SEARCH

SEARCH BY CITATION

Keywords:

  • leprosy;
  • ecological studies;
  • socio-economic factors;
  • epidemiology;
  • spatial analysis
  • lèpre;
  • études écologiques;
  • facteurs socioéconomiques;
  • épidémiologie;
  • analyse spatiale
  • lepra;
  • estudios ecológicos;
  • factores socioeconómicos;
  • epidemiología;
  • análisis espacial

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Objective

To analyse the ecological association between the demographic and socio-economic characteristics of the Brazilian municipalities and average leprosy incidence rate in the period 2009–2011.

Methods

An ecological study taking the Brazilian municipalities as its units of analysis. The local empirical Bayes estimation method was used to obtain smoothed incidence rates (SIR) for leprosy. The mean, median, first quartile (Q1) and third quartile (Q3) of the SIR were calculated per 100 000 inhabitants. Hierarchical log-linear negative binomial regression models were used to estimate the incidence rate ratios (IRR).

Results

In the period 2009–2011, the average SIR of leprosy in Brazil was 20.2 per 100 000 inhabitants, and the median incidence rate among municipalities was 9.1 per 100 000 inhabitants. Significantly higher adjusted IRR were identified for large municipalities (IRR = 1.67) compared to small municipalities, as well as in municipalities with higher illiteracy rates (IRR = 2.15), more urbanised municipalities (IRR = 1.53), those with greater social inequality as per the Gini index (IRR = 1.26), high percentage of households with inadequate sanitation (IRR = 1.63), higher average number of people per room (IRR = 1.41), high proportions of Family Health Programme coverage (IRR = 1.29), high percentage of household contacts investigated (IRR = 2.30) and those with percentage of cases with grade 2 disability considered to be the medium (IRR = 1.26).

Conclusions

In this study, SIR was significantly associated with municipalities with low socio-economic status. Disease control activities need to be focused on these municipalities, and investments need to be made in improving the population's living conditions.

Objectif

Analyser l'association écologique entre les caractéristiques démographiques et socioéconomiques des municipalités brésiliennes et le taux moyen d'incidence de la lèpre pour la période 2009–2011.

Méthodes

Une étude écologique prenant les municipalités brésiliennes comme unités d'analyse. La méthode d'estimation bayésienne empirique locale a été utilisée pour obtenir les taux d'incidence lissée (TIL) de la lèpre. La moyenne, la médiane, le premier quartile (Q1) et le troisième quartile (Q3) des TIL ont été calculés par 100 000 habitants. Des modèles de régression binomiale hiérarchique log-linéaire négative ont été utilisés pour estimer les rapports de taux d'incidence (IRR).

Résultats

Dans la période 2009–2011, le taux moyen d'incidence lissée de la lèpre au Brésil était de 20.2 pour 100 000 habitants et le taux médian d'incidence des municipalités était de 9.1 pour 100 000 habitants. D'importants rapports ajustés de taux d'incidence (IRR) ont été identifiés pour les grandes municipalités (IRR = 1.67) par rapport aux petites, ainsi que pour les municipalités avec des taux d'analphabétisme plus élevés (IRR = 2.15), les municipalités plus urbanisées (IRR = 1.53), celles avec plus d'inégalités sociales selon l'indice de Gini (IRR = 1.26), celles avec un pourcentage élevé de ménages ayant un assainissement inadéquat (IRR = 1.63), celles avec un plus grand nombre moyen de personnes par pièce (IRR = 1.41), celles avec une forte proportion de couverture du Programme de Santé Familiale (IRR = 1.29), celles avec un pourcentage élevé de contacts familiaux investigués (IRR = 2.30) et celles avec un pourcentage considéré comme moyen de cas avec un handicap de grade 2 (IRR = 1.26).

Conclusions

Dans cette étude, le TIL a été significativement associé avec les municipalités avec un faible statut socioéconomique. Les activités de lutte contre la maladie devraient être axées sur ces municipalités et des investissements devraient être réalisés dans l'amélioration des conditions de vie de la population.

Objetivo

Analizar la asociación ecológica entre las características demográficas y socioeconómicas de los municipios Brasileros y la tasa promedio de incidencia de lepra durante el periodo 2009–2011.

Métodos

Estudio ecológico que utiliza los municipios Brasileros como unidad de análisis. Se utilizó el método Bayesiano empírico para obtener la razón de incidencias estandarizada (SIR) para lepra. Se calcularon la media, la mediana, el primer cuartil (Q1) y el tercer cuartil (Q3) de las SIR por 100 000 habitantes. Se utilizaron modelos jerárquicos de regresión binomial negativa para calcular la razón de tasas de incidencia (RDI).

Resultados

Durante el periodo 2009–2011, la SIR promedio de lepra en Brasil era del 20.2 por 100 000 habitantes, y la media de la tasa de incidencia entre municipios era de 9.1 por 100 000 habitantes. Se identificaron las RDI ajustadas significativas en los municipios grandes (RDI = 1.67), en comparación con los municipios pequeños, al igual que los municipios con mayores tasas de analfabetismo (RDI = 2.15), los municipios más urbanizados (RDI = 1.53), aquellos con mayor inequidad social según el índice de Gini (RDI = 1.26), mayor porcentaje de hogares con un saneamiento inadecuado (RDI = 1.63), mayor número promedio de personas por habitación (RDI = 1.41), mayor proporción de cobertura en el Programa Familiar de Salud (RDI = 1.29), mayor porcentaje de contactos en el hogar investigados (RDI = 2.30) y aquellos con un porcentaje de casos con discapacidad de grado 2 considerados como el medio (RDI = 1.26).

Conclusiones

En este estudio, SIR estaba significativamente asociado con aquellos municipios con un estatus socioeconómico bajo. Las actividades para el control de la enfermedad deberían centrarse en estos municipios y las inversiones deberían hacerse mejorando las condiciones de vida de la población.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

In 2011, about 220 000 new cases of leprosy were identified worldwide. India (58%), Brazil (16%) and Indonesia (9%) account for 83% of these cases. Brazil is the second country in the world with the largest number of new notified cases and accounts for 92% of total new cases in the Americas (World Health Organization 2012). Brazil is the largest country in South America with a population of 191 million inhabitants (IBGE 2011). The country is divided into five regions (North, Northeast, Center-West, Southeast and South), 26 states and the Federal District.

In Brazil in 2012, the prevalence rate of leprosy was 1.5 cases per 10 000 inhabitants, the detection rate was 17.2 per 100 000 inhabitants and the detection rate in those aged under 15 was 4.8 per 100 000 inhabitants (Ministério da Saúde 2013a). The Center-West, the North and the Northeast regions were considered to be the most outstanding areas in terms of the continuing transmission of the disease in the country. An ecological study conducted in the Brazilian state of Ceará, located in the Northeast region, found that leprosy is associated with higher levels of poverty and uncontrolled urbanisation (Kerr-Pontes et al. 2004). Control of leprosy must take into consideration its multiple associated factors. In addition to adequate care in treating cases, socio-economic conditions also need to be considered, including effective efforts to improve housing, sanitation and education (Souza et al. 2009).

A better understanding of the epidemiology of leprosy will help to explain why Brazil, along with other countries in the world, has not yet been able to eliminate this disease as a public health problem. The study aimed to analyse the ecological association between the demographic and socio-economic characteristics of the Brazilian municipalities and average leprosy incidence rate in the period 2009–2011.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

An ecological study was conducted with all 5565 Brazilian municipalities as its units of analysis. Cases of leprosy are recorded by the Brazilian Notifiable Diseases Surveillance System (Sistema de Informação de Agravos de Notificação – SINAN) based on data abstracted from forms used in the investigation of cases. SINAN is the main survey system for collection and analysis of national data on leprosy in Brazil (Ministério da Saúde 2007).

The dependent variable is the average smoothed incidence rate (SIR) of leprosy. Incidence rates of leprosy were calculated per 100 000 inhabitants, whereby the numerator was the total number of new cases recorded on the SINAN system in the period 2009–2011, and the denominator was the size of the resident population as per the 2010 demographic census (IBGE 2011), multiplied by three. Although it was a strategy for reducing the overall fluctuation of the study measure, the 3-year average estimator did not adequately reduce the variance in the smaller municipalities (<5000 inhabitants), which represent about 25% of all Brazilian municipalities. Therefore, the local empirical Bayes estimator (Marshall 1991) was then used to obtain the smoothed rates. This estimator is a good alternative for mitigating crude rate fluctuation associated with small areas using neighbourhood information from each municipality (Assunção et al. 2005). The neighbourhood list was based on geographic unities (municipalities) with contiguous boundaries.

The independent variables, obtained from the 2010 demographic census, were assessed and allocated into two levels: Level 1 (demographic and socio-economic) variables included region (North, Northeast, Southeast, Center-West, South), illiteracy rate (percentage of people aged 15 or over unable to read or write at least a simple note), poverty rate (percentage of people with per capita household income of up to ½ the minimum wage), percentage of the population living in extreme poverty (percentage of people below the extreme poverty line defined for Brazil, i.e. nominal monthly per capita household income of up to R$ 70), average monthly household income per capita (total monthly income of all family members living in the household divided by the number of these family members), urbanisation rate (percentage of the population living in urban areas), sex ratio (ratio of males to females), unemployment rate (percentage of the population ≥16 years without employment), the Gini index of per capita household income (DATASUS 2013), 20–20 income ratio (ratio between the income of the wealthiest 20% and the poorest 20%), average number of dwellers in permanent private households, average number of dwellers per room (delimited space in the household) and percentage of households with inadequate sanitation (percentage of households not connected to the water supply main, and not connected to the sewer system, and without access to garbage collection). Level 2 (income redistribution programme and health services) variables included percentage coverage of the Programa Bolsa Família (Family Allowance Programme, PBF) over the eligible population in 2010 (MDS 2013), percentage average coverage of the Programa Saúde da Família (Family Health Programme, PSF) in 2010 (Ministério da Saúde 2013b) and Sinan operational indicators – percentage of household contacts investigated between 2009 and 2011 and percentage of cases with grade 2 disability (Alberts et al. 2011) among new cases detected and assessed between 2009 and 2011.

All continuous variables were categorised into approximate quartiles, except the variable percentage of cases with grade 2 disability, which was categorised according to the following parameters of interpretation: low (<5%), medium (5% and <10%) and high (≥10%) (Ministério da Saúde 2009). Municipalities categorisation by size followed the classification used by the Brazilian Institute of Geography and Statistics (Instituto Brasileiro de Geografia e Estatística – IBGE): small 1 (municipalities with up to 20 000 inhabitants), small 2 (municipalities with between 20 001 and 50 000 inhabitants), medium (municipalities with between 50 001 and 100 000 inhabitants), large (municipalities with between 100 001 and 900 000 inhabitants) and metropolises (municipalities with more than 900 000 inhabitants).

In the descriptive stage, the average smoothed leprosy incidence rate distributions, according to municipal characteristics, were described based on their mean, median, quartiles (first = Q1, third = Q3) and minimum and maximum values. The existence of multicollinearity was then verified using the independent variable correlation matrix and the variance inflation factor (Montgomery 2013).

The distribution of incidence rates in municipalities was non-symmetric. For this reason in the analytical stage, the option was taken to use the log-linear model with negative binomial response. Multivariate analysis was conducted using a hierarchical approach (Victora et al. 1997) with the model incorporating two levels: the distal, including demographic and socio-economic variables; and the proximal including income redistribution programme and health service variables. Leprosy is a disease with a complex chain of causation (Bhat & Prakash 2012). The construction of the hierarchical model may be a better way to capture the inter-relationships between its determinants. The backward method was used on each of the hierarchical levels to select the independent variables. Distal variables were added first to the multivariable model, and retained as long as they remained significantly (P ≤ 0.05). Akaike's criterion was used when selecting the models. To assess model adjustment quality, normal probability plots were analysed relating to the negative binomial log-linear model (Paula 2004).

Analysis was conducted with the aid of the following programs: STATA 12 (StataCorp 2011), R 3.0.2 (R Core Team 2013) and ArcGis 9.2 (Environmental Systems Research Institute, Redlands, CA, USA) (ESRI 2010). This study was approved by the Ethics Committee of the Health Sciences Faculty of the University of Brasília.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

In Brazil between 2009 and 2011, the mean crude rate and the mean SIR of leprosy in the municipalities were 18.6 and 20.2 new cases per 100 000 inhabitants, respectively. Median SIR was 9.1 (Q1–Q3: 3.3–22.8) per 100 000 inhabitants (Table 1). Great SIR variability was found in the small municipalities (up to 50 000 inhabitants), whilst it gradually gained stability in the larger municipalities (Figure 1). The municipality of São Luís (MA) is noteworthy in that it had a high SIR – 69.2 per 100 000 inhabitants – compared to the remaining Brazilian metropolises. The municipality of Marituba (PA) had the highest SIR – 253.5 per 100 inhabitants – among the large municipalities.

Table 1. Average smoothed leprosy incidence rates (per 100 000 inhabitants) according to municipal characteristics. Brazil, 2009–2011
VariablesAverage smoothed incidence rate of leprosy (per 100 000 inhabitants)
TotalMeanMedianQ1Q3MinimumMaximum
Brazil556520.29.13.322.80.0377.0
Region
North44953.142.621.276.00.1253.5
Northeast179520.512.45.625.00.0233.6
Southeast166810.05.53.010.90.0227.2
South11886.52.90.79.30.0377.0
Center-West46559.246.124.378.30.3311.7
Size of the municipalities
Small municipalities 1391419.18.43.021.20.0377.0
Small municipalities 2104323.111.74.327.70.0251.9
Medium-sized municipalities32524.610.04.131.00.0204.9
Large municipalities26620.18.13.423.80.0253.5
Metropolises1720.411.44.426.02.569.2
Illiteracy rate (%)
<814728.43.81.58.30.0253.5
≥8 to <13131921.48.83.521.20.0267.1
≥13 to <24141331.617.67.044.30.0377.0
≥24136120.012.15.224.30.0209.1
Poverty rate (%)
<2515548.64.41.79.80.0156.2
≥25 to <43122621.37.73.021.50.0311.7
≥43 to <66134828.414.44.837.00.0377.0
≥66143722.113.66.427.80.0191.6
Percentage of the population living in extreme poverty
<315788.84.62.09.70.0171.9
≥3 to <8122522.38.22.923.20.0311.7
≥8 to <20134628.714.34.735.20.0377.0
≥20141623.114.66.530.20.0191.6
Average monthly household income per capita
<R$ 260138720.312.76.025.30.0171.1
≥R$ 260 to <R$ 421139427.514.14.934.70.0377.0
≥R$ 421 to <R$ 571138821.78.03.121.60.0311.7
≥R$ 571139611.34.51.711.10.0267.1
Urbanization rate (%)
<47138817.18.42.621.00.0246.5
≥47 to <65142322.310.63.726.60.0377.0
≥65275420.78.73.522.20.0311.7
Unemployment rate (%)
<4151713.14.91.412.70.0266.9
≥4 to <6138621.79.83.924.00.0377.0
≥6 to <8121223.610.64.228.30.0261.4
≥8145023.512.95.428.60.0285.3
Gini index of per capita household income
<0.50253014.05.82.314.10.0311.7
≥0.50 to <0.55162021.110.23.925.10.0233.6
≥0.55141530.417.67.140.00.0377.0
(20–20) income ratio
<12151210.95.22.111.50.0261.4
≥12 to <17124318.37.02.817.60.0311.7
≥17 to <27143723.311.03.927.20.0285.3
≥27137329.017.07.739.50.0377.0
Sex ratio
<1.0214415.47.73.017.60.0253.5
≥1.0 to <1.1293920.59.13.323.00.0377.0
≥1.148240.123.97.564.70.0266.9
Average number of dwellers in permanent private households
<3.1121115.26.11.715.70.0311.7
≥3.1 to <3.3294921.69.83.824.20.0377.0
≥3.3140521.610.93.925.80.0261.4
Average number of dwellers per room
<0.51143211.74.91.412.40.0267.1
≥0.51 to <0.57136916.67.12.916.90.0311.7
≥0.57 to <0.65135923.710.64.425.10.0377.0
≥0.65140529.117.67.539.70.0253.5
Percentage of households with inadequate sanitation
<613499.74.32.010.10.0143.9
≥6 to <16145119.98.63.220.90.0285.3
≥16276525.513.55.431.30.0377.0
Percentage of contacts investigated
<4627689.74.51.510.60.0261.4
≥46 to <83142633.820.08.746.70.0285.3
≥83137127.314.67.332.30.0377.0
Percentage of coverage of the Family Allowance Programme (PBF)
<1613868.94.11.69.50.0189.9
≥16 to <30143521.58.23.222.80.0311.7
≥30274425.214.25.831.40.0377.0
Percentage of coverage of the Family Health Programme (PSF)
<5090114.65.12.414.50.0267.1
≥50 to <8085223.211.04.324.20.0266.9
≥80381220.99.93.624.20.0377.0
Percentage of cases with grade 2 disability
<5428816.37.02.617.50.0285.3
≥5 to <1031652.638.518.279.81.1251.9
≥1096127.016.17.932.10.5377.0
image

Figure 1. Boxplot of average smoothed leprosy incidence rates (per 100 000 inhabitants), by municipality size. Brazil, 2009–2011.

Download figure to PowerPoint

The highest average SIRs were found in municipalities in the Center-Western and Northern regions (Figure 2), where 442 (48.4%) of 914 municipalities had rates >40 new cases per 100 000 inhabitants and were considered to be hyperendemic. The municipalities of the South and Southeast regions had the lowest rates.

image

Figure 2. Map of average smoothed leprosy rate (per 100 000 inhabitants), by municipality. Brazil, 2009–2011.

Download figure to PowerPoint

In the period 2009–2011, median SIR was highest in the Center-West region (46.1 per 100 000 inhabitants), followed by the North (42.6 per 100 000 inhabitants) and Northeast (12.4 per 100 000 inhabitants) (Table 1). The lowest median SIR was found in the South (2.9 per 100 000 inhabitants). Higher median SIR was also found in municipalities which had larger populations (>20 000 inhabitants), were more urbanised (≥65% of the population living in urban areas), had a median illiteracy rate (≥13% to <24%), higher poverty rates (≥43%), higher unemployment rates (≥8%), higher average number of dwellers per household (≥3.3), and significant income inequality (Gini index of per capita household income ≥0.55). Also noteworthy is that the municipalities with the highest sex ratios – more males than females – had the highest SIR (23.9 per 100 000 inhabitants).

After the descriptive analysis had been performed, the following variables were excluded owing to high collinearity with other variables: percentage PBF coverage, average number of dwellers in permanent private households, poverty rate and average monthly household income per capita.

The adjusted analysis provided incidence rate ratios (IRR) of 4.62 (95% CI: 4.12–5.17) and 3.14 (95% CI: 2.77–3.57) for the Center-West and North regions, respectively, in comparison with the Southeast region (Table 2). With regard to medium-sized municipalities and metropolises compared to small-sized municipalities, IRR was 1.67 (95% CI: 1.44–1.95) and 1.92 (95% CI: 1.15–3.18), respectively.

Table 2. Ecological associations between average leprosy rates (per 100 000 inhabitants) and selected variables, in Brazilian municipalities. Brazil, 2009–2011
LevelVariablesHierarchical log-linear negative binomial regression model
Crude IRR (95% CI)P-valueAdjusted IRR (95% CI)P-value
  1. CI, confidence interval.

Level 1Region
Southeast1.00 1.00 
South0.65 (0.60, 0.70)<0.0010.77 (0.71, 0.84)<0.001
Northeast2.05 (1.92, 2.20)<0.0011.35 (1.20, 1.52)<0.001
North5.32 (4.78, 5.91)<0.0013.14 (2.77, 3.57)<0.001
Center-West5.93 (5.34, 6.58)<0.0014.62 (4.12, 5.17)<0.001
Size of the municipalities
Small municipalities 11.00 1.00 
Small municipalities 21.21 (1.13, 1.30)<0.0011.19 (1.11, 1.29)<0.001
Medium-sized municipalities1.29 (1.15, 1.45)<0.0011.41 (1.24, 1.60)<0.001
Large municipalities1.05 (0.93, 1.19)0.4461.67 (1.44, 1.95)<0.001
Metropolises1.07 (0.66, 1.74)0.7841.92 (1.15, 3.18)0.012
Illiteracy rate (%)
<81.00 1.00 
≥8 to <132.54 (2.35, 2.74)<0.0011.51 (1.37, 1.66)<0.001
≥13 to <243.76 (3.49, 4.06)<0.0012.41 (2.12, 2.74)<0.001
≥242.38 (2.20, 2.57)<0.0012.15 (1.83, 2.53)<0.001
Urbanization rate (%)
<471.00 1.00 
≥47 to <651.30 (1.21, 1.41)<0.0011.27 (1.17, 1.37)<0.001
≥651.21 (1.13, 1.29)<0.0011.53 (1.40, 1.67)<0.001
Gini index of per capita household income
<0.501.00 1.00 
≥0.50 to <0.551.51 (1.42, 1.61)<0.0011.10 (1.02, 1.18)0.010
≥0.552.17 (2.03, 2.32)<0.0011.26 (1.16, 1.37)<0.001
Percentage of households with inadequate sanitation
<61.00 1.00 
≥6 to <162.06 (1.91, 2.23)<0.0011.42 (1.30, 1.56)<0.001
≥162.64 (2.46, 2.82)<0.0011.63 (1.47, 1.81)<0.001
Average number of dwellers per room
<0.511.00 1.00 
≥0.51 to <0.571.42 (1.31, 1.53)<0.0011.14 (1.05, 1.24)0.002
≥0.57 to <0.652.03 (1.88, 2.19)<0.0011.25 (1.14, 1.37)<0.001
≥0.652.49 (2.30, 2.68)<0.0011.41 (1.26, 1.58)<0.001
Level 2Percentage of contacts investigated
<461.00 1.00 
≥46 to <833.49 (3.26, 3.72)<0.0012.09 (1.95, 2.25)<0.001
≥832.81 (2.63, 3.00)<0.0012.30 (2.14, 2.48)<0.001
Percentage of coverage of the Family Health Programme (PSF)
<501.00 1.00 
≥50 to <801.58 (1.44, 1.74)<0.0011.19 (1.07, 1.32)0.001
≥801.43 (1.32, 1.54)<0.0011.29 (1.17, 1.41)<0.001
Percentage of cases with grade 2 disability
<51.00 1.00 
≥5 to <103.22 (2.87, 3.62)<0.0012.00 (1.77, 2.26)<0.001
≥101.65 (1.54, 1.77)<0.0011.26 (1.17, 1.36)<0.001

Incidence rate ratio was 2.15 (95% CI: 1.83–2.53) in municipalities with higher illiteracy rates (24% or over) compared to those with lower rates (under 8%) (Table 2). More urbanised municipalities (65% or over) had an IRR of 1.53 (95% CI: 1.40–1.67) compared to those which were less urbanised. Municipalities with higher social inequality indices (Gini index of 0.55 or more) had an IRR of 1.26 (95% CI: 1.16–1.37) compared to those with less inequality. IRR was 1.63 (95% CI: 1.47–1.81) for municipalities where the percentage of households with inadequate sanitation was ≥16%, compared to those where this percentage was <6%. In municipalities where the average number of dwellers per room was ≥0.65, IRR was 1.41 (95% CI: 1.26–1.58) compared to municipalities with <0.51 dwellers per room on average.

As the proportion of PSF coverage and the percentage of household contacts investigated increase in the Brazilian municipalities, a persistent gradient of increasing SIR can be seen after adjusting for the other municipal socio-economic variables (Table 2). IRR was 1.26 (95% CI: 1.17–1.36) and 2.00 (95% CI: 1.77–2.26) in municipalities with a high (≥10%) and medium (≥5% and <10%) percentage of cases with grade 2 physical disability compared to municipalities with a lower percentage (<5%).

The variables relating to the percentage of the population in extreme poverty, unemployment rate, 20–20 income ratio and sex ratio lost their statistical significance following the adjusted analysis (Table 2). The normal probability graph with simulated envelopes for an adjustment of negative binomial distribution showed that the model fitted well the data because the points (residuals) were within the confidence bands.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

In Brazil between 2009 and 2011, average municipal leprosy incidence rates were high (crude mean rate of 18.6 and an SIR of 20.2 new cases per 100 000 inhabitants). The low median SIR (9.1 per 100 000 inhabitants) in relation to the estimated average indicates a heavily right-skewed distribution. Higher SIR are found in municipalities that have larger populations (over 50 000 inhabitants), are more urbanised and are located in the Center-West, North and Northeast regions, with poorer socio-economic indicators (high rates of illiteracy, high average number of dwellers per room, high percentage of households with inadequate sanitation) and with greater social inequality according to the Gini index. With regard to health services, a gradient of increasing SIR was found in the Brazilian municipalities as the proportion of PSF coverage and the percentage of household contacts investigated. In addition, there are densely populated areas of Brazil (such as South and Southeast) which did not present a high incidence rate. This result could suggest that the socio-economic status has more influence on increasing the incidence of leprosy than population density.

Since the 1980s, studies have indicated the impact of economic development on reducing the incidence of leprosy. Indeed, few new cases of this disease are recorded in developed countries. When leprosy is detected in these contexts, it is found mainly among immigrants from countries where the disease is still endemic (Saikawa 1981; Cossermelli-Messina 1993). On the other hand, in developing countries, the socio-economic and demographic characteristics of some municipalities can contribute to the reproduction and distribution of leprosy. Diverse studies have shown significant association between the incidence of leprosy and socio-economic factors – poor housing conditions, and agglomeration of people in households, unavailability of water supply and garbage collection, as well as poverty – suggesting that improved socio-economic conditions could contribute to reduce the disease incidence (Ponnighaus et al. 1994; Kerr-Pontes et al. 2006; Queiroz et al. 2010; Sales et al. 2011; Suzuki et al. 2012; Moura et al. 2013). Moreover, more populous municipalities are generally subject to disorganised growth and high demographic density in poverty pockets, usually on the outskirts of large cities. Within these contexts, the probability of contact with the disease is greater and the persistence of its transmission is aggravated even more by living conditions and precarious access to health services (Imbiriba et al. 2009).

A study in Spain reported that socio-economic development has a known impact on leprosy's epidemiological behaviour (Alfonso et al. 2005). In Bangladesh, the deprived socio-economic circumstances and especially nutritional deficits are known risk factors for leprosy in general (Feenstra et al. 2011).

Along with these phenomena of deprived socio-economic, higher SIR were found in this study in municipalities having greater social inequalities, even after controlling for the other socio-economic and demographic variables. A study in an endemic area in Brazil suggests that unplanned and uncontrolled urbanisation increases social inequality by excluding people from social and material opportunities, making them susceptible to several diseases – including leprosy (Kerr-Pontes et al. 2004). Indeed, in addition to absolute poverty, it is possible to suppose that social exclusion and isolation are an impediment to the adequate enjoyment of citizenship and being reached by social policies – including those relating to education and health, thus making it difficult to access resources relevant to ensuring good health.

Although determined socially, the guarantee of equal access to health services – prevention, diagnosis, rehabilitation and treatment – is important for addressing leprosy (Rodrigues & Lockwood 2011). Continuous surveillance of those who have had contact with leprosy cases also appears to be a particularly relevant strategy for reducing its incidence rate (Sarno et al. 2012). In this study, the fact that municipalities with better structured surveillance services – greater PSF coverage and tracing of examined contacts – had the highest leprosy incidence rates is worthy of reflection. One hypothesis that can be raised to explain this finding is that in these municipalities, at least in the short term, a consequence of increased surveillance actions may be the initial increase in the incidence of ‘detected’ cases of the disease. In turn, this increase may lead to increased tracing of people who have had contact with it and greater detection of cases with grade 2 disability which previously were not identified. This hypothesis may explain the finding that municipalities with a greater proportion of cases presenting with grade 2 disability also had higher average leprosy incidence rates. Furthermore, a study into spatial approaches to define highly vulnerable areas in an endemic area in Brazil between 2001 and 2009 identified several clusters where late diagnosis of the disease occurred (Alencar et al. 2012). Another factor which can hinder early diagnosis is that the symptoms of leprosy are similar to those of many kinds of skin diseases and neuropathic problems (Pfaltzgraff & Ramu 1994).

Nevertheless, it is expected that in the long term, these conditions can be overturned and that the impact of active surveillance can be measured through the reduction in the percentage of cases with grade 2 disability. Moreover, it should be emphasised that in Brazil, high PSF coverage is mainly found in municipalities with precarious living conditions and this may also influence the findings.

Our study has some limitations. Its main limitation relates to the use of secondary data, which may often result in inconsistencies in the rates estimated. Case underreporting is expected, so that incidence rates only take into consideration ‘detected’ cases of the disease. In addition, owing to the characteristics of the health services in the municipalities, differences are expected in the coverage and quality of the information held on the SINAN system. Municipalities having a higher proportion of their populations residing in rural areas and municipalities in regions where the population is dispersed, as in the case of the Brazilian Amazon region, have reduced capacity for diagnosing leprosy as a result of the difficulties faced by inhabitants in accessing health services that have adequately trained professional staff. A potential bias in detection may have overestimated incidence differentials between these municipalities and large municipalities. On the other hand, some adjustments made to the models may have minimised this possible bias, such as the adjustments for region and for municipal urbanization rates, which may indirectly have reduced the effect of the difficulties in diagnosis associated with these characteristics. Another limitation, relating to the use of crude rates in spatial analysis, is their high instability in expressing the risk of a rare event or when the population in the region of occurrence is small. With the aim of minimising this limitation, this study used incidence rate smoothing by means of the local empirical Bayes estimation method (Marshall 1991). Another point to be considered is the difficulty establishing a temporal relationship as one criterion for causality, whereby the independent variable may in fact be a consequence of the dependent variable. For example, it can be presumed that places with higher incidence rates may be prioritised for PSF implantation. As such, higher PSF coverage could be a consequence of high incidence rates and not a determinant of them. On the other hand, PSF implantation, this being a primary healthcare model emphasising linkage between the health team and families and the community, may promote the detection of leprosy cases, thus accentuating the association found.

In the present study were not included climatic variables in the analysis. Studies have suggested that the population living in endemic areas with poor conditions is at highest risk for acquiring Mycobacterium leprae irrespective of season (Bhat & Prakash 2012; Lavania et al. 2013). Future studies, including field-collected validation data (temperature, rivers, water sources, humidity), may help to identify the environmental factors associated with the incidence rate of leprosy.

In Brazil, studies have suggested that the migration has been considered a possible factor associated with continued leprosy transmission. However, socio-economic factors influence the initial decision to migrate (Murto et al. 2013, 2014). The leprosy incidence and control involves a very complex causal chain. It includes biologic aspects of transmission, development of clinical disease in those infected and disease detection. This manuscript focused on the association between the average leprosy incidence rate and the demographic and socio-economic characteristics of the municipalities. The associations found in this study do not imply causation.

Considering that higher IRR were found in municipalities with poorer socio-economic status, it is recommended that the efforts of control programmes be directed towards these areas which are considered to be highly endemic. There is clearly a need to intensify measures aimed at early diagnosis of the disease, as well as actions to prevent and control it, whilst taking into consideration existing social inequalities. We suggest that health service and social programme actions be integrated. Improving the population's living conditions remains fundamental to interrupting the continuing transmission of leprosy, and consequently, its incidence could be reduced.

Acknowledgement

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

The authors would like to thank the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for granting a PhD scholarship to Lucia R. S. Freitas.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References
  • Alberts CJ, Smith WCS, Meima A, Wang L & Richardus JH (2011) Potential effect of the World Health Organization's 2011–2015 global leprosy strategy on the prevalence of grade 2 disability: a trend analysis. Bulletin of the World Health Organization 89, 487495.
  • Alencar CH, Ramos AN Jr, dos Santos ES, Richter J & Heukelbach J (2012) Clusters of leprosy transmission and of late diagnosis in a highly endemic area in Brazil: focus on different spatial analysis approaches. Tropical Medicine & International Health 17, 518525.
  • Alfonso JL, Vich FA, Vilata JJ & de las Aguas JT (2005) Factors contributing to the decline of leprosy in Spain in the second half of the twentieth century. International Journal of Leprosy and Other Mycobacterial Diseases: Official Organ of the International Leprosy Association 73, 258268.
  • Assunção RM, Schmertmann CP, Potter JE & Cavenaghi SM (2005) Empirical Bayes estimation of demographic schedules for small areas. Demography 42, 537558.
  • Bhat RM & Prakash C (2012) Leprosy: an overview of pathophysiology. Interdisciplinary Perspectives on Infectious Diseases 2012, 181089.
  • Cossermelli-Messina W (1993) Leprosy: statistics in the Americas. The Journal of Rheumatology 20, 893896.
  • DATASUS (2013) http://tabnet.datasus.gov.br/cgi/ibge/censo/cnv/ginibr.def (accessed 26 June 2013).
  • ESRI (2010) ArcGis 9.2. Environmental Systems Research Institute, Inc 1982–2011, Redlands, CA.
  • Feenstra SG, Nahar Q, Pahan D, Oskam L & Richardus JH (2011) Recent food shortage is associated with leprosy disease in Bangladesh: a case-control study. PLoS Neglected Tropical Diseases 5, e1029.
  • IBGE (2011) Sinopse do Censo Demográfico 2010. Instituto Brasileiro de Geografia e Estatística, IBGE, Brasślia.
  • Imbiriba EN, Silva Neto AL, Souza WV, Pedrosa V, Cunha Mda G & Garnelo L (2009) Social inequality, urban growth and leprosy in Manaus: a spatial approach. Revista de Saúde Pública 43, 656665.
  • Kerr-Pontes LR, Montenegro AC, Barreto ML, Werneck GL & Feldmeier H (2004) Inequality and leprosy in Northeast Brazil: an ecological study. International Journal of Epidemiology 33, 262269.
  • Kerr-Pontes LR, Barreto ML, Evangelista CM, Rodrigues LC, Heukelbach J & Feldmeier H (2006) Socioeconomic, environmental, and behavioural risk factors for leprosy in North-east Brazil: results of a case–control study. International Journal of Epidemiology 35, 9941000.
  • Lavania M, Turankar RP, Karri S, Chaitanya VS, Sengupta U & Jadhav RS (2013) Cohort study of the seasonal effect on nasal carriage and the presence of Mycobacterium leprae in an endemic area in the general population. Clinical Microbiology & Infection 19, 970974.
  • Marshall RJ (1991) Mapping disease and mortality rates using empirical Bayes estimators. Journal of the Royal Statistical Society. Series C, Applied statistics 40, 283294.
  • MDS (2013) Ministério do Desenvolvimento Social Bolsa Família. http://www.mds.gov.br/bolsafamilia/ (accessed 26 June 2013).
  • Ministério da Saúde (2007) Secretaria de Vigilância em Saúde. Departamento de Vigilância Epidemiológica. Sistema de Informação de Agravos de Notificação – Sinan: normas e rotinas. Ministério da Saúde, Brasília.
  • Ministério da Saúde (2009) Secretaria de Vigilância em Saúde. Secretaria de Atenção à Saúde. Portaria n°. 125/SVS-SAS, de 26 de março de 2009. Define ações de controle da hanseníase (Ordinance No. 125/SVS-SAS of March 26, 2009. Defines actions to control leprosy). Ministério da Saúde, Brasília.
  • Ministério da Saúde (2013a) Doenças decorrentes e perpetuadoras da pobreza. In: Saúde Brasil 2012: uma análise de situação de saúde e dos 40 anos do Programa Nacional de Imunizações no Brasil. Brasília: Editora do Ministério da Saúde, 169185.
  • Ministério da Saúde (2013b) Manual de orientações sobre o Bolsa Família na Saúde. http://bvsms.saude.gov.br/bvs/publicacoes/ma-nualorientacao_sobre_bolsa_familia.pdf (accessed 26 June 2013).
  • Montgomery DC (2013) Design and Analysis of Experiments, 8th edn. John Wiley & Sons, Inc., Hoboken, NY.
  • Moura ML, Dupnik KM, Sampaio GA et al. (2013) Active surveillance of Hansen's Disease (leprosy): importance for case finding among extra-domiciliary contacts. PLoS Neglected Tropical Diseases 7, e2093.
  • Murto C, Chammartin F, Schwarz K, da Costa LMM, Kaplan C & Heukelbach J (2013) Patterns of migration and risks associated with leprosy among migrants in Maranhão, Brazil. PLoS Neglected Tropical Diseases 7, e2422.
  • Murto C, Ariza L, Alencar CH, Chichava OA, Oliveira AR, Kaplan C, da Silva LFM & Heukelbach J (2014) Migração entre pessoas com hanseníase: estudo de base populacional no Centro-Oeste do Brasil. Cadernos de Saúde Pública 30, 487501.
  • Paula GA (2004) Modelos de regressão com apoio computacional. Universidade de São Paulo, São Paulo.
  • Pfaltzgraff RE & Ramu G (1994) Clinical leprosy. In: Leprosy, 2nd edn (ed. Hastings RC) Churchill Livingstone, New York, pp. 237290.
  • Ponnighaus JM, Fine PE, Sterne JA et al. (1994) Incidence rates of leprosy in Karonga District, northern Malawi: patterns by age, sex, BCG status and classification. International Journal of Leprosy and Other Mycobacterial Diseases: Official Organ of the International Leprosy Association 62, 1023.
  • Queiroz JW, Dias GH, Nobre ML et al. (2010) Geographic information systems and applied spatial statistics are efficient tools to study Hansen's disease (leprosy) and to determine areas of greater risk of disease. The American Journal of Tropical Medicine and Hygiene 82, 306314.
  • R Core Team (2013) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria.
  • Rodrigues LC & Lockwood D (2011) Leprosy now: epidemiology, progress, challenges, and research gaps. The Lancet Infectious Diseases 11, 464470.
  • Saikawa K (1981) The effect of rapid socio-economic development on the frequency of leprosy in a population. Leprosy Review 52, 167175.
  • Sales AM, Ponce de Leon A, Dupre NC et al. (2011) Leprosy among patient contacts: a multilevel study of risk factors. PLoS Neglected Tropical Diseases 5, e1013.
  • Sarno EN, Duppre NC, Sales AM, Hacker MA, Nery JA & de Matos HJ (2012) Leprosy exposure, infection and disease: a 25-year surveillance study of leprosy patient contacts. Memórias do Instituto Oswaldo Cruz 107, 10541059.
  • Souza AD, el-Azhary RA & Foss NT (2009) Management of chronic diseases: an overview of the Brazilian governmental leprosy program. International Journal of Dermatology 48, 109116.
  • StataCorp (2011) Stata Statistical Software College Station, Release 12th edn. Stata Corp LP, Texas.
  • Suzuki K, Akama T, Kawashima A, Yoshihara A, Yotsu RR & Ishii N (2012) Current status of leprosy: epidemiology, basic science and clinical perspectives. The Journal of Dermatology 39, 121129.
  • Victora CG, Huttly SR, Fuchs SC & Olinto MT (1997) The role of conceptual frameworks in epidemiological analysis: a hierarchical approach. International Journal of Epidemiology 26, 224227.
  • World Health Organization (2012) Global leprosy situation. Weekly Epidemiological Record 87, 317328.