The relationship between nutritional and sociodemographic factors ...

Investigación original / Original research
The relationship between nutritional and
sociodemographic factors and the likelihood
of children in the Dominican Republic
having a BCG scar
Eddy Pérez-Then,1 Gail Shor-Posner,2 Lee Crandall,2 and James Wilkinson2
Suggested citation
Pérez-Then E, Shor-Posner G, Crandall L, Wilkinson J. The relationship between nutritional and socio-
demographic factors and the likelihood of children in the Dominican Republic having a BCG scar. Rev
Panam Salud Publica. 2007;21(6):365–72.
ABSTRACT
Objectives.
To critically assess the prevalence among schoolchildren 6 to 9 years of age
throughout the Dominican Republic of a bacille Calmette-Guérin (BCG) vaccination scar, and
to examine the relationship between nutritional and sociodemographic factors and the likeli-
hood of having a BCG scar.
Methods.
This correlational study used the database of the Second National Census on
Height and Weight of Elementary School First Grade Students, which was conducted in the
Dominican Republic August 2001–May 2002, to provide a critical assessment of BCG cover-
age nationwide. The Census information for the children included the presence of BCG scar,
their nutritional status, and basic demographic data. We developed a new sociodemographic
indicator, the “Rosa Index,” to examine the potential influence of poverty and other environ-
mental characteristics on scar presence. We used logistic regression models to predict the pres-
ence of a BCG scar.
Results.
An overall BCG scar prevalence of 55.3% (85 644/154 887) was found. Malnour-
ished children were less likely to have a BCG scar than were children with adequate nutritional
status (odds ratio = 0.91; 95% confidence interval: 0.87, 0.95, P < 0.05). Children who were
7–9 years old were less likely to have a BCG scar than were children 6 years old. Children in
the areas of the country more than two hours’ driving distance from the capital city of Santo
Domingo more often exhibited lower BCG scar prevalence levels than did children in Santo
Domingo. A higher Rosa Index (better level of socioeconomic characteristics) was correlated
with higher BCG scar prevalence values (r = 0.54, P < 0.05).
Conclusions.
Our study findings indicate that BCG coverage appears to be inadequate for
schoolchildren in the Dominican Republic. Nevertheless, the presence of a scar in a higher pro-
portion of younger children suggests that coverage has improved in recent years. More pro-
grammatic and economic emphasis needs to be placed on extending early BCG vaccination
coverage to the areas of the country where vaccination coverage is lower, and on examining the
potential role that poverty may have on vaccination effectiveness.
Key words
Tuberculosis, BCG vaccine, cicatrix, vaccination, population surveillance, socio-
economic factors, Dominican Republic.
1
National Research Center on Maternal and Child
respondence to: Eddy Pérez-Then; e-mail: ept26@
2
Department of Epidemiology and Public Health,
Health (CENISMI), Avenida Abraham Lincoln #2
yahoo.com; telephone: (809) 533-2873; fax: (809)
Miller School of Medicine, University of Miami,
Esq. Ave. Independencia, Centro de Los Héroes,
532-6450.
Miami, Florida, United States of America.
Santo Domingo, República Dominicana. Send cor-
Rev Panam Salud Publica/Pan Am J Public Health 21(6), 2007
365

---

Original research
Pérez-Then et al. • BCG scar in the Dominican Republic
Bacille Calmette-Guérin (BCG) vac-
tuberculosis meningitis (less than 2
using standard anthropometric meth-
cination has been used to prevent tu-
years of age) per 100 000 cases in the
ods. Height was calculated with a lo-
berculosis (TB) since 1921, and it was
Dominican Republic increased from
cally made centimeter measuring
incorporated in the World Health Or-
0.1 per 100 000 cases in 2004 to 0.7 per
board placed on the wall at a 90-
ganization (WHO) Expanded Program
100 000 cases in 2005 (11).
degree angle, as recommended by the
on Immunization (EPI) in 1974 to
This study was conducted to assess
Pan American Health Organization.
strengthen the fight against childhood
nationwide BCG scar prevalence in
The interviewer examined both of
TB in developing countries (1–3).
Dominican children, and it also exam-
the child’s arms for the presence of a
Though there are controversies re-
ined the relationship between nutri-
BCG scar and recorded the scar’s pres-
garding the vaccine’s efficacy, it is
tional and sociodemographic factors
ence or absence on a data collection
generally agreed that it is effective
and the likelihood of having a BCG
form. Basic demographic data (age,
against disseminated disease and
scar.
sex, and place of residence (school re-
meningitis in childhood TB (4–7). Fa-
gion)) of the subject were obtained, as
vorable outcomes of BCG vaccination
well as information about the legal
have also been reported in the preven-
METHODS
guardian’s socioeconomic status. In-
tion of leprosy (8) and in the treatment
formation about the school building,
of superficial bladder carcinoma (7).
National Census on Height
such as type of toilet, trash collection,
The presence or absence of a BCG
and Weight of Elementary School
and type of construction (wood or
scar is often used in clinical settings as
First Grade Students
cement), was also collected, and re-
an indicator of vaccine effectiveness,
corded on the data collection form.
and the same is done in surveys car-
In August 2001–May 2002, Soriano
The Census on Height and Weight
ried out by EPI and other health pro-
et al. carried out the Second National
data included the prevalence of mal-
grams in order to assess vaccine up-
Census on Height and Weight of Ele-
nutrition by school region. To estimate
take (9). However, BCG scar as a
mentary School First Grade Students
nutritional parameters (height-for-
measure of vaccine coverage is still a
(II Censo Nacional de Talla y Peso en Es-
age), percentiles were calculated using
subject of controversy. It has been re-
colares de Primer Grado de Básica) in the
the United States National Center for
ported that approximately 10% of chil-
Dominican Republic (13). All children
Health Statistics tables with < 2 stan-
dren successfully immunized with
from 6 to 9 years of age attending pub-
dard deviations as the cutpoint for
BCG don’t develop a scar. Discrepan-
lic or private elementary school were
malnutrition estimators.
cies that are found in vaccine effective-
included in the Census on Height and
ness can be explained by vaccine char-
Weight. A trained school teacher inter-
acteristics (nonpotent BCG strain dose,
viewed the caregivers of the children,
A secondary data analysis of the
or quality), poor vaccination tech-
who accompanied the children to the
BCG scar prevalence in Dominican
niques, recipient characteristics (e.g.,
school. The interviews typically lasted
children
age at vaccination, immunity and nu-
30 minutes and were performed in a
tritional status, race), and information
private classroom in the school.
This research was a correlational,
bias or falsified child records (10).
A pretest was conducted with 50
secondary data analysis study in-
The Dominican Republic has a high
schoolchildren from the different
tended to determine the effect of so-
TB incidence rate, approximately 45
school regions of the Dominican Re-
ciodemographic and nutritional para-
cases per 100 000 inhabitants (11). BCG
public. (There are 16 school regions in
meters on the likelihood of having
vaccination has been carried out in ac-
the Dominican Republic, with each
BCG scar.
cordance with WHO guidelines, under
region comprised of different prov-
We performed the statistical analy-
the aegis of the immunization pro-
inces and municipalities (Figure 1).
sis of the collected data using SPSS
gram of the Ministry of Health. Since
The school regions were created by
11.5 (SPSS, Chicago, Illinois, United
1978, intradermal administration of
the Secretariat of Education to fa-
States of America) and NCSS 2004 and
BCG has been recommended at birth
cilitate logistics (food distribution,
PASS 2002 (NCSS, Kaysville, Utah,
in any hospital, health post, or other
data collection, etc.) among rural and
United States). BCG prevalence pro-
health facility. Since 1991, the same
urban areas of the country). As part of
portions were calculated, and the ef-
strain of BCG has been used across the
the training procedures, school teach-
fects of the study variables (basic de-
country (11).
ers were instructed on how to recog-
mographic data, residence of the
As of 2005, the Dominican health au-
nize the BCG scar, perform anthropo-
subject, and nutritional indicators) on
thorities estimated the BCG coverage
metric measures, and conduct the
the likelihood of vaccination (as indi-
in children less than 2 years of age to be
interviews. Methods were standard-
cated by the scar’s presence) were esti-
98% (12). Despite this high vaccination
ized for filling out and reviewing the
mated by means of odds ratios (ORs)
rate, and an HIV prevalence among the
study forms.
and 95% confidence intervals (95%
general population (15–60 years of age)
The teacher interviewers were trained
CIs), obtained through logistic regres-
of 1% (12), the incidence of pediatric
to obtain measurements of height,
sion, employing univariate and multi-
366
Rev Panam Salud Publica/Pan Am J Public Health 21(6), 2007

---

Pérez-Then et al. • BCG scar in the Dominican Republic
Original research
FIGURE 1. BCG scar prevalence among schoolchildren in the Dominican Republic, by school region, 2001–2002
variate analyses (14). Logistic regres-
The Rosa Index. To evaluate the im-
sociodemographic data collected in
sion models were also developed to
pact of sociodemographic ecological
the Census on Height and Weight in
predict the presence of a BCG scar. In
variables on scar presence, a new
August 2001–May 2002. Possible index
each model, the category most posi-
quantitative sociodemographic indica-
scores could range from zero (mini-
tively associated with the presence of
tor, the “Rosa Index,” was developed
mum) to 12 (maximum). The higher
BCG scar was generally used as the
by the main author (E. Pérez-Then) in
the index score, the better the “eco-
reference category.
collaboration with Rosa Urania Abreu
logical” characteristics of the school
To facilitate interpretation of the
and the authors of the Census on
region. The Rosa Index was also in-
school data for univariate and multi-
Height and Weight. (The term “Rosa”
cluded in a multivariate model to pre-
variate models, the school regions
was used to acknowledge the contri-
dict the presence of BCG scar in Do-
were grouped into three geographic
bution of Rosa Urania Abreu in devel-
minican children.
categories: (1) Santo Domingo (refer-
oping the index.) This new approach
ence group), which is the capital of the
assigned a possession score to every
Multivariate analysis. For the multi-
Dominican Republic; (2) school re-
school region based on the mean val-
variate analysis, variables were en-
gions one to two hours’ driving dis-
ues of monthly household income, as
tered according to their statistical sig-
tance from the capital (“Mid School
well as the mean proportional values
nificance (in univariate analysis) and
Regions”); and (3) school regions more
of residential and schools characteris-
their biologic relevance. Various mod-
than two hours’ driving distance from
tics for each region. The Rosa Index
els were evaluated, including a full
the capital (“Other School Regions”).
values were assigned according to the
model containing all the variables
Rev Panam Salud Publica/Pan Am J Public Health 21(6), 2007
367

---

Original research
Pérez-Then et al. • BCG scar in the Dominican Republic
studied, in order to select the best pos-
TABLE 1. Scar prevalence, nutritional status, age, and place of residence (school region),
sible model to fit the data. The good-
by gender, in study of BCG scar prevalence, Dominican Republic, 2001–2002
ness of fit of the logistic models was
assessed taking into consideration the
Males
Females
changes introduced into the log likeli-
Variable
No.a
%
No.a
%
hood function. Alpha was set at 0.05.
BCG scar
Yes
45 567
55.3
40 077
55.3
Ethical issues. Ethical approval for the
No
36 826
44.7
32 417
44.7
study was obtained, in the Dominican
Nutritional statusb
Malnutrition
6 568
7.4
4 015
5.5
Republic, from the Human Research
Adequate
82 393
92.6
68 479
94.5
Subject Committee of the Centro Na-
Age (yr)b
cional de Investigaciones en Salud Ma-
6
34 838
45.7
34 083
50.8
terno Infantil (CENISMI ) (National Re-
7
24 633
32.2
21 246
31.7
search Center on Maternal and Child
8
11 469
15.0
8 158
12.2
9
5 512
7.2
3 626
5.4
Health), and, in the United States,
Place of residence (school region)
from the Institutional Review Board of
Santo Domingo (capital)
14 389
17.5
13 030
18.5
the University of Miami.
Mid School Regionsc
Santiago
7 391
8.9
6 816
9.4
La Vega
7 228
8.8
6 101
8.4
Azua
6 319
7.7
5 545
7.7
RESULTS
San Cristóbal
5 663
6.9
5 062
6.9
San Pedro de Macorís
5 735
6.9
5 003
6.9
BCG scar prevalence
Monte Plata
1 908
2.3
1 577
2.2
Other School Regionsd
The overall BCG scar prevalence
San Juan de La Maguana
5 148
6.3
4 421
6.1
Barahona
4 879
5.9
4 497
6.2
found through the examination of the
San Francisco de Macorís
4 424
5.4
3 743
5.2
children’s arms was 55.3% (85 644/
Cotui
4 151
5.0
3 464
4.8
154 887).
Puerto Plata
3 678
4.5
3 428
4.7
Higuey
3 194
3.9
2 751
3.8
Nagua
3 275
3.9
2 721
3.8
Monte Cristi
2 530
3.1
2 123
2.9
Sociodemographic characteristics
Mao
2 346
2.9
2 104
2.9
and BCG scar presence
a Total sample size is different for each variable (e.g., scar, nutritional status, age, and place of residence).
b P < 0.05. For nutritional status: odds ratio = 1.36; 95% confidence interval: 1.3, 1.4, P < 0.05. For age: χ2 = 601, P < 0.0001.
The characteristics of the children
c The Mid School Regions are ones that are one to two hours’ driving distance from the capital, Santo Domingo.
d
are shown, by gender, in Table 1.
Other School Regions are ones that are more than two hours’ driving distance from the capital, Santo Domingo.
There was a similar presence of BCG
scar among males and females. Statis-
tically significant gender differences
were found for nutritional status and
living in Mid School Regions (less than
pattern to fit the data. The multivariate
age (P < 0.05).
two hours’ driving distance from the
analysis revealed the same pattern of
capital) were more likely to have a
significance observed in the univariate
BCG scar than were children living in
analysis, except for nutritional status,
Univariate analysis
Santo Domingo, the capital of the Do-
which no longer had a statistically sig-
minican Republic (reference category).
nificant affect on the scar’s presence.
The univariate analysis showed as-
In contrast, children living in the
sociations between several variables
Other School Regions (more than two
and the presence of BCG scar (Table 2).
hours’ driving distance from the capi-
The Rosa Index
Malnourished children were less likely
tal) were generally less likely to pre-
to have BCG scar than were children
sent a scar than were children living in
To further elucidate variance in the
in the reference group (OR = 0.91; 95%
Santo Domingo (Table 2).
likelihood of having a BCG scar, the
CI = 0.87, 0.95). Age was also an indi-
Rosa Index was used (Table 3). The
cator, with children 7-9 years old being
Rosa Index evaluated the impact of 12
less likely to have a BCG scar than
Multivariate analysis
environmental variables (residence
were children 6 years old (the refer-
characteristics and school characteris-
ence group for age) (P < 0.05).
The multivariate analysis used a full
tics) on scar prevalence. The highest
Another factor that was associated
three-model variable since removal of
score for a particular school region
with scar presence was geographic lo-
each of the variables did not result in
could be 12 points, and the lowest
cation (Table 2). In general, children
a different and significantly improved
could be zero. The higher the Rosa
368
Rev Panam Salud Publica/Pan Am J Public Health 21(6), 2007

---

Pérez-Then et al. • BCG scar in the Dominican Republic
Original research
TABLE 2. Nutritional status, age, and place of residence (school region), by BCG scar
affects scar development through its
prevalence and univariate odds ratio, with 95% confidence interval (CI), Dominican Repub-
impact on the immune system (16).
lic, 2001–2002
However, from our study, it was not
possible to determine the schoolchil-
No.
BCG scar
Odds
dren’s nutritional status in infancy or
Characteristic
positive/total
(%)
ratio
95% CI
to ascertain if there was a relationship
Nutritional statusa
between this variable and the presence
Adequateb
80 755/145 462
55.5
1.00
of a BCG scar. This highlights the im-
Malnutrition
5 627/10 583
53.2
0.91
0.87, 0.95
portance of obtaining information re-
Age (yr)a
6b
40 453/69 328
58.4
1.00
garding vaccination history in future
7
24 900/46 209
53.9
0.75
0.73, 0.78
census programs implemented in the
8
10 176/19 799
51.4
0.73
0.71, 0.75
Dominican Republic.
9
4 461/9 204
48.5
0.67
0.64, 0.70
A second major limitation is related
Place of residence (school region)
to the representation of average expo-
Santo Domingo (capital)b
16 129/27 553
58.5
1.00
Mid School Regionsc
sure levels, rather than actual individ-
Santiagoa
9 845/13 503
72.9
1.53
1.46, 1.59
ual values. Thus, while it may appear
San Cristóbala
6 824/10 764
63.4
1.23
1.17, 1.28
from correlational data that there is an
Azuaa
7 284/11 898
61.2
1.12
1.07, 1.17
overall positive or negative associa-
San Pedro de Macorísa
5 934/10 874
54.6
0.85
0.81, 0.89
tion, this might actually be masking
La Vegaa
7 279/13 384
54.4
0.84
0.81, 0.88
Monte Plataa
1 256/3 516
35.7
0.39
0.37, 0.42
a more complicated relationship be-
Other School Regionsd
tween exposure and disease (15), in
San Francisco de Macorísa
5 341/8 212
65.0
1.32
1.25, 1.39
this particular case, vaccine adminis-
Nagua
3 529/6 069
58.1
0.98
0.93, 1.04
tration and the presence of BCG scar.
Mao
2 576/4 454
57.8
0.97
0.91, 1.04
Therefore, in this type of study design,
Higueya
3 121/6 026
51.8
0.76
0.72, 0.81
Monte Cristia
2 346/4 659
50.4
0.72
0.67, 0.76
malnutrition should be considered as
Cotuia
3 571/7 744
46.1
0.61
0.58, 0.64
an indicator of health inequalities
Barahonaa
4 134/9 421
43.9
0.59
0.56, 0.63
rather than as a disturbance in a par-
Puerto Plataa
3 085/7 154
43.1
0.54
0.51, 0.57
ticular individual’s immune system.
San Juan de La Maguanaa
3 985/9 656
41.3
0.50
0.47, 0.52
The univariate and multivariate anal-
a P < 0.05.
yses indicate an interesting age-related
b Reference group.
c The Mid School Regions are one to two hours’ driving distance from the capital, Santo Domingo.
pattern. Children from 7 through 9
d The Other School Regions are ones that are more than two hours’ driving distance from the capital, Santo Domingo.
years of age were less likely to have the
BCG scar than were children 6 years of
age. To some extent this may reflect
changes in the quality of the BCG vac-
Index score, the better were the “eco-
cates that a large proportion of the
cination program (e.g., periods of high
logical” characteristics of the school re-
Dominican children aged 6–9 years
and low coverage) caused by such fac-
gion. The correlation between the Rosa
old did not appear to have been vacci-
tors as vaccine storage conditions and
Index scores and the BCG scar preva-
nated against TB.
vaccine availability (17, 18). The higher
lences for the school regions was sta-
Several factors appeared to con-
proportion of young children with a
tistically significant (r = 0.54, P < 0.05).
tribute to the lack of BCG scar pres-
BCG scar may also reflect a general im-
When the Rosa Index was included
ence. Malnourished children were less
provement in vaccination coverage.
in the multivariate model (Table 4),
likely to have a BCG scar than were
Other factors at the individual or pro-
nutritional status was reduced in sig-
children with adequate nutritional sta-
grammatic level could also be associ-
nificance (P = 0.052), and the subjects’
tus, suggesting that population groups
ated with this trend, and they need to
age and the Rosa Index were indepen-
with better nutritional status may have
be considered in future studies, in
dently associated with the presence a stronger response to BCG vaccine, or
which the unit of analysis should be
of BCG scar (P < 0.05). Elevated Rosa
that poorer children may have been
students, and not school regions.
Index scores were found in school re-
less likely to be vaccinated and/or im-
There were also statistical differ-
gions with higher BCG scar prevalence
munized. In evaluating these findings,
ences in scar prevalence according to
values.
it is worth recalling common limita-
geographical location (school region).
tions of correlational studies, particu-
Better access to the health care ser-
larly the inability to link exposure with
vices in urban areas, greater availabil-
DISCUSSION
disease in particular individuals (15).
ity of pediatricians, and inability of
The higher likelihood of malnourished
the Dominican EPI to cover all regions
Our analysis of the data from the
children lacking a BCG scar raises the
of the country, especially the very
Census on Height and Weight indi-
possibility that malnutrition negatively
poor regions along the border with
Rev Panam Salud Publica/Pan Am J Public Health 21(6), 2007
369

---

Original research
Pérez-Then et al. • BCG scar in the Dominican Republic
TABLE 3. Factors and their point value for the Rosa Index possession score used to cate-
amined in this study and that will need
gorize school regions in the Dominican Republic according to the characteristics of the
to be considered in future research.
homes and the schools in each of the school regions, 2001–2002a
To increase understanding of other
variables that could be impacting cov-
Points
erage, the Rosa Index, which included
Factor
0
0.5
1.0
more “ecological” variables, was cre-
House characteristics
ated. Our results indicate a relatively
Monthly household income (< US$ 83)
> 80%
50%–80%
< 50%
strong correlation between BCG scar
Homeownership
< 35%
35%–60%
> 60%
prevalence and the Rosa Index (r =
Poor quality of the house
> 50%
35%–50%
< 35%
0.54), suggesting that BCG prevalence
Source of water is tap in the house
< 40%
40%–65%
> 65%
could be related to factors impacted by
Flush toilet in the house
< 10%
10%–20%
> 20%
Trash collection (public/government)
< 25%
25%–40%
> 40%
poverty and other environmental
Electricity
< 50%
50%–80%
> 80%
characteristics. Nutritional status was
Stove
< 60%
60%–80%
> 80%
only significant when the Rosa Index
School characteristics
was included in the multivariate
School with good (cement) construction
< 25%
25%–40%
> 40%
Flush toilet in the school
model. There was a positive correla-
< 25%
25%–40%
> 40%
Trash collection for the school (public/government)
< 20%
20%–35%
> 35%
tion pattern between BCG scar preva-
School receives foodb
> 70%
50%–70%
< 50%
lence and the Rosa Index, with the
a
school regions with a higher Rosa
Using the Rosa Index approach, a total possession score was assigned to each school region based on points given for the
mean value of monthly household income for the region as well as the residential characteristics and the school characteris-
Index score having a higher BCG scar
tics for each region. For example, a score of 0 points was given to a particular school region if more than 80% of the resi-
prevalence. This finding suggests a po-
dents reported having a monthly household income of less than US$ 83 (using the 2002 US$ exchange rate in the Domini-
can Republic). In contrast, a score of 1 point was given to a particular school region if more than 60% of the residents reported
tential role for poverty in vaccination
homeownership. The highest score for a particular school region could be 12 points, and the lowest could be zero. The higher
effectiveness.
the Rosa Index score was, the better were the “ecological” characteristics of the school region.
b The poorest areas were prioritized to receive food (breakfast) from the government.
It is important to note that a number
of factors that were not represented
in the Rosa Index could have had an
important impact on scar presence.
These factors include the type of legal
TABLE 4. Adjusted odds ratio and 95% confidence interval (CI) from
guardian (mother vs. stepmother, fa-
multivariate logistic regression predicting the presence of BCG scar, in-
ther, etc.), legal guardian’s education,
cluding the Rosa Index in the model, Dominican Republic, 2001–2002
legal guardian’s employment status,
child’s history of BCG vaccination,
Predictor variable
Odds ratio
95% CI
P
and child’s birthplace (home, rural
Nutritional status
clinic, private clinic, etc.). Future cen-
Adequatea
1.00
sus work in the Dominican Republic
Malnutrition
0.96
0.92, 1.00
0.052
will need to evaluate multiple ecologi-
Age (yr)
cal factors in order to provide a broad
6a
1.00
7
0.86
0.84, 0.88
0.02
perspective.
8
0.79
0.77, 0.82
0.03
This was a population-based study,
9
0.75
0.72, 0.78
0.03
and the rate of attendance in the
Rosa Index scoreb
1.11
1.10, 1.12
0.02
Dominican Republic for children in el-
a Reference group.
ementary school is 99% (19). Conse-
b The highest Rosa Index score, 11, was for the Santo Domingo school region, which was the ref-
quently, the findings represent Do-
erence group.
minican children between 6 and 9
years of age among different school re-
gions of the country. The findings also
overcome the limitations reported in
Haiti (17), may also have contributed
children living in some of the school
previous, hospital-based surveys of
to this pattern.
regions approximately one or two
BCG coverage implemented in the
BCG coverage in Santo Domingo,
hours’ distance from Santo Domingo
Dominican Republic (17, 18). Never-
the capital, was expected to be higher
were actually more likely to have a
theless, we recognize that including
than in other areas of the country since
BCG scar than were children living in
additional variables related to EPI lo-
a variety of factors affecting the proba-
the capital. These unexpected results
gistical procedures, children of differ-
bility of being vaccinated, including
may be related to different logistic pro-
ent ages (e.g., newborn, infants), and
better socioeconomic conditions and
cedures of the Dominican EPI and to
household sociodemographic condi-
improvements in public health poli-
other, unknown factors inherent to
tions could increase the predictive
cies, are present in this area. However,
those school regions that were not ex-
value of the statistical models for the
370
Rev Panam Salud Publica/Pan Am J Public Health 21(6), 2007

---

Pérez-Then et al. • BCG scar in the Dominican Republic
Original research
presence of BCG scar, and might pro-
and vaccine availability. Otherwise,
cover the very poor regions of the
vide a more thorough explanation for
community “ownership” of the EPI ef-
country.
the vaccine coverage levels.
forts is unlikely, and it is doubtful that
In summary, it appears that an ap-
the EPI objectives will be achieved,
Acknowledgments. This research
propriate goal for the Dominican
even though health care services are,
was supported by a United States
health authorities would be to ensure
in principle, available. Qualitative sur-
National Institutes of Health Fogarty
that all children receive the protection
veys could be included in the Census
International Center research and
afforded by the BCG vaccine. The Cen-
on Height and Weight, in order to
training grant on HIV/AIDS and
sus on Height and Weight could be an
learn what caregivers perceive as their
TB (D43-TW000017-17). The authors
important tool to monitor coverage
priorities for increasing vaccination
thanks Dr. John Lewis for his statisti-
not only with BCG vaccine but also
coverage, how vaccination programs
cal advice, and Dr. Jeannette Báez and
with other vaccines in the Dominican
fit into these priorities, and how the
Mélida Pérez for data management
EPI. The effectiveness of the Domini-
perceived priorities could be achieved
and programming. The authors also
can EPI interventions needs to be mon-
within the context of the existing con-
acknowledge the invaluable contribu-
itored and to be measured in terms of
straints such as poorer access to the
tion of Rosa Urania Abreu in the cre-
community perceptions on such prior-
health care services in rural areas and
ation of the Rosa Index, and Belkis
ities as access to health care facilities
the inability of the Dominican EPI to
Cuello for all her support.
REFERENCES
1. Lugosi L. Theoretical and methodological as-
in Yavatmal District, India. Public Health.
Secretaría de Estado de Educación, Banco
pects of BCG vaccine from the discovery of
2005;119(3):209–16.
Mundial; 2002.
Calmette and Guérin to molecular biology. A
9. World Health Organization. Expanded Pro-
14. Tabachnick BG, Fidell LS. Using multivariate
review. Tuber Lung Dis. 1992;73:252–61.
gramme on Immunization. Programme re-
statistics. 4th edition. Boston: Allyn and
2. Fine PEM. BCG vaccination against tuber-
view. Wkly Epidemiol Rec. 1994;69(12):87–90.
Bacon; 2001.
culosis and leprosy. Br Med Bull. 1988;44(3):
10. Santiago EM, Lawson E, Gillenwater K,
15. Hennekens CH, Buring JE. Epidemiology in
704–16.
Kalangi S Lescano AG, Du Quella G, et al. A
medicine. Philadelphia: Lippincott Williams
3. Smith PG. Case-control studies of the efficacy
prospective study of bacillus Calmette-
& Wilkins; 1987.
of BCG against tuberculosis. In: International
Guerin scar formation and tuberculin skin
16. Moore SE, Goldblatt, D, Bates CJ, Prentice
Union Against Tuberculosis, ed. Proceedings
test reactivity in infants in Lima, Peru. Pedi-
AM. Impact of nutritional status on antibody
of the XXVIth IUAT World Conference on
atrics. 2003;112:e298–e302.
responses to different vaccines in undernour-
Tuberculosis and Respiratory Diseases. Sin-
11. República Dominicana, Secretaría de Estado
ished Gambian children. Acta Paediatr. 2003;
gapore: Professional Postgraduate Services
de Salud Pública y Asistencia Social, División
92(2):170–6.
International; 1987. Pp. 73–9.
Nacional de Epidemiología, Programa Na-
17. Espinal MA, Cruz-Bello J, Pérez EN, Mar-
4. Camargos PAM, Guimaraes MDC, Antunes
cional de Control de la Tuberculosis. Esta-
ranzini B. A hospital-based survey of BCG
CMF. Risk assessment for acquiring meningi-
dísticas sobre tuberculosis. Santo Domingo:
coverage in Santo Domingo, the Dominican
tis tuberculosis among children not vacci-
Secretaría de Estado de Salud Pública y Asis-
Republic. Bull Pan Am Health Organ. 1994;
nated with BCG: a case-control study. Int J
tencia Social; 2005.
28(1):1–8.
Epidemiol. 1988;47(1):193–7.
12. Centro de Estudios Sociales y Demográficos,
18. Pérez-Then E, Shor-Posner G, Campa A,
5. Myint TT, Win H, Aya HH, Kyaw-Mint TO.
Secretaría de Estado de Salud Pública y Asis-
Cedano J, Garrido C, Liz V. Cobertura de
Case-control study on evaluation of BCG
tencia Social, Comisión Ejecutiva para la
BCG en dos hospitales pediátricos de referen-
vaccination of newborns in Rangoon, Burma.
Reforma del Sector Salud, Consejo Presiden-
cia nacional en Santo Domingo, República
Ann Trop Paediatr. 1987;7(2):159–66.
cial del SIDA, Agencia de los Estados Uni-
Dominicana. Bol CENISMI. 2001;11(2):15–6.
6. Bannon MJ. BCG and tuberculosis. Arch Dis
dos para el Desarrollo Internacional, Banco
19. República Dominicana, Oficina Nacional de
Child. 1999;80:80–3.
Mundial/CERSS, Banco Interamericano de
Estadística. Resultados definitivos. VIII Censo
7. Wang L, Turner MO, Elwood RK, Schulzer
Desarrollo/CERSS, Programa MEASURE
de Población y Vivienda 2002. Vol. IV: Carac-
M, Fitzgerald JM. A meta-analysis of the ef-
DHS+/ORC Macro. Encuesta Demográfica y
terísticas educativas. Santo Domingo: ONE;
fect of Bacille Calmette Guérin vaccination on
de Salud 2002. Santo Domingo: MEASURE
2004.
tuberculin skin test measurements. Thorax.
DHS+/ORC Macro: 2003.
2002;57:804–9.
13. Soriano G, Chahín D, Báez J. II Censo Nacio-
8. Zodpey SP, Ambadekar NN, Thakur A. Ef-
nal de Talla y Peso en Escolares de Primer
fectiveness of Bacillus Calmette Guérin
Grado de Básica, República Dominicana,
(BCG) vaccination in the prevention of lep-
2001–2002. Santo Domingo: Centro Nacional
Manuscript received 17 November 2005. Revised manu-
rosy: a population-based case-control study
de Investigaciones en Salud Materno Infantil,
script accepted for publication 5 June 2007.
Rev Panam Salud Publica/Pan Am J Public Health 21(6), 2007
371

---

Original research
Pérez-Then et al. • BCG scar in the Dominican Republic
RESUMEN
Objetivos.
Evaluar críticamente la prevalencia de cicatrices por la vacunación con el
bacilo de Calmette-Guérin (BCG) en niños de 6 a 9 años de la República Dominicana
y examinar la relación entre los factores nutricionales y socioeconómicos y la proba-
Relación entre los factores
bilidad de tener cicatriz de BCG.
nutricionales y
Métodos.
Para este estudio correlacional se empleó la base de datos del II Censo Na-
cional de Talla y Peso en Escolares de Primer Grado de Básica, realizado en la Repú-
sociodemográficos y la
blica Dominicana entre agosto de 2001 y mayo de 2002, para evaluar críticamente el
probabilidad de que los niños
nivel de cobertura nacional de la vacunación con BCG. Entre la información censal de
de la República Dominicana
los niños estaban si tenían cicatriz de BCG, su estado nutricional y sus datos demo-
tengan cicatriz de BCG
gráficos básicos. Se desarrolló un nuevo indicador sociodemográfico, el “índice de
Rosa”, para analizar la posible influencia de la pobreza y de otras características am-
bientales en la presencia de esa cicatriz. Se emplearon modelos de regresión logística
para predecir la presencia de la cicatriz de BCG.
Resultados.
La prevalencia general de cicatrices de BCG fue de 55,3% (85 644/
154 887). Los niños desnutridos presentaron una menor probabilidad de tener cicatriz
de BCG que los niños con un adecuado estado nutricional (razón de posibilidades =
0,91; intervalo de confianza de 95%: 0,87 a 0,95; P < 0,05). Los niños de 7–9 años tu-
vieron menor probabilidad de tener cicatriz de BCG que los niños de 6 años. Los niños
de zonas del país que se encuentran a más de dos horas de viaje de Santo Domingo,
la capital, presentaron menor prevalencia de cicatrices de BCG con mayor frecuencia
que los niños de Santo Domingo. Se encontró correlación entre tener un mayor índice
de Rosa (mejor nivel en las características socioeconómicas) y una mayor prevalencia
de cicatrices de BCG (r = 0,54; P < 0,05).
Conclusiones.
Los resultados del presente estudio indican que la cobertura de va-
cunación de escolares con la vacuna BCG parece no ser la adecuada en la República
Dominicana. Sin embargo, la presencia de la cicatriz de BCG en una mayor propor-
ción de niños más pequeños puede indicar que esa cobertura ha mejorado en años re-
cientes. Se debe hacer un mayor énfasis programático y económico para extender la
vacunación temprana con BCG a las áreas del país donde la cobertura de vacunación
es menor y para analizar el papel que puede estar desempeñando la pobreza en la efi-
cacia de la vacunación.
Palabras clave
Tuberculosis, vacuna BCG, cicatriz, vacunación, vigilancia de la población, facto-
res socioeconómicos, República Dominicana.
TUBERCULOSIS. Detección de casos, tratamiento y vigilancia
La segunda edición de este libro de referencia práctico y con información fidedigna pro-
porciona una base racional para el diagnóstico y tratamiento de la tuberculosis. Escrito por
varios expertos en el campo, sigue siendo fiel al original de Kurt Toman con un formato de
preguntas y respuestas y los capítulos agrupados en tres secciones: detección de casos,
tratamiento y vigilancia.
2006, 396 pp.
Adquiera esta publicación por medio de la librería en línea de la OPS:
ISBN: 92 75 31617 1
http://publications.paho.org; fax: (301) 206-9789; correo electrónico: paho@pmds.com;
US$ /24.00 en países de América Latina y el
Caribe/US$ 32.00 en el resto del mundo
OPS/OMS oficina de país
Código: PC 617
372
Rev Panam Salud Publica/Pan Am J Public Health 21(6), 2007

---