F'Efc;;SPECTJVE ultim:n<.'ly. c:~nnot lc3\'e bec:~use il1L') cmnm rep:~y their high do\\0 (ll). ln acldi1ion, returning to their p::~rems'lwrnc m:1y not be an option because di\orcc is consickrcd unacccpr::Jblc and lcJ\ ing their hucbands may h:n e serious implicatil'ns on the soci:ll or trib::~l ties th:Jt were Jen:lop·~d dming the m:1rriage. Cervical Cancer Cl1ild marri:1ge and pol: gam; play an important role in :tnother de3dly disease. ccn ical cancer. HPV infection l1:1s be cum e enclem ic to sub-S:llnran Africa (J ]-2.f). A lihough m:my Africm nations do not hZJ\'e the c1pacity 10 ackqu3iely or cffecli\ ely screen for cer\'ical cancer or HPV. the incidence of cervical cancer in Africa is estimoted to be extrEmely high. Common risks for cervical C3ncer are child marriage, low socioeconomic status, poor access to he:Jlth G1re, and husbands \vho had multiple sex partners. for example, in Mali, cervical cancer is the most common cancer in women, l1ZJS 3n age-standardized incidence rate of24.4 per 100.000, 3nd is the second most common cause of death frum cancer (.?5). In a case-control study of 200 participants with and \1 ithout cervical cancer, among \I hom the mean age at marriage was 15 years, HPV was deieCJed in 97SL, of the c3ses and 40% ofthe contlols. The risk factors identified were cl1ild marriage, high parity (> 10 children). polygamous husbands (>2 wives). and poor genital hygiene (no tap water available and reuse of sanitary n~'pkins). Another study in l\1orocco had similar Jindings (]6), with cervical cancer risk factors identified as child marri:Jge, high p:nity, long-term use of oral contraceptives, and poor gcnit<Jl hygiene (contml participants bathed nwre freCJUently, and case-pm1icipants used homemade sanitary napkins more freCJUCntly). Other studies have also implicated hygiene as a possible factor (JJ,27). Children Bearing Children Pregnancy poses m:my challenges for youmr 12i1l~ Because pregnancy suppresses tl1e immune system (J8). pregnant girls are at increased risk of acquiring diseases like mabria. l\hlaria kills > l million people eacl1 year. 90% of them in Africa. Approxim3tely :25 million pregnant women me t:';poscd to malaria per year, and pregnant women are among the most severely affected by malaria. About 10.5 million become infected during their second or third trimestc·r (29). and among these, the moJiality rate is ==50~0 (30). Not only are pregn3nt women moo.t susceptible to malaria during tl1eir first pregnancy (31), but they 3lso have higlwr rates of mZJbri3-related complications (predominantly pulmonary edema and hypoglycemia) and death than do nonpregnant women. J\1alaria parasite density is significantly higher in pregn:mt girls <']9 years than in pref'nant \I omen > 19 years. (32) Howe\'er, a \\·oman \lho has h0d malari::J during pregnancy is less susceptible 1646 to mcol:m;i dmmg \\Om~m i' also J--!1\' init.'Cltd (3}), The intcr8c1ic1n bL'i\\ ecn l-11\ ·and rnzd~nizl in yuun~ nLJrried girls is cic\ E:tlcs of coinfeClic'n He u• Cenir:ll Ai!·iuu1 Republic. J\lcdm~.i. Z:m1b1:t. ;md limb::bwe. where >90°[, ofthe pc>pubtion are to malari:: and >lO':o ~!rt Hl\' pwitiYc. H!\'-infcued p;nients :1re much murc 10 infection \\ i1h 1_) ll ;s J.i 2odi z; 717 Pregnalll women ]JJ\ c l1igh maLJria p::ra:citcomia in the pbcenia and rnorc• senTc clinical dise3c.c, which afrects not just tlle i1rst pregnancy but :tll subsequent prcgn:mcies. HJV-infccled pl1ic-nts also elL' ncn respond as \\icll to standard ::mimabriZJ treatmenJ Fin:dly. malaria increases Hl\' \ira] load and r:Jises the risk fur mother-to-child HI\' transmission (29). The biologic interaction between these diseases not only complicz1tes trc:11mcnt in an alre(ldy cktlknging sc·tting but also presents a serious risk for death to pregnant girls <19 years of age. Chiidren Delivering Children Binhs resulting from cl1ild marriages are said to be "wo soon, too close, too many, or too l<Jte'. (33). For ex<1mple. a high percentage of girls in Ethiopia (:25%), U (4:2%,), and M2,li (45%) have given bi1ih by the a."e of ! f) compared with only lS·u in Germ:my. 2% in France, ::nd l 0% in the United St<~les (1). The problem \\ith children delivering children is that the yollng mothers are at a significantly higher risk than older women for debilitating illness and even death. Comp:1rcd with women >20 ye:trs of age, girls 10-14 years of age <Jre 5-7 times more likely to die from childbirt1J, and girls J 5-19 years of age nre twice as Ji].;ely (3.f). for exflmple, in Mali, the malcrn3lmm1ality rate for girls aged 15-19 is 178 pcr l 00.000 lin bi1ihs and for women aged :20-34, only 3:2 per 100,000. ln Togo. for the same age grollps, these ra1es are 286 and3CJ. respectively (1). Rez;sons for these high death rates inclucle eclampsia. postp2rtum hemorrhage. HJV infection. m::lari3, and ob~tructecll3bor. Obstructed labc>r is the result of a girl's pelvis being too smaJl to deliver a fe!lE. The fctw,'s head passes into thE 'a gina. bu1 its shoulders CZlllllOt fii thmugh the mother's pelvic Lones. \\'ilhout a ccsarczlll section. the neonate dies. and tl1e mother is fortun3ie if she survives. lf sepsis or hemorrh;::gc clocs not occur anclthe girl does survive. the tissue and bones of1l1e neomne will eventually soften and the remains will pas:o lhrougl-~ the vagn1a. l\1any times. obstructed lc:bor lendo to fistulas: tlK pressure of the fetal heZid on the vagin::1l woll causes ti.scue necrosis, and fistulas dn•elop between the vagina ancl iht blz1dder or rectum afier the necrotic tissue slough5. l\'lorcthan 2 million adolescents are living with fi'luh:s. ,,nd fistulas deYelop in == l OO.ODO more each year (35). Girls ~1gcs 10- J 5 ) ears arc especiall:v v1llnerable bcc:mse 11Kir pe h ic Emerging Infectious Dissc;ses • vrww.cdc.gov/eid ·Vol. 12, No. 11. Nc·1ernbsr 200iJ