Professional Documents
Culture Documents
Non-Specific Effects of Diphtheria-Tetanus-Pertussis and Measles Vaccinations? An Analysis of Surveillance Data From Navrongo, Ghana
Non-Specific Effects of Diphtheria-Tetanus-Pertussis and Measles Vaccinations? An Analysis of Surveillance Data From Navrongo, Ghana
Abstract objectives Studies from low-income countries have suggested that routine vaccinations may have non-
specific effects on child mortality; measles vaccine (MV) is associated with lower mortality and
diphtheria-tetanus-pertussis (DTP) with relatively higher mortality. We used data from Navrongo,
Ghana, to examine the impact of vaccinations on child mortality.
methods Vaccination status was assessed at the initiation of a trial of vitamin A supplementation and
after 12 and 24 months of follow-up. Within the placebo group, we compared the mortality over the first
4 months and the full 2 years of follow-up for different vaccination status groups with different like-
lihoods of additional vaccinations during follow-up. The frequency of additional vaccinations was
assessed among children whose vaccination card was seen at 12 and 24 months of follow-up.
results Among children with a vaccination card, more than 75% received missing DTP or MV during
the first 12 months of follow-up, whereas only 25% received these vaccines among children with no
vaccination card at enrolment. Children without a card at enrolment had a significant threefold
higher mortality over the 2-year follow-up period than those fully vaccinated. The small group of
children with DTP3-4 but no MV at enrolment had lower mortality than children without a card and
had the same mortality as fully vaccinated children. In contrast, children with 1–2 DTP doses but no MV
had a higher mortality during the first 4 months than children without a card [MRR = 1.65 (0.95,
2.87)]; compared with the fully vaccinated children, they had significantly higher mortality after
4 months [MRR = 2.38 (1.07, 5.30)] and after 2 years [MRR = 2.41 (1.41, 4.15)]. Children with 0–2
DTP doses at enrolment had higher mortality after 4 months (MRR = 1.67 (0.82, 3.43) and after 2 years
[MRR = 1.85 (1.16, 2.95)] than children who had all three doses of DTP at enrolment.
conclusions As hypothesised, DTP vaccination was associated with higher child mortality than
measles vaccination. To optimise vaccination policies, routine vaccinations need to be evaluated in
randomised trials measuring the impact on survival.
subsequent mortality (Kristensen et al. 2000; Aaby et al. tion visits was not updated. Comparing information from
2004a,b; Veirum et al. 2005). These results have been rounds 1 and 4 for the same children, there were minor
controversial because they question basic assumptions of inconsistencies (Appendix).
the current vaccination programme and suggest that
changes in the programme might be needed (Fine & Smith
Follow-up and verbal autopsies
2007).
More data sets from other countries are therefore needed All children in the study were visited every 4 months.
to assess the magnitude of NSE and whether different Deaths were identified through these visits and indepen-
routine vaccinations have different NSE. Surprisingly, few dently by key informants based in the community. With the
longitudinal studies in low-income countries have collected intensive follow-up and the need to report, the status of all
routine information on vaccinations. The Ghana vitamin A children in the study it is unlikely that deaths were missed.
supplementation trial (Ghana VAST)’s Survival Study in The verbal autopsy procedure for all deaths makes it
Kassena-Nankana District, Ghana, in 1989–1991 assessed unlikely that surviving children were erroneously classified
vaccination status at the initiation of the trial. Previously, as dead. The follow-up procedures were independent of the
we have demonstrated that the impact of vitamin A vaccination status of the child.
supplementation (VAS) may be modified by interaction When a child had died an attempt was made to complete
with vaccines (Benn et al. 2009). In the present analysis, we a verbal autopsy questionnaire with the mother ⁄ guardian
have focused on the impact of vaccines on child mortality to ascertain the circumstances of the death and the
among children who received placebo to study the impact symptoms associated with the final illness. No information
of vaccines in the absence of VAS. on vaccines was collected. Questionnaires were read and
classified by two physicians (DAR, Fred Binka). If any of
the physicians classified a death as because of measles
Methods and subjects infection, the death has been classified as measles in the
Ghana VAST Survival Study present analysis.
Round 4 not only reflected which vaccines the child had of vaccine status because children who survive have better
already received, but also which vaccines they were likely information than dead children because health cards for
to receive in the coming months. Hence, any vaccination dead children were often not available for vaccination
effects on mortality are likely to be a product of both information to be updated. This makes vaccine informa-
vaccination status at enrolment and the vaccines likely to tion for the dead children incomplete making vaccination
be received during follow-up. We have particularly focused to be automatically associated with a strong beneficial
on the impact on survival in the first 4 months when the effect.
information on immunisation status would be most valid.
However, we have also reported the mortality effect for the
Results
full 2 years duration of the trial. For the present analyses,
children who had no health card at Round 1 have been Study population
assumed to be ‘unvaccinated’.
A total of 6882 children were enrolled in the placebo group
We have emphasised two aspects of vaccination status:
in Round 1; 49% had a health card seen, 13% were said to
first, children who had not yet received MV were likely to
have a card but it was not seen, 37% did not have a card,
receive MV during follow-up; second, children who had
and 0.4% had no information recorded on their health
not received all doses of DTP were likely to receive
card status. The present analysis is restricted to children
additional doses of DTP during follow-up. We have
whose card was seen (3397) or who had no card (2577).
therefore distinguished between children who had received
Within the first 4 months 91 (1.5%) of these children died
0–2 doses (DTP0-2) or 3 or more doses of DTP (DTP3-4)
and over the 2 years of the trial, 274 (4.6%) children died;
at enrolment. As indicated in Table 2, vaccination status at
51 ⁄ 274 (19%) were because of measles and the two
enrolment was categorised into 6 groups.
physicians agreed on 47 of 51 diagnoses of measles.
Statistical analyses
Vaccination coverage
Comparisons of background factors for children with and
without a health card were carried out using logistic Fewer than 50% of the children had received BCG, DTP
regression, linear regression or nonparametric tests. For the and MV (Table 1). For those with a health card seen,
assessment of the risk of dying within 4 months or 2 years, however, the coverage continued to increase until around
the period at risk was defined as the time period between 24 months of age. Most children had received their routine
the enrolment date and either the date the child died or was vaccinations out-of-sequence; 76% had received BCG and
censored or 4 months ⁄ 2 years later, whatever came first. DTP simultaneously, and 86% (1771 ⁄ 2057) of those who
The survival data were analysed in Cox-proportional had received both MV and DTP had received MV and
hazards models with age as the underlying time and DTP simultaneously or at least one DTP after MV.
reported as mortality rate ratios (MMRs) with 95% Among children aged 3 years or older at enrolment,
confidence interval (CI). We also analysed the data using subsequent vaccination intensity was limited. In the
time since visit as the timescale in a Cox model and following analysis, we have therefore focused on children
controlled for age as a categorised variable (6–11, 12–23 under 3 years of age in the placebo group (N = 3082;
and 24–35 months). The results are essentially the same as Figure 1).
in the Cox model using age as the timescale (see Appendix).
Cox-proportional hazard assumptions were tested and
Vaccinations during follow-up
were not violated. Background factors associated with a
P < 0.10 were included in the final model using backward The vaccination intensity in the first 4 months after
selection. The study area was divided into four geograph- enrolment was high for children with a health card
ical Zones: North, South, East and West. As the WAZ (Table 2). During the first 4 months, 50% (408 ⁄ 817)
score was better for children who did not have a vaccina- received additional doses of DTP if they had a vaccination
tion card – possibly due to less accurate age assessment (see card but had not received DTP3 at enrolment (Table 2,
Appendix) – we also controlled for WAZ in all analyses. Group 2, 3 and 5); 76% (621 ⁄ 817) received DTP within
Data were analysed in Stata 11. 12 months. Among children who had not had MV at
To minimise the effect of survival bias, we used a enrolment, 50% (276 ⁄ 549) received this vaccine during the
landmark approach with vaccinations status at enrolment next 4 months (Table 2, Group 2, 3 and 4); 79%
being a fixed time variable during follow-up (Jensen et al. (431 ⁄ 549) received MV within 12 months. In contrast,
2007). Survival bias leads to differential misclassification children who had received DTP3-4 before enrolment
Table 1 Vaccination coverage at enrolment into the trial according to age in the placebo group
Coverage of vaccines
Children
Vaccine All children with card Coverage by age among children with a health card n (%)
5–11 months 12–17 months 18–23 months 24–35 months 36+ months
5974 3397 N = 358 N = 484 N = 409 N = 738 N = 1408
All n (%) n (%) n (%) n (%) n (%) n (%) n (%)
BCG 2639 (44) 2639 (78) 305 (85) 409 (85) 353 (86) 589 (80) 983 (70)
BCG with DTP* 2003 (76)* 258 (85) 341 (83) 291 (82) 466 (79) 647 (66)
Any DTP 2795 (47) 2795 (82) 308 (86) 436 (90) 368 (90) 636 (86) 1047 (74)
DTP3 ⁄ DTP4 1503 (25) 1503 (44) 87 (24) 183 (38) 191 (47) 378 (51) 664 (47)
Any OPV 2781 (47) 2781 (82) 312 (87) 424 (88) 369 (90) 629 (85) 1047 (74)
OPV3-5 1426 (24) 1426 (42) 80 (22) 165 (34) 180 (44) 366 (50) 635 (45)
MV 2208 (37) 2208 (65) 86 (24) 266 (55) 307 (75) 556 (75) 993 (71)
MV with DTP* 1352 (66)* 64 (76) 179 (70) 206 (69) 374 (71) 529 (60)
DTP after MV* 1096 (53)* 8 (10) 73 (28) 132 (44) 294 (55) 589 (66)
237 had health cards 1093 did not have 1989 had health 11 had no information
but were not seen health cards cards seen on health card
207 deaths recorded at 2727 were alive 148 moved out of the
end of 2 years follow-up study area
Figure 1 Study population of children aged 6–35 months at enrolment into the placebo group.
Table 2 Proportion of children receiving vaccination during follow-up, according to vaccination status at enrolment, among children
<36 months of age at enrolment into the placebo group
Age (months)
at entry n (%) % Received DTP (N) % Received MV (N)
6–11
Vaccination status at 12–23
enrolment 24–35 First 4 months First year 2 years First 4 months First year
1 No health card 265 (24.3) 18 (152 ⁄ 863) 25 (214 ⁄ 863) 24 (169 ⁄ 714) 14 (124 ⁄ 863) 25 (218 ⁄ 863)
409 (37.4)
419 (38.3)
2 Health card 48 (24.5) 59 (73 ⁄ 123) 78 (96 ⁄ 123) 81 (87 ⁄ 108) 45 (55 ⁄ 123) 77 (95 ⁄ 123)
(no DTP, no MV) 72 (36.7)
76 (38.8)
3 Health card + DTP1 ⁄ 2 182 (38.8) 57 (191 ⁄ 335) 81 (271 ⁄ 335) 83 (250 ⁄ 301) 47 (158 ⁄ 335) 78 (262 ⁄ 335)
(no MV) 201 (42.9)
86 (18.3)
4 Health card + DTP3 ⁄ 4 42 (38.5) 10 (9 ⁄ 91) 11 (10 ⁄ 91) 13 (10 ⁄ 80) 69 (63 ⁄ 91) 81 (74 ⁄ 91)
(no MV) 47 (43.1)
20 (18.4)
5 Health card + MV 41 (8.5) 40 (144 ⁄ 359)* 71 (254 ⁄ 359)* 74 (233 ⁄ 315)* NA NA
and DTP0-2 246 (50.7)
198 (40.8)
6 Health card + MV 45 (6.2) 5 (30 ⁄ 572)* 6 (33 ⁄ 572)* 5 (28 ⁄ 514)* NA NA
and DTP3-4 327 (44.8)
358 (49.0)
Group 1 consisted of those with no health card and was considered ‘unvaccinated’. Groups 2–6 all had health cards at Round 1.
Group 2 had not received DTP or MV (some of these children had, however, received BCG); Group 3 had received DTP1-2 but not
MV; Group 4 had received DTP3-4, but not MV; Group 5 had received DTP1-2 and MV; and Group 6 had received DTP3-4 and MV
(fully vaccinated).
*DTP vaccinations after MV.
Table 3 Background characteristics of children with and without a health card, among children <36 months of age at enrolment into
the placebo group
Sex (Males ⁄ N) 49.5% (984 ⁄ 1989) 50.9% (557 ⁄ 1093) 0.94 (0.81, 1.09)
Median age (months) 20.0 [IQR 13.6; 27.5] 19.6 [IQR 12.3; 28.0] P = 0.163*
Ever breastfed 100% (1989 ⁄ 1989) 99.9% (1092 ⁄ 1093) P = 0.164*
Still breastfeeding 81.0% (1612 ⁄ 1989) 75.4% (824 ⁄ 1093) 1.40 (1.16, 1.67)
Mean arm circumference MUAC z-scores )1.47 ()1.52; )1.41) )1.40 ()1.47; )1.33) )0.07 ()0.16, )0.02)
Mean weight-for-age z-score )1.94 ()2.00; )1.88) )1.67 ()1.76; )1.58) )0.27 ()0.37, )0.16)
BCG scar 80.9% (1609 ⁄ 1988) 15.0% (164 ⁄ 1093) 24.0 (19.6, 29.5)
Ever had measles before enrolment 2.0% (39 ⁄ 1989) 1.9% (21 ⁄ 1091) 1.02 (0.58, 1.83)
Previously admitted to hospital 4.6% (92 ⁄ 1989) 2.5% (27 ⁄ 1093) 1.91 (1.23, 3.08)
Radio in compound 31.3% (622 ⁄ 1985) 18.6% (202 ⁄ 1089) 2.0 (1.67, 2.41)
breastfed. The children with a health card had lower children without a health card having been underesti-
weight-for-age z-scores (WAZ) than children with no card. mated. Also, children with a card were more likely to have
The difference in WAZ could be due to current age of the been hospitalised, presumably indicating that communities
with high coverage had better contact with health services. DTP3-4 but no MV at enrolment (Group 4) had similar
Fifteen per cent of the children without a health card had a mortality as fully vaccinated children (Group 6), and
BCG scar. These children may have lost the card or never roughly half the mortality rate compared with the children
been issued with one. without a card both after 4 months and 2 years, though
neither of these differences were statistically significant
(Table 5). Hence, missing vaccines was not per se a risk
Mortality of different vaccination groups
factor for child mortality. The prevention of measles deaths
In the first 4 months, the MRR for children with a did not appear to explain the lower mortality among the
health card compared with children without a card and measles-vaccinated children and the children with DTP3-4
adjusted for zone, WAZ and ownership of a radio was but no MV (Table 5).
0.97 (95%CI 0.61, 1.56); 1.10 (95%CI 0.56, 2.16) for In contrast, children with 1–2 doses of DTP but no MV
vaccinated girls and 0.86 (95%CI 0.44, 1.69) for vacci- at enrolment (Table 5, Group 3) had a higher overall
nated boys. Over the 2 years of follow-up, the children mortality than children without a card (Group 1) during
with a card had significantly lower mortality than children the first 4 months of follow-up [MRR = 1.65 (95%CI
without a card [MRR = 0.61 (0.46, 0.82)]. 0.95, 2.87)], with mortality being significantly higher for
Children who had received MV by enrolment had deaths that were not attributed to measles [MRR = 1.74
slightly better WAZ than children who had received at (95%CI 1.00, 3.05)]. The DTP1-2-vaccinated children
least one dose of DTP but not MV (DTP-vaccinated (Group 3) also had a twofold higher mortality rate than the
children), the RR of WAZ < )2 being 0.93 (95%CI 0.88, fully vaccinated children (Group 6) in the first 4 months
1.00) for MV children. Controlled for WAZ, over the first [MRR = 2.38 (95%CI 1.07, 5.30)] and over the 2 years
4 months, the children who were MV-vaccinated at [MRR = 2.41 (95%CI 1.41, 4.15)].
enrolment had half the mortality of the children who were Controlling for MV status (Table 6), we compared
initially only DTP-vaccinated [MRR = 0.51 (95%CI 0.27, children with missing doses of DTP (Table 5, Groups 2, 3
0.97)] (Table 4). The reduction in mortality could not be and 5) with children who had already received DTP3-4 at
explained by protection against measles deaths, as defined enrolment (Table 5, Groups 4 and 6). Those missing doses
by the verbal autopsies, since the MRR was 0.46 (95%CI of DTP tended to have higher mortality during the first
0.23, 0.89) when measles deaths were excluded from the 4 months [MRR = 1.67 (95%CI 0.82, 3.43)] and had
analysis (Table 4). significantly higher mortality over the full duration of the
As vaccination status group at enrolment influenced the study [(MRR = 1.85 (95%CI 1.16, 2.95)] than those with
likelihood of receiving additional vaccinations during DTP3-4 at enrolment (Table 6).
follow-up (Table 2), we also assessed the relative mortality
of these different vaccination status groups (Table 5). Fully
Discussion
vaccinated children (Group 6) had lower mortality than
children without a card (Group 1), and this effect was The data from Navrongo provided several indications that
significant for the 2-year follow-up period [MRR = 0.37 vaccines may have important NSE on child survival. The
(95%CI 0.23, 0.62)]. The small group of children with children who had received MV at enrolment had twofold
Table 4 Mortality rate ratio during the first 4 months of follow-up for children with MV at enrolment vs. children with DTP without MV
at enrolment among children aged 6–35 months at enrolment into the placebo group
MV at enrolment 33 (7 ⁄ 209) [593] {1} 45 (10 ⁄ 220) [622] {1} 40 (17 ⁄ 429) [1215] {2}
(Table 2, Group 5 and 6)
DTP (No MV at enrolment) 127 (13 ⁄ 102) [289] 139 (14 ⁄ 101) [289] 133 (27 ⁄ 203) [578]
(Table 2, Group 3 and 4)
MRR MV ⁄ DTP 0.50 (0.19, 1.31) 0.52 (0.22, 1.23) 0.51 (0.29, 0.97)
(No MV at enrolment)*
MRR* (measles deaths excluded) 0.44 (0.16, 1.22) 0.48 (0.20, 1.16) 0.46 (0.23, 0.89)
weight-for-age 1000 years MRR (95% CI) No heath card) MR per 1000 years MRR (95% CI) health card)
Group: Vaccination z-score at (deaths ⁄ pyrs) {MRR – without {MRR – without (deaths ⁄ pyrs) {MRR – without {MRR – without
status at enrolment N enrolment {measles deaths} measles deaths) measles deaths} {measles deaths} measles deaths) measles deaths}
1. No card 1093 )1.7 [)2.6; )0.8] 80 (31 ⁄ 389) {2} 1.0 (Reference) 1.0 (Reference) 51 (101 ⁄ 1998) {23} 1.0 (Reference) 1.0 (Reference)
2. Health card: 196 )1.9 [)2.9; )1.0] 58 (4 ⁄ 69) 0.72 (0.25, 2.05) 0.69 (0.24, 1.97) 30 (11 ⁄ 370) {2} 0.59 (0.32, 1.10) 0.55 (0.29, 1.02)
noDTP, noMV {0.77 (0.27, 2.19)} {0.75 (0.26, 2.13)} {0.63 {0.59 (0.29, 1.17)}
(0.31, 1.25)}
3. DTP1-2, no MV 469 )2.0 [)3.0; )1.2] 152 (25 ⁄ 164) 1.90 (1.12, 3.22) 1.65 (0.95, 2.87) 54 (46 ⁄ 854) {6} 1.07 (0.75, 1.51) 0.90 (0.63, 1.30)
{2.03 (1.19, 3.47)} {1.74 (1.00, 3.05)} {1.19 {1.03 (0.69, 1.52)}
(0.82, 1.75)}
4. DTP3-4, no MV 109 )1.8 [)3.0; )1.2] 52 (2 ⁄ 38) 0.66 (0.16, 2.74) 0.57 (0.13, 2.42) 24 (5 ⁄ 207) {1} 0.48 (0.20, 1.19) 0.44 (0.18, 1.08)
{0.70 (0.17, 2.94)} {0.61 (0.14, 2.61)} {0.50 {0.46 (0.17, 1.28)}
(0.18, 1.36)}
5. MV, DTP0-2 485 )1.9 [)2.8; )1.1] 47 (8 ⁄ 171){2} 0.58 (0.27, 1.27) 0.80 (0.36, 1.78) 26 (24 ⁄ 923) {3} 0.52 (0.33, 0.81) 0.62 (0.39, 0.97)
{0.47 (0.19, 1.23)} {0.66 (0.27, 1.61)} {0.58 {0.75 (0.46, 1.23)}
(0.36, 0.94)}
6. MV, DTP3-4 730 )1.8 [)2.6; )1.0] 35 (9 ⁄ 257) 0.44 (0.21, 0.92) 0.70 (0.32, 1.53) 14 (20 ⁄ 1408) 0.28 (0.18, 0.46) 0.37 (0.23, 0.62)
{0.47 (0.22, 0.98)} {0.76 (0.34, 1.69)} {0.36 {0.51 (0.30, 0.85)}
(0.22, 0.60)}
P. Welaga et al. Non-specific effects of diphtheria-tetanus-pertussis and measles vaccinations
Table 6 Mortality during follow-up according to number of doses of DTP received before enrolment, among children aged 6-35 months
who had a health card at enrolment into the placebo group
4 months of follow-up
DTP0-2 124 (29 ⁄ 234) [665] {0} 47 (8 ⁄ 171) [485] {2}
DTP3-4 53 (2 ⁄ 38) [109] 35 (9 ⁄ 257) [730]
MRR (DTP0-2 ⁄ DTP3-4)* 2.29 (0.53, 9.78) 1.08 (0.38, 3.06)
[without measles deaths] [2.29 (0.53, 9.78)] [0.88 (0.30, 2.59)]
MRR (DTP0-2 ⁄ DTP3-4)* 1.67 (0.82, 3.43)
[without measles deaths] [1.60 (0.78, 3.28)]
24 months of follow-up
DTP0-2 47 (57 ⁄ 1224) [665] {8} 26 (24 ⁄ 923) [485] {3}
DTP3-4 24 (5 ⁄ 207) [109] {1} 14 (20 ⁄ 1408) [730]
MRR (DTP0-2 ⁄ DTP3-4)* 1.89 (0.75, 4.77) 1.70 (0.91, 3.17)
[without measles deaths] [1.93 (0.69, 5.41] [1.53 (0.81, 2.90]
MRR (DTP0-2 ⁄ DTP3-4)* 1.85 (1.16, 2.95)
[without measles deaths] [1.68 (1.03, 2.73)]
*Adjusted for age, zone, weight-for-age and ownership of radio and MV status at enrolment.
lower mortality than children who had received only relevant for assessing the impact of the primary series of
DTP and this difference was not explained by the DTP vaccinations administered before 9 months of age.
prevention of measles deaths. Furthermore, adjusted for Second, in Navrongo a large part of the population
MV vaccination status at enrolment, those with incomplete remained unvaccinated at that time, and therefore, it
doses of DTP had higher mortality than the children was possible to compare vaccinated with unvaccinated
who had already received DTP3 before enrolment. These children over a wide age range; the unvaccinated chil-
trends are consistent with observations made in several dren were not merely a small particularly frail subgroup,
other studies from Africa (Aaby et al. 1995, 2006a,b,c; which had been too weak to get vaccinated. Third, in
Kristensen et al. 2000). other studies, we have analysed the impact of a vaccine,
while it was presumed to be the latest vaccine received
(Aaby et al. 2004b,c, 2006a,b,c; Veirum et al. 2005).
Strengths and weaknesses
This was not possible in this study because the time interval
This was an observational study of the effects of vaccina- for collection of vaccination data was too long. Hence,
tion. Misclassification of vaccination status did occur. A vaccinations were evaluated as risk factors for mortality
small proportion of vaccinated children were probably taking into consideration the initial vaccination status and
wrongly classified as unvaccinated. Information on vacci- whether additional vaccines may have been given during
nation dates may also have been missed or noted wrongly follow-up.
for a few of the children. Assuming these misclassifications It could be speculated that the increased mortality
to be random, such errors should tend to make the associated with DTP merely represented poor social
estimated differences more conservative. conditions, disorganised parenting, differential ascertain-
The Navrongo data set is special in several ways. First, ment of vaccinations or differential access to the EPI
the study only included children older than 6 months and programme. However, the DTP-vaccinated children would
there was very little follow-up time and only three deaths have had better families or more compliant parents than
before 9 months of age when BCG and DTP will usually the unvaccinated children. Nonetheless, the DTP-
have their strongest NSE. Hence, the DTP vaccines vaccinated children had higher mortality than the unvac-
administered in the present study were likely to have been cinated children. The late 1980s when these data were
given out-of-sequence, that is, DTP co-administered with collected was the introduction period for the EPI
MV or DTP given after MV. The study is therefore less programme, and many vaccines were given in campaigns
and out-of-sequence. The process of who got vaccinated NSE of vaccines should be interpreted with caution.
when are likely to have been much more random depend- Furthermore, selection biases may have explained some
ing on when there was a drive to organise vaccinations; for of the mortality differentials, but are unlikely to explain
example, the incidence of vaccinations was very high just all of them because the mortality effect of different
after the initiation of the trial. vaccines went in opposite directions. Specifically, the
Navrongo data set documented similar differential
effects of DTP and MV as have been reported in several
Interpretation: differential mortality impact of DTP and
previous studies (Velema et al. 1991 Aaby et al. 1995,
MV vaccinations
2002, 2006c, 2012; Veirum et al. 2005).
Compared with children who were MV-vaccinated at Following studies suggesting a possible negative effect
enrolment, the DTP-vaccinated but MV-unvaccinated of DTP on survival (Aaby et al. 1995; Kristensen et al.
children had twofold higher mortality. Using vaccination 2000), several WHO-sponsored studies from Africa
status as a predictor of subsequent vaccinations we found (Nyarko et al. 2001; Vaugelade et al. 2004; Elguero
that the DTP0-2-vaccinated children had the highest et al. 2005) have argued that DTP had strong beneficial
mortality, whereas those who had already received MV effects, including a study from Navrongo based on data
and DTP3 had the lowest mortality, the difference being collected in 1996–2000 (Nyarko et al. 2001). However,
2.5-fold. these studies had survival bias, with the information on
Interestingly, the prevention of deaths attributed to vaccinations received being better for children who
measles by verbal autopsy explained little of this reduction survived (Aaby et al. 2007; Fine & Smith 2007; Jensen
in mortality. Similar observations have been made in et al. 2007) because health cards for dead children were
previous studies from Guinea-Bissau (Aaby et al. 1996b), often not inspected for vaccination information after the
Senegal (Aaby et al. 1996a) and Bangladesh (Aaby et al. death of a child. The family often destroys the card after
2003a). MV-vaccinated children had worse WAZ z-score the child’s death making vaccine status information for
than the unvaccinated children, and it seems unlikely that the dead children incomplete. With survival bias, vacci-
the beneficial effect of MV is merely owing to a positive nation will automatically be associated with a strong
selection bias. As supported by previous studies (Aaby beneficial effect.
et al. 1995), MV may have a beneficial effect which is not These data were collected 20 years ago when most
explained by the prevention of measles infection. children received vaccinations out-of-sequence and most
On the other hand, being only DTP-vaccinated at had DTP as the last vaccination, and it could be
enrolment was associated with higher mortality than the speculated whether they had any relevance in the current
rates of both unvaccinated and fully vaccinated children. situation. However, it needs to be emphasised that there
These children were likely to receive both MV and DTP are still many children getting vaccines out-of-sequence
during follow-up and usually administered simultaneously. in rural Africa, particularly in states with less well-
Previous research has suggested that children who had had organised EPI (Aaby & Benn 2009b). Furthermore, the
simultaneous administration of MV and DTP had higher problem of having DTP as the last vaccination may
subsequent mortality than children who had received MV become more common again. WHO’s SAGE (the Stra-
on its own as the most recent vaccine (Aaby & Benn tegic Advisory Group of Experts on immunisation) has
2009b). Also, in a small randomised study, girls rando- recently recommended a booster dose of pertussis vaccine
mised to receive MV and DTP simultaneously had signif- given in the second year of life (WHO 2010), which
icantly poorer growth than children randomised to receive would again make DTP the last vaccination through
MV only (Agergaard et al. 2011). In the present study, the most of childhood.
small group of DTP3-vaccinated children who were likely
to receive only MV during follow-up also had low
Differences between sexes
mortality. Control for baseline nutritional status did not
substantially reduce the mortality differences between the In most studies of NSE of vaccines, we have emphasised
various vaccination groups. The possibility that additional that they tend to differ for boys and girls (Aaby et al. 1995,
DTP vaccinations during follow-up may have had a 2002, 2004b, 2006a,c, 2007, 2012; Benn et al. 2009). The
negative effect was strengthened by the fact that children data set in this study was too small to have been able to
with incomplete DTP doses tended to have higher mortality detect anything but very large differences. There may be
than children with complete DTP doses at enrolment. other reasons that the sex-differential pattern was less clear
Given the uncertainty regarding who received additional in the present study. Firstly, because of the infrequent
vaccinations during follow-up, the Navrongo data on the collection of vaccination status data, we could not evaluate
the NSE of the last known vaccine that the child received Aaby P, Bhuyia A, Nahar L, Knudsen K, Francisco A & Strong M
with certainty. During follow-up, the children may have (2003a) The survival benefit of measles immunisation may not
received different vaccines with opposite sex-differential be explained entirely by the prevention of measles disease.
effects, for example, MV after DTP or DTP after MV. International Journal of Epidemiology 32, 106–115.
Aaby P, Jensen H, Samb B et al. (2003b) Differences in female-
Secondly, the majority of children in Navrongo received a
male mortality after high-titre measles vaccine and association
live and inactivated vaccine simultaneously, that is, BCG
with subsequent vaccination with diphtheria-tetanus-pertussis
and DTP or DTP and MV together. We have previously and inactivated poliovirus: a re-analysis of the West African
reported that a combined live and inactivated vaccine may studies. Lancet 361, 2183–2188.
be better for girls than for boys (Aaby et al. 2004a, 2009a). Aaby P, Jensen H, Rodrigues A et al. (2004a) Divergent female-
male mortality ratios associated with different routine vaccina-
tions among female-male twin pairs. International Journal of
Conclusion
Epidemiology 33, 367–373.
The present study adds to the growing evidence that the Aaby P, Jensen H, Gomes J, Fernandes M & Lisse IM (2004b) The
impact of routine vaccinations cannot be explained merely introduction of diphtheria-tetanus-pertussis vaccine and child
mortality in rural Guinea-Bissau: an observational study.
by prevention of targeted diseases or selection biases (Aaby
International Journal of Epidemiology 33, 374–380.
et al. 2007) and that DTP and MV may have different
Aaby P, Rodrigues A, Biai S et al. (2004c) Oral polio vaccination
effects for child survival (Velema et al. 1991; Kristensen and low case fatality at the paediatric ward in Bissau, Guinea-
et al. 2000; Aaby et al. 2002, 2004a, 2006a,c, 2009a, Bissau. Vaccine 22, 3014–3017.
2012; Veirum et al. 2005; Benn et al. 2009). Randomised Aaby P, Jensen H & Walraven G (2006a) Age-specific changes in
controlled trials have documented that BCG and MV have female-male mortality ratio related to the pattern of vaccina-
non-specific beneficial effects (Aaby et al. 2010, 2011; tions: an observational study from rural Gambia. Vaccine 24,
Roth et al. 2010). There is a need for others to assess the 4701–4708.
impact of routine vaccinations in further randomised trials Aaby P, Ibrahim S, Libman M & Jensen H (2006b) The sequence
measuring mortality to improve the vaccination policy of vaccinations and increased female mortality after high-titre
currently implemented in high-mortality low-income measles vaccine: trials from rural Sudan and Kinshasa. Vaccine
24, 2764–2771.
countries.
Aaby P, Vessari H & Nielsen J et al. (2006c) Non-specific and sex-
differential effects of routine immunizations in rural Malawi.
Acknowledgements The Pediatric Infectious Disease Journal 25, 721–727.
Aaby P, Benn CS, Nielsen J, Lisse IM, Rodrigues A & Jensen H
Funding was provided by grants from DANIDA and
(2007) Estimating the effect of DTP vaccination on mortality
European Union FP7 support for OPTIMUNISE. CB is in observational studies with incomplete vaccination data.
funded through an ERC Starting Grant. Tropical Medicine and International Health 12, 15–24.
Aaby P, Benn CS, Nielsen J & Ravn H (2009a) Sex-differential
References non-specific effects of BCG and DTP in Cebu, The Philippines.
International Journal of Epidemiology 38, 320–323.
Aaby P & Benn CS (2009b) Assessment of childhood immunisa- Aaby P, Martins CL, Garly ML et al. (2010) Non-specific effects
tion coverage? Lancet 373, 1428. of standard measles vaccine at 4.5 and 9 months of age on
Aaby P, Samb B, Simondon F, Coll Seck AM, Knudsen K & childhood mortality: randomised controlled trial. BMJ 341,
Whittle H (1995) Non-specific beneficial effect of measles c6495.
immunisation: analysis of mortality studies from developing Aaby P, Roth A, Ravn H et al. (2011) Randomised trial of BCG
countries. British Medical Journal 311, 481–485. vaccination at birth to low-birth-weight children: beneficial non-
Aaby P, Samb B, Andersen M & Simondon F (1996a) No long- specific effects in the neonatal period? The Journal of Infectious
term excess mortality after measles infection: a community Diseases 204, 245–252.
study from Senegal. American Journal of Epidemiology 143, Aaby P, Benn CS, Nielsen J, Lisse IM, Rodrigues A & Ravn H
1035–1041. (2012) Testing the hypothesis that diphtheria-tetanus-pertussis
Aaby P, Lisse I, Mølbak K, Knudsen K & Whittle H (1996b) vaccine has negative non-specific and sex-differential effects on
No persistent T lymphocyte immunosuppression or increased child survival in high-mortality countries. BMJ Open 2,
mortality after measles infection: a community study from e000707.
Guinea-Bissau. The Pediatric Infectious Disease Journal 15, Agergaard J, Nante E, Poulstrup G et al. (2011) Diphtheria-teta-
39–44. nus-pertussis vaccine administered simultaneously with measles
Aaby P, Garly ML, Balé C, Martins C, Lisse I & Jensen H (2002) vaccine is associated with increased morbidity and poor growth
Routine vaccinations and child survival in war situation with in girls. A randomised trial from Guinea-Bissau. Vaccine 29,
high mortality: effect of gender. Vaccine 21, 15–20. 487–500.
Benn CS, Aaby P, Nielsen J, Binka FN & Ross DA (2009) Does WHO (2010) Meeting of the Strategic Advisory Group of Experts
vitamin A supplementation interact with routine vaccinations? on immunization, April 2010 – conclusions and recommenda-
An analysis of the Ghana vitamin A supplementation trial. The tions. Weekly Epidemiological Record 85, 197–212.
American Journal of Clinical Nutrition 90, 629–639.
Binka FN, Ross DA, Morris SS et al. (1995) Vitamin A supple-
mentation and childhood malaria in northern Ghana. The Appendix
American Journal of Clinical Nutrition 61, 853–859. Ghana VAST Survival Study
Elguero E, Simondon F, Simondon K & Vaugelade J (2005) Non-
specific effects of vaccination on survival: a prospective study in The GhanaVAST Survival Study was conducted between
Senegal. Tropical Medicine and International Health 10, 1989 and 1991 (Ghana VAST Study Team 1993; Binka et al.
956–960. 1995; Ross et al. 1995; Benn et al. 2009). A census was
Fine PEM & Smith PG (2007) ‘Non-specific effects of vaccines’ – carried out shortly before the trial started. The study area
an important analytical insight, and a call for a workshop.
was exclusively rural and extended families lived in com-
Tropical Medicine and International Health 12, 1–4.
pounds. The area was divided in 185 clusters, which were
Ghana VAST Study Team (1993) Vitamin A supplementation in
northern Ghana: effects on clinic attendances, hospital admis-
randomised to vitamin A supplementation (VAS) or placebo
sions, and child mortality. Lancet 342, 7–12. http:// within a double-blind trial. Children aged 6–90 months of
www.who.int/childgrowth/software/en age were enrolled into an open cohort and were visited at
Jensen H, Benn CS, Nielsen J et al. (2007) Survival bias in home every 4 months by trained fieldworkers, in a total of 7
observational studies of the impact of routine vaccinations on rounds. They were dosed with vitamin A or placebo at each
childhood survival. Tropical Medicine and International Health round. Over the 2 years, 21 906 children were enrolled in
12, 5–14. the trial by receiving at least one dose of either VAS or
Kim-Farley RJ (1992) EPI for the 1990s. The Expanded Pro- placebo; 13 462 being enrolled in Round 1. Children who
gramme on Immunization Team. Vaccine 10, 940–948. moved out of the area were censored in the analysis at the
Kristensen I, Aaby P & Jensen H (2000) Routine vaccinations and
time of moving; children who moved from a cluster with one
child survival: follow up study in Guinea-Bissau, West Africa.
treatment to a cluster with the opposite treatment were
BMJ 321, 1435–1438.
Nyarko P, Pence B & Debpuur C (2001) Immunization Status and censored at the time of receiving the new treatment. Children
Child Survival in Rural Ghana. Population Council: Working who developed xerophthalmia were censored at the time of
papers No. 147, New York. diagnosis and received special intensive vitamin A treatment.
Ross DA, Kirkwood BR, Binka FN et al. (1995) Child morbidity VAS was associated with a 19% (95%CI, 2–32%) reduction
and mortality following vitamin A supplementation in Ghana: in mortality (Ghana VAST Study Team 1993).
time since dosing, number of doses and time of the year.
American Journal of Public Health 85, 1246–1251.
Roth A, Sodemann M, Jensen H et al. (2006a) Tuberculin reac- Age determination
tion, BCG scar and lower female mortality. Epidemiology 17, At enrolment, the date of birth of the child was determined
562–568.
from the health card if available. Otherwise, the date of
Roth AE, Garly ML, Jensen H, Nielsen J & Aaby P (2006b) Bacille
Calmette Guerin vaccination and infant mortality. Expert
birth was assessed by means of detailed local events
Review of Vaccines 5, 277–293. calendars. If no exact date could be obtained, the date were
Roth A, Benn CB, Ravn H et al. (2010) A randomised trial of the set to day 15 of the month of birth. The date of birth may
effect of revaccination with BCG in early childhood and have been determined more precisely among children with
mortality. BMJ 340, c671. a health card. For example, in Round 1, there were more
Vaugelade J, Pinchinat S, Guielle G, Elguero E & Simondon F children who had day 15 in a month as date of birth among
(2004) Lower mortality in vaccinated children: follow up study children without a health card (3772 ⁄ 5066, 74.5%) than
in Burkina Faso. BMJ 329, 1309–1311. among those with a card (2764 ⁄ 6656, 41.5%) [Prevalence
Veirum JE, Sodemann M, Biai S et al. (2005) Diphtheria-tetanus- ratio 1.79 (1.69, 1.91)], and the month of birth was
pertussis and measles vaccinations associated with divergent
grouped close to the date of the enrolment month or
effects on female and male mortality at the paediatric ward in
6 months earlier (P = 0.003), suggesting that the children
Bissau, Guinea-Bissau. Vaccine 23, 1197–1203.
Velema JP, Alihonou EJ, Gandaho T & Hounye FH (1991) had been reported to be for instance 1 year or 1½ year,
Childhood mortality among users and non- users of primary rather than a year and some months. This may have made
health care in a rural West African community. International the children in the group with no health card appear to
Journal of Epidemiology 20, 474–479. have been slightly younger than they truly were.
4. Health
Vaccination status at 2. Health card, 3. Health card, card, DTP3-4, 5. Health card, 6. Health card,
enrolment (Round 1) 1. No card no DTP, no MV DTP1-2, no MV no MV MV, DTP 0-2 MV, DTP3-4 Total
1. No card 1093 )1.7 80 (31 ⁄ 389) {2} 1.0 (Reference) 1.0 (Reference) 51 (101 ⁄ 1.0 1.0 (Reference)
[)2.6; )0.8] 1998) {23} (Reference)
2. Health card: 196 )1.9 58 (4 ⁄ 69) 0.72 (0.25, 2.05) 0.65 (0.23, 1.84) 30 (11 ⁄ 0.59 0.53 (0.28, 0.99)
noDTP, noMV [)2.9; )1.0] {0.77 (0.27, 2.19)} {0.71 (0.25, 2.04)} 370) {2} (0.32, 1.10) 0.63 {0.57 (0.28, 1.13)}
(0.31, 1.25)}
3. DTP1-2, 469 )2.0 152 (25 ⁄ 164) 1.90 (1.12, 3.22) 1.79 (1.03, 3.11) 54 (46 ⁄ 1.07 0.93 (0.65, 1.33)
no MV [)3.0; )1.2] {2.03 (1.19, 3.47)} {1.87 (1.07, 3.26)} 854) {6} (0.75, 1.51) {1.05 (0.71, 1.56)}
{1.19 (0.82, 1.75)}
4. DTP3-4, 109 )1.8 52 (2 ⁄ 38) 0.66 (0.16, 2.74) 0.65 (0.15, 2.74) 24 (5 ⁄ 0.48 0.45 (0.18, 1.11)
no MV [)3.0; )1.2] {0.70 (0.17, 2.94)} {0.68 (0.16, 2.90)} 207) {1} (0.20, 1.19) {0.47 (0.17, 1.30)}
{0.50 (0.18, 1.36)}
5. MV, DTP0-2 485 )1.9 47 (8 ⁄ 171){2} 0.58 (0.27, 1.27) 0.74 (0.34, 1.65) 26 (24 ⁄ 0.52 0.58 (0.37, 0.91)
[)2.8; )1.1] {0.47 (0.19, 1.23)} {0.60 (0.24, 1.47)} 923) {3} (0.33, 0.81) {0.70 (0.42, 1.14)}
{0.58 (0.36, 0.94)}
6. MV, DTP3-4 730 )1.8 35 (9 ⁄ 257) 0.44 (0.21, 0.92) 0.61 (0.28, 1.35) 14 (20 ⁄ 0.28 0.35 (0.21, 0.57)
[)2.6; )1.0] {0.47 (0.22, 0.98)} {0.67 (0.31, 1.48)} 1408) (0.18, 0.46) {0.47 (0.28, 0.79)}
{0.36 (0.22, 0.60)}
Total 3082 )1.8 73 (79 ⁄ 1090){4} 36 (207 ⁄
[)2.7; )1.0] 5760){35}
P. Welaga et al. Non-specific effects of diphtheria-tetanus-pertussis and measles vaccinations
Corresponding Author Paul Welaga, Navrongo Health Research Centre, PO Box 114, Navrongo, Ghana. E-mail: pwelaga@yahoo.com
1505
volume 17 no 12 pp 1492–1505 december 2012