRef Search ADS PubMed  15 Monge S, Jarrin I, Mocroft A, et al.   Mortality in migrants living with HIV in western Europe (1997-2013): a collaborative cohort study. Lancet HIV  2015; 2: E540– 9. Google Scholar CrossRef Search ADS PubMed  16 Antinori A, Coenen T, Costagiola D, et al.   Late presentation of HIV infection: a consensus definition. HIV Med  2011; 12: 61– 4. Google Scholar CrossRef Search ADS PubMed  17 Aebi-Popp K, Mulcahy F, Glass TR, et al.   Missed opportunities among HIV-positive women to control viral replication during pregnancy and to have a vaginal delivery. J Acquir Immune Defic Syndr  2013; 64: 58– 65. Google Scholar CrossRef Search ADS PubMed  18 Floridia M, Ravizza M, Pinnetti C, et al.   Treatment change in pregnancy is a significant risk factor for detectable HIV-1 RNA in plasma at end of pregnancy. HIV Clin Trials  2010; 11: 303– 11. Google Scholar CrossRef Search ADS PubMed  19 Katz IT, Leister E, Kacanek D, et al.   Factors associated with lack of viral suppression at delivery among highly active antiretroviral therapy-naive women with HIV: a cohort study. Ann Intern Med  2015; 162: 90– 9. Google Scholar CrossRef Search ADS PubMed  20 Soriano-Arandes A, Noguera-Julian A, Lopez-Lacort M, et al.   Pregnancy as an opportunity to diagnose human-immunodeficiency virus immigrant women in Catalonia. Enferm Infec Microbiol Clin  2016; doi:10.1016/j.eimc.2016.07.011 21 Jasseron C, Mandelbrot L, Tubiana R, et al.   Prevention of mother-to-child HIV transmission: similar access for sub-Sahara African immigrants and for French women?. AIDS  2008; 22: 1503– 11. Google Scholar CrossRef Search ADS PubMed  22 Chiappini E, Galli L, Lisi C, et al.   Risk of perinatal HIV infection in infants born in Italy to immigrant mothers. Clin Infect Dis  2011; 53: 310– 3. Google Scholar CrossRef Search ADS PubMed  23 Izzo I, Forleo MA, Casari S, et al.   Maternal characteristics during pregnancy and risk factors for positive HIV RNA at delivery: a single-cohort observational study (Brescia, Northern Italy). BMC Public Health  2011; 11: 124. Google Scholar CrossRef Search ADS PubMed  24 Floridia M, Tamburrini E, Masuelli G, et al.   Brief report: consequences of presentation with advanced HIV disease in pregnancy: data from a National Study in Italy. J Acquir Immune Defic Syndr  2015; 70: 452– 5. Google Scholar CrossRef Search ADS PubMed  25 Burns FM, Imrie JY, Nazroo J, et al.   Why the(y) wait? Key informant understandings of factors contributing to late presentation and poor utilization of HIV health and social care services by African migrants in Britain. AIDS Care  2007; 19: 102– 8. Google Scholar CrossRef Search ADS PubMed  26 Burns FM, Johnson AM, Nazroo J, et al.   Missed opportunities for earlier HIV diagnosis within primary and secondary healthcare settings in the UK. AIDS  2008; 22: 115– 22. Google Scholar CrossRef Search ADS PubMed  27 Fakoya I, Reynolds R, Caswell G, Shiripinda I. Barriers to HIV testing for migrant black Africans in Western Europe. HIV Med  2008; 9(Suppl 2): 23– 5. Google Scholar CrossRef Search ADS PubMed  28 Briand N, Warszawski J, Mandelbrot L, et al.   Is intrapartum intravenous zidovudine for prevention of mother-to-child HIV-1 transmission still useful in the combination antiretroviral therapy era? Clin Infect Dis  2013; 57: 903– 14. Google Scholar CrossRef Search ADS PubMed  29 Reitter A, Stücker A, Linde R, et al.   Pregnancy complications in HIV‐positive women: 11‐year data from the Frankfurt HIV Cohort. HIV Med  2014; 15: 525– 36. Google Scholar CrossRef Search ADS PubMed  30 COHERE. Delayed HIV diagnosis and initiation of antiretroviral therapy: inequalities by educational level, COHERE in EuroCoord, Socio-economic - Inequalities HIV Writing Group for Collaboration of Observational HIV Epidemiological Research in Europe in EuroCoord. AIDS  2014; 28: 2297– 306. PubMed  31 Mocroft A, Lundgren JD, Sabin ML, et al.   Risk factors and outcomes for late presentation for HIV-positive persons in Europe: results from the Collaboration of Observational HIV Epidemiological Research Europe Study (COHERE). PLoS Med  2013; 10: e1001510. Google Scholar CrossRef Search ADS PubMed  32 TEMPRANO ANRS 12136 Study Group. A Trial of Early Antiretrovirals and Isoniazid Preventive Therapy in Africa. New England Journal of Medicine  2015; 373: 808– 22. CrossRef Search ADS PubMed  33 Tariq S, Elford J, Cortina-Borja M, et al.   Childhood. The association between ethnicity and late presentation to antenatal care among pregnant women living with HIV in the UK and Ireland. AIDS Care  2012; 24: 978– 85. Google Scholar CrossRef Search ADS PubMed  34 World Health Organisation. Consolidated Guidelines on the Use of Antiretroviral Drugs for Treating and Preventing HIV Infection: Recommendations for a Public Health Approach. 2013. Available at: http://apps.who.int/iris/bitstream/10665/85321/1/9789241505727_eng.pdf 35 Saunders P, Goodman A, Smith C, et al.   Does gender or mode of HIV acquisition affect virological response to modern antiretroviral therapy (ART)?. HIV Med  2016; 17: 18– 27. Google Scholar CrossRef Search ADS PubMed  36 Alvarez-del Arco D, Fakoya I, Monge S, et al.   HIV Acquisition among migrants living in Europe: Results from aMASE - Advancing Migrant Access to Health Services in Europe. 15th European AIDS Conference. Barcelona, 21–24 October, 2015. Abstract PS3/5 37 Desgrées-du-Loû A, Pannetier J, Ravalihasy A, et al.   Sub-Saharan African migrants living with HIV acquired after migration, France, ANRS PARCOURS study, 2012 to 2013. Eurosurveillance  2015; 20: © The Author 2017. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.

Journal

The European Journal of Public HealthOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial