Secondary Logo

Journal Logo

Original Study

Increased Female-Male Mortality Ratio Associated With Inactivated Polio and Diphtheria-Tetanus-Pertussis Vaccines

Observations From Vaccination Trials in Guinea-Bissau

Aaby, Peter MSc*†; Garly, May-Lill MD, PhD*†; Nielsen, Jens MSc*†; Ravn, Henrik PhD*†; Martins, Cesario MD, MSc*; Balé, Carlitos MD*; Rodrigues, Amabelia PhD*; Benn, Christine Stabell MD, PhD*†; Lisse, Ida Maria MD*†

Author Information
The Pediatric Infectious Disease Journal: March 2007 - Volume 26 - Issue 3 - p 247-252
doi: 10.1097/01.inf.0000256735.05098.01



Click on the links below to access all the ArticlePlus for this article.

Please note that ArticlePlus files may launch a viewer application outside of your web browser.

In the prevaccination era, female infants in West Africa appear to have had the same or slightly lower postneonatal mortality than boys.1–3 However, several studies have suggested that vaccines may affect the mortality ratio (MR) between girls and boys.1–16 Standard measles vaccine (MV) is associated with a lower female-male MR,1–3,8,9,12–14 whereas diphtheria-tetanus-pertussis vaccine (DTP)11–13,15 and hepatitis B vaccine (HBV)16 are associated with higher mortality for girls than boys. The observations of higher female mortality after DTP suggested a surprising explanation of the 2-fold higher mortality among girls after high-titer measles vaccination (HTMV). In the routine immunization program recommended by WHO, 3 doses of DTP have been followed by MV at 9 months of age. However, HTMV was given early at 4–5 months of age, and was usually followed by DTP or IPV. The excess female mortality associated with HTMV was limited to children who had received DTP or IPV after HTMV.12

If DTP and IPV increase the female-male MR, modifications in the current routine immunization program could possibly lead to lower child mortality. We therefore examined whether IPV and DTP have an impact on the female-male MR when these vaccines are administered prior to the first MV.

IPV has been used as a control vaccine7,17 in several randomized trials of early MV at 4–6 months of age in Guinea-Bissau. All control children were offered MV at 9 months of age but many were traveling and only received MV later. Many children received additional doses of DTP and OPV during the trials and before receiving MV. In a MV trial in Sudan,18 missing doses of DTP at enrollment, and therefore presumably receiving these doses after enrollment, were associated with increased mortality after enrollment. In the present analysis, we therefore examined whether the DTP vaccination status prior to enrollment had an influence on the effect of IPV.


Measles Vaccine Trials.

We have conducted 4 trials of early MV in the capital of Guinea-Bissau. In the 1980s, we conducted 2 trials using medium or high-titer Edmonston-Zagreb (EZ) measles vaccines from 4 months of age as described in previous publications.1,4,19 In the 1990s, we examined the effect of 2 doses of MV given at 6 and 9 months of age.17,20 In all trials, we used IPV in the control group. The control group was offered standard MV from 9 months of age. At any given time, many children would be traveling with their mother because of work elsewhere or for family reasons. Hence, many children only received their MV later. Main features of relevance to the present reanalysis of the studies are summarized below.

The first trial in Bissau included children born between August 1984 and September 1985 and registered before 4 months of age in Bandim 1, the area in which the Bandim Health Project (BHP) has maintained a surveillance system since 1978.7,19 The BHP has conducted 3-monthly anthropometric examinations of the children less than 3 years of age since 1980. These examinations were conducted at different meeting places in the study area. Whenever vaccines were available the anthropometric team was accompanied by a nurse from the local health center who provided vaccinations to children missing the routine immunizations recommended in Guinea-Bissau, ie, BCG, 3 doses of oral polio (OPV) and DTP vaccines and MV (Fig. 1A). From 1985, children were randomized to receive a medium dose of EZ or an IPV from 4 months of age. At 9 months, they were invited to come back; the EZ-group then received IPV, whereas children in the control group received standard Schwarz (SW) MV (Fig. 1A). Ninety-five percent of the children still living in the study area at 9 months of age took part in the study, the percentage not being different for boys (95.0%) and girls (95.3%).

Design of early measles vaccination trials: A, EZ measles vaccination trials. Children were recruited at 4 months of age and were offered the second vaccine at 9 months of age, IPV in the EZ vaccination group and standard MV in the control group. B, Two-dose measles vaccination trials. Children were recruited at 6 months of age and were offered the second vaccine at 9 months of age, standard MV in both groups. DTP, diphtheria-tetanus-pertussis vaccine; EZ, Edmonston-Zagreb; HBV, hepatitis B vaccine; IPV, inactivated polio vaccine; MV, measles vaccine; OPV, oral polio vaccine; SMV, standard measles vaccine.

The second EZ trial in Bissau enrolled children born between May 1986 and April 19874,7; the study had a similar design except that two thirds of the EZ group received high-titer EZ MV, whereas a medium dose of EZ vaccine had been used in the first part of the second trial.12 Ninety-five percent of the children still living in the study area at 9 months of age took part in the study, the percentage being the same for boys (95.8%) and girls (94.8%). The children in both EZ studies were followed to May to June 1990.

In the two-dose trials, children were randomized at 6 months of age to standard-titer MV or IPV. The trial team only provided these vaccines at enrollment. If children were missing other routine vaccinations, they were advised to go to the health center the next day. At 9 months, the children were offered standard MV (Fig. 1B). In the first of these trials, 300 children born between September 1992 and July 1993 were also randomized to vitamin A or placebo.20,21 Among the children who received IPV and placebo, no child died before they received MV and this trial did therefore not contribute to the estimate of the female-male MR in the IPV group.21

The second two-dose trials was initiated in 1995 and included all children born between September 1994 and October 2001 and registered before 6 months of age in the 4 urban districts followed by the Bandim Health Project, ie, Bandim 1, Bandim 2, Belem and Mindara.17,22 Children born until mid-February 1995 received EZ standard MV, whereas the subsequent cohorts received SW MV.17,22 Eighty-five percent of the children still living in the study area at 9 months of age took part in the study, this percentage being the same for boys (85.4%) and girls (85.2%). A subgroup of children in the two-dose trial received HBV at 7.5, 9 and 10.5 months of age as part of an immunologic study. HBV affected the female-male MR,16 and the children were therefore censored in the present analysis when they received HBV. During the second two-dose trial, there was a war in Bissau starting on June 7, 1998; most people had to take refuge in theinterior of the country and mortality was considerably higher during this period.8,13,23 The war ended in May 1999. We conducted a subanalysis in which the war period was censored.

The vaccination status for other vaccines was assessed at the time of enrollment. In the 2 EZ studies, missing doses of DTP and OPV were administered at the same time as the children were enrolled in the trial. In the two-dose MV study, children missing DTP or OPV were advised to come back to the health center the following day to receive the missing vaccines. Many children came and received these vaccines. These vaccinations were not registered directly by the project, but only when the child’s vaccination card was seen at a subsequent 3-monthly home visit. Hence, it was not possible to control for these vaccinations in the survival analysis as the information would be missing for children who died or moved before their vaccination card was seen again. To assess the likely impact of DTP vaccinations during these trials, we classified the children according to whether they had received the 3 initial doses of DTP prior to enrollment. The children missing doses of DTP would be likely to receive them during follow-up, whereas the fully DTP-vaccinated children would not.18 We assessed the prevalence of such subsequent DTP vaccinations among children who remained in the area to at least 3 years of age and who had their vaccination card inspected at least once after enrollment.

Epidemiologically, it has been difficult to separate the effect of DTP and OPV since the vaccines are usually given together in the routine immunization program. However, there was no negative effect when OPV was used alone24,25 and the potential negative effect of DTP and OPV11,15 is therefore most likely to be caused by DTP. We shall therefore only refer to the effect of DTP in the following presentation.

To ensure that female mortality was not particularly high in these trial cohorts, we examined first, the female-male MR in the early MV groups and second, the female-male MR from the time the IPV children got standard MV until 3 years of age.

For the present analysis, we reviewed verbal autopsy forms to identify accidents. All deaths except one in the IPV groups were likely to be caused by infectious diseases. In the second study, one boy with Down’s syndrome was considered a “spirit” child and abandoned; we censored the child in the survival analysis.


In the first 2 EZ studies, we used IPV kindly donated by Merieux, France. In the two-dose trials, we used IPV purchased from Statens Serum Institut (SSI), Denmark, in the first trial, and from SSI and SMI, Sweden, in the second trial. During a 2-month period in 1995, 182 children received a batch of IPV, which had been frozen. Exclusion of these children had no impact on the estimates (data not shown).

Statistical Analysis.

The children were followed from the time of the first IPV or MV at 4–6 months of age and until migration, death, receiving measles vaccine, receiving HBV, or becoming 3 years of age, whichever came first. Many children were traveling at 9 months of age and were only measles vaccinated later. Hence, the IPV group could be followed beyond 9 months of age. In additional analyses, we censored at 3 months after vaccination, at 9 months of age and at 18 months of age.

Anthropometric z-scores for weight-for-age and height-for-age were calculated using WHO’s Anthro program. We used a Cox proportional hazards model stratifying for trial to estimate the mortality rate ratios (MR) for girls and boys.


Female-Male Mortality Ratio After MV and IPV.

The number of children enrolled, follow-up time and deaths are presented in Table 1. In the groups receiving early MV, the female-male MR was 1.01 (0.69–1.46) between enrollment and the end of follow-up (Table 1). In all IPV groups, the pattern was higher mortality for girls, the female-male MR being 1.52 (1.02–2.28). The IPV-group was called for MV at 9 months of age but many only attended later because of traveling.22 Results in the IPV group were similar if follow-up was censored at 3 months after vaccination (MR 1.73; 0.97–3.10) at 9 months or at 18 months of age (Table 1). If follow-up was censored during the war, the female-male MR was 1.62 (0.91–2.90).

Female-Male Mortality Ratios Among Recipients of Measles Vaccine and IPV in Vaccine Trials in Guinea-Bissau, 1985–2003

When children in the IPV group received MV at 9 months or later, the female-male MR changed significantly to 0.88 (0.68–1.14) (P = 0.025) (Table 2).

Female-Male Mortality Ratios Among Recipients of Standard Measles Vaccine in Vaccine Trials in Guinea-Bissau, 1995–2003

We tested several background factors, including district, ethnic group, schooling of mother, schooling of father, number in family, number of persons in bed, roof, electricity, season and weight-for-age, but none of these affected the female-male estimates by more than 4%. There was no interaction between sex and these background factors. There was no difference in nutritional status for boys and girls at enrollment in the IPV groups that could explain the difference in mortality; both the weight-for-age and the height-for-age z-scores were significantly better for girls than for boys at enrollment (data available online: Appendix).

DTP Vaccination Status and Mortality.

Since DTP has been associated with increased female mortality, mortality in the IPV group could be caused by DTP given after IPV vaccination but before MV. To test this, we distinguished children who had received 0–2 or 3 doses of DTP, respectively, before enrollment (Table 3). Only 8 children in the first 2 EZ trials had received DTP3, whereas 49% of the children in the two-dose trial had received DTP3 before enrollment. Of the children from the two-dose study who remained in the area, at least 84% (1154 of 1379) received one or more of the missing doses of DTP according to the routine registration system.

Mortality Among Children Randomized to Measles Vaccine or IPV According to Whether They Had Received All Doses of DTP Before Enrollment in Guinea-Bissau, 1985–2003

Children who were not fully DTP vaccinated at enrollment tended to have higher mortality than DTP3 vaccinated children (MR 1.30; 0.97–1.73), the MR being 1.61 (1.08–2.40) for girls and 1.02 (0.67–1.54) for boys (test of homogeneity, P = 0.12).

DTP Vaccination Status and the Female-Male Mortality Ratio.

In the IPV group, the female-male MR was significantly increased for children who had not received DTP3 (MR 1.69; 1.01–2.83) but not for children with DTP3 vaccination prior to enrollment (MR 1.28; 0.67–2.46) (Table 3). In the MV group, the female-male MR tended to be higher for children who had received less than DTP3 (MR 1.25; 0.80–1.96) than for children who had received DTP3 before enrollment (MR 0.63; 0.32–1.24) (test of homogeneity, P = 0.10) (Table 3).

DTP Vaccination Status, Sex-Differential Effects and the Comparison of IPV and MV Vaccinations.

To facilitate overview, the groups compared have been indicated in Table 3 and in the text. The overall effect of IPV compared with MV was limited; IPV may have been associated with slightly higher mortality for girls (MR 1.25; 0.87–1.80) (Table 3, groups 3 and 4 vs. groups 1 and 2) but slightly lower mortality for boys (MR 0.83; 0.55–1.25) (Table 3, groups 7 and 8 vs. groups 5 and 6). For children missing doses of DTP, there was no difference between the IPV and the MV groups (Table 3, groups 4 and 8 vs. groups 2 and 6). Among children who had received DTP3 at enrollment, IPV-vaccinated girls had a MR of 1.86 (0.95–3.67) compared with MV-vaccinated girls (Table 3, group 3 vs. group 1). It made no difference for boys (MR 0.92; 0.48–1.77) (P = 0.14) (Table 3, group 7 vs. group 5). Exclusion of measles deaths had no impact on these estimates (data not shown).

It should be noted that there was little difference in mortality rate for boys (Table 3, groups 5, 6, 7 and 8). However, mortality varied markedly among the girls depending on the vaccination status. Girls in the MV group who had received DTP3 before enrollment (and were unlikely to have received DTP after MV) had a lower mortality than girls who had not received DTP3 (MR 0.50, 0.26–0.95) (Table 3, group 1 vs. group 2). Similarly, these girls had a lower mortality than girls whose last vaccine received was likely to be DTP (because they had not received DTP3 at enrollment) or IPV (MR 0.49; 0.28–0.87) (Table 3, group 1 vs. groups 2, 3 and 4).


The main focus in these trials was the effect of MV on measles infection and general mortality.9 However, the IPV and DTP vaccinations administered at enrollment and during follow-up might have had very important effects on the observed mortality patterns. Vaccine trials conducted in developing countries with high childhood mortality have often used another vaccine as control, for example, IPV, rabies, meningococcal vaccine26 or DTP.27 This is easier to explain to the population, and it has been considered unethical to administer a control vaccine that confers no benefit.26 It has been assumed that the control vaccine had no effect on the morbidity or mortality pattern for the disease studied; for example, with a high OPV coverage, IPV would be unlikely to prevent any death from polio infection. However, if vaccines have nonspecific effects, this practice as well as the unrelated routine vaccinations given during the course of the trial might in fact bias the evaluation. The impact of DTP or IPV vaccinations administered after enrollment could easily distort the estimated mortality effect of the “vaccine of interest.”

These trials suggest that IPV and DTP are associated with increased female-male MRs. First, children randomized to receive IPV had a significantly increased female-male MR (1.52; 1.02–2.28), whereas children randomized to receive MR had no increase in the female-male MR (1.01; 0.69–1.46); after the children in the IPV group had been given MV at 9 months of age or later, the female-male MR fell from 1.52 to 0.88 (0.68–1.14). Second, children who had not received DTP3 when they were enrolled (and were therefore likely to receive DTP after MV or IPV) had a higher mortality than children who had received all 3 doses of DTP (MR 1.30; 0.97–1.73); this effect was seen only in girls (MR 1.61; 1.08–2.40) and not in boys (MR 1.02; 0.67–1.54). Furthermore, girls in the MV group who had not received DTP3 (and therefore were likely to have received DTP after MV) had a higher mortality than girls in the MV group who had received DTP3 before enrollment (MR 2.00; 1.05–3.85). This trend seems consistent with the observation from the HTMV trials that DTP or IPV administered after HTMV were associated with increased female mortality.12

There was no differential recruitment of boys and girls in these randomized studies. It seems unlikely that parental gender preferences would lead to increased female mortality, when there was no such trend among measles-vaccinated children. In the prevaccination era, girls had, if anything, slightly lower mortality than boys.3 The trend cannot be explained by unusually low mortality for boys in the IPV group. Girls were not more malnourished than boys, rather the opposite. The increased female mortality appeared to be stronger in the first 2 EZ studies than in the two-dose study; however, in the two-dose study, 49% had received all 3 doses of DTP before enrollment. Adjusting for DTP vaccination status at enrollment, there was little difference in the female-male MR in the IPV group between the trials (Table 3).

In the present study, the children were randomized to IPV, and it seems unlikely that uncontrolled confounding might explain why IPV but not MV was associated with increased female mortality. We have previously found DTP3,10–13,15 and HBV16 to be associated with an increased female-male MR in observational studies. In contrast to IPV, both DTP and HBV have aluminum as an adjuvant. In the present analysis, children who received HBV were censored, so the increased female mortality in the IPV group is unrelated to the effect of HBV. On the other hand, the relative importance of IPV and DTP is difficult to assess. All children had received both DTP and IPV. However, some aspects of the present study suggest that DTP is definitely associated with increased female mortality. First, the female-male MR was particularly increased for children who presumably received missing doses of DTP after enrollment. Second, female mortality was also increased among the MV-vaccinated children missing doses of DTP, even though the female-male MR was low among DTP3-vaccinated children who received MV at enrollment. “Missing doses of DTP” could be a marker of poor health or poor compliance but was only important for girls. Furthermore, “missing doses of DTP” were also important in other trials18; in Sudan, the number of “missing doses of DTP” was associated with a significant linear trend in the female-male MR. IPV was not used in Sudan.

The female-male MR of 1.01 among children randomized to MV at 6 months of age could be said to contradict our previous observation of lower female-male mortality among recipients of MV.1–3,8,9,13 However, the present children received MV early and most had not received DTP3 at enrollment; these children had higher female mortality. Among children having received DTP3 before enrollment, girls did have lower mortality than measles-vaccinated boys as observed in previous studies.1–3,13

These observations question assumptions underlying current vaccination policy. The possibility that vaccines could have gender-differential effects has not been fully considered. According to the Global Advisory Committee on Vaccine Safety, there is no sex-differential effect of DTP.28–30 Given the impact of DTP and IPV after HTMV,12,18 this is surprising. WHO-sponsored studies31–33 did not analyze gender effects according to the last vaccination; and since DTP and MV vaccines have opposite tendencies, such studies are unlikely to distinguish differences.34 Lower female mortality after MV or increased female mortality after DTP has been observed in several countries, including Guinea-Bissau, Senegal,12 Gambia,35 Sudan,18 Congo,18 Malawi,14 India,36 and Bangladesh.37 These trends have not been contradicted. If the female-male MR is increased after DTP as last vaccination but low after MV, providing MV vaccination before 9 months of age might reduce female infant mortality considerably.


1. Aaby P, Samb B, Simondon F, et al. Divergent mortality for male and female recipients of low-titre and high-titre measles vaccines in rural Senegal. Am J Epidemiol. 1993;138:746–755.
2. Desgrées du Loû A, Pison G, Aaby P. The role of immunizations in the recent decline in childhood mortality and the changes in the female/male mortality ratio in rural Senegal. Am J Epidemiol. 1995;142:643–652.
3. Aaby P, Jensen H, Rodrigues A, et al. Divergent female-male mortality ratios associated with different routine vaccinations among female-male twin pairs. Int J Epidemiol. 2004;33:367–373.
4. Aaby P, Knudsen K, Whittle H, et al. Long-term survival after Edmonston-Zagreb measles vaccination: increased female mortality. J Pediatr. 1993;122:904–908.
5. Aaby P, Samb B, Simondon F, et al. Sex specific mortality after high titre measles vaccines in rural Senegal. Bull WHO. 1994;72:761–770.
6. Holt EA, Moultin LH, Siberry GK, Halsey NA. Differential mortality according to measles vaccine titer and sex. J Infect Dis. 1993;168:1987–1996.
7. Knudsen KM, Aaby P, Whittle H, et al. Child mortality following standard, medium and high titre measles vaccination in West Africa. Int J Epidemiol. 1996;25:665–673.
8. Aaby P, Garly ML, Balé C, et al. Survival of previously measles-vaccinated and measles-unvaccinated children in an emergency situation: an unplanned study. Pediatr Infect Dis J. 2003;22:798–805.
9. Aaby P, Samb B, Simondon F, Coll Seck AM, Knudsen K, Whittle H. Non-specific beneficial effect of measles immunization: analysis of mortality studies from developing countries. BMJ. 1995;311:481–485.
10. Aaby P, Jensen H. Routine vaccinations and child survival: effect of gender. BMJ. 13 Dec 2002;
11. Aaby P, Jensen H, Gomes J, Fernandes M, Lisse IM. The introduction of diphtheria-tetanus-pertussis vaccine and child mortality in rural Guinea-Bissau: an observational study. Int J Epidemiol. 2004;33:374–380.
12. Aaby P, Jensen H, Samb B, et al. Differences in female-male mortality after high-titre measles vaccine and association with subsequent vaccination with diphtheria-tetanus-pertussis and inactivated poliovirus: a re-analysis of the West African studies. Lancet. 2003;361:2183–2188.
13. Aaby P, Jensen H, Garly ML, Balé C, Martins C, Lisse I. Routine vaccinations and child survival in war situation with high mortality: effect of gender. Vaccine. 2002;21:15–20.
14. Ashorn P, Maleta K, Espo M, Kulmala T. Male-biased mortality among 1–2 years old children in rural Malawi. Arch Dis Child. 2002;87:386–387.
15. Kristensen I, Aaby P, Jensen H. Routine vaccinations and child survival: follow-up study in Guinea-Bissau. BMJ. 2000;321:1435–1438.
16. Garly ML, Jensen H, Martins CL, et al. Hepatitis-B vaccination may be associated with an increased female mortality in Guinea-Bissau: an observational study. Pediatr Inf Dis J. 2004;23:1086–1092.
17. Garly ML, Balé C, Martins CL, et al. Measles antibody responses after early two dose trial in Guinea-Bissau with Edmonston-Zagreb and Schwarz standard-titre measles vaccine: better antibody increase from booster dose of Edmonston-Zagreb vaccine. Vaccine. 2001;19:1951–1959.
18. Aaby P, Ibrahim S, Libman M, Jensen H. The sequence of vaccinations and increased female mortality after high-titre measles vaccine: trials from rural Sudan and Kinshasa. Vaccine. 2006;24:2764–2771.
19. Aaby P, Lisse IM, Whittle H, et al. Long-term survival in trial of medium-titre Edmonston-Zagreb measles vaccine in Guinea-Bissau: five-year follow-up. Epidemiol Infect. 1994;112:413–420.
20. Benn CS, Aaby P, Balé C, et al. Effect of vitamin A supplementation on the antibody response to measles vaccine in Guinea-Bissau, West Africa. Lancet. 1997;350:101–105.
21. Benn CS, Bale C, Sommerfelt H, Friis H, Aaby P. Vitamin A supplementation and childhood mortality: amplification of the non-specific effects of vaccines? Int J Epidemiol. 2003;32:822–828.
22. Garly ML, Martins CL, Balé C, et al. Early two-dose measles vaccination schedule in Guinea-Bissau: good protection and coverage in infancy. Int J Epidemiol. 1999;28:347–352.
23. Aaby P, Gomes J, Fernandes M, Djana Q, Lisse I, Jensen H. Nutritional status and mortality of refugees and residents in a non-camp setting during conflict: follow up study. BMJ. 1999;319:878–881.
24. Aaby P, Rodrigues A, Biai S, et al. Oral polio vaccination and low case fatality at the paediatric ward in Bissau, Guinea-Bissau. Vaccine. 2004;22:3014–3017.
25. Aaby P, Hedegaard K, Sodemann M, et al. Childhood mortality after oral polio immunisation campaign in Guinea-Bissau. Vaccine. 2005;23:1746–1751.
26. Hoskins EW. Edmonston-Zagreb measles vaccine: randomized controlled clinical trial in Sudan [Thesis]. University of Oxford, 1993.
27. Hartfield J, Morley D. Efficacy of measles vaccine. J Hyg. 1963;61:143–147.
28. Global Advisory Committee on Vaccine Safety. Wkly Epidemiol Rec. 2002;77:393–394.
29. Hall AJ. Vaccination and child mortality. Lancet. 2004;364:2156–2157.
30. WHO Task Force on Routine Infant Vaccination and Child Survival. Report of a meeting to review evidence for a deleterious effect of DTP vaccination on child survival. London, 2004. (accessed 20 Dec 2004).
31. Lehmann D, Vail J, Firth MJ, de Klerk NH, Alpers MP. Benefits of routine immunisations on childhood survival in Tari, Southern Highlands Province, Papua New Guinea. Int J Epidemiol. 2005;34:138–148.
32. Breiman RF, Streatfield PK, Phelan M, Shifa N, Rashi M, Yunus M. Effect of infant immunization on childhood mortality in rural Bangladesh: analysis of health and demographic surveillance data. Lancet. 2004;364:2204–2211.
33. Vaugelade J, Pinchinat S, Guielle G, Elguero E, Simondon F. Lower mortality in vaccinated children: follow up study in Burkina Faso. BMJ. 2004;329:1309–1311.
34. Aaby P, Benn CS, Nielsen J, Lisse IM, Rodrigues A, Jensen H. Estimating the effect of DTP vaccination on mortality in observational studies with incomplete vaccination data. Trop Med Int Health. 2007;12:15–24.
35. Aaby P, Jensen H, Walraven G. Age-specific changes in female-male mortality ratio related to the pattern of vaccinations: an observational study from rural Gambia. Vaccine. 2006;24:4701–4708.
36. Moulton LH, Rahmathullah L, Halsey NA, Thulasiraj RD, Katz J, Tielsch JM. Evaluation of non-specific effects of infant immunizations on early infant mortality in a southern Indian population. Trop Med Int Health. 2005;10:947–955.
37. Benn CS, Fisker AB, Jørgensen MJ, Aaby P. Why worry: vitamin A with DTP vaccine? Vaccine. 2007;25:777–779.

DTP; IPV; measles vaccine; nonspecific effects of vaccination; sex-differential effects

Supplemental Digital Content

© 2007 Lippincott Williams & Wilkins, Inc.