– SP, Brazil
Este estudo tem por objetivo analisar a incidência de câncer infantil e a correspondente sobrevivência em crianças (0 – 14 anos) e adolescentes (15–19 anos) residentes em Campinas. Para tanto, foram utilizados diferentes métodos estatísticos para identificar o padrão de incidência da doença, bem como para analisar variáveis sociodemográficas associadas à sobrevivência dos indivíduos acometidos.
76
Produção Bibliográfica referente ao Capítulo 3
Trabalhos completos publicados em anais de congressos
OLIVEIRA FRIESTINO, J.K.; MOREIRA-FILHO, D. C. Câncer em crianças e adolescentes - Análise de sobrevida e distribuição espacial de casos registrados em um município brasileiro. In: GeoSaúde 2014, 1 Congresso de Geografia da Saúde dos Países de Língua Portuguesa, 2014, Coimbra - Portugal. A Geografia da Saúde no cruzamento de saberes. Coimbra Pt: Fundação para a Ciência e a Tecnologia, 2014. p. 729-732.
Resumos publicados em anais de congressos
OLIVEIRA FRIESTINO, J.K.; MENDONÇA, D., OLIVEIRA, P.; MOREIRA-FILHO, D. C. Cancer Incidence and survival among children and adolescents in Campinas: SP, Brazil. In: European Congress of Epidemiology: Healthy Living Conference on 25-27 June 2015 in Maastricht, the Netherlands. Abstracts [recurso eletrônico] / European Congress of Epidemiology 2015 Healthy Living. Maastricht, 2015. p. 2019-220.
OLIVEIRA FRIESTINO, J.K.; MOREIRA-FILHO, D. C. Leucemia Linfóide Aguda em crianças e adolescentes no município de Campinas: estudo da distribuição espacial dos casos. In: IX Congresso Brasileiro de Epidemiologia - EpiVix 2014, 2014, Vitória ES. Anais [recurso eletrônico] / IX Congresso Brasileiro de
Epidemiologia: as fronteiras da epidemiologia contemporânea: do conhecimento científico à ação. Rio de Janeiro: ABRASCO, 2014. p. 1613-1613.
OLIVEIRA SILVA, J.K.; MOREIRA-FILHO, D. C. Distribuição Espacial dos
Tumores do Sistema Nervoso Central em crianças e adolescentes no município de Campinas SP. In: 10º Congresso Brasileiro de Saúde Coletiva: Ciência para
cidadania, 2012, Porto Alegre. Anais Saúde Coletiva. Porto Alegre: Abrasco, 2012. OLIVEIRA SILVA, J.K.; MOREIRA-FILHO, D. C. Childhood Cancer and Spatial Distribution in a Brazilian City. In: 34th Annual Meeting of International Association of Cancer Registries, 2012, Cork - Ireland. Programme & Book of abstracts Annual Meeting of International Association of Cancer Registries. Lyon, França: IACR, 2012. v. 1. p. 116-116.
Apresentações de Trabalho
OLIVEIRA FRIESTINO, J.K.; MENDONÇA, D.; OLIVEIRA, P.; MOREIRA-FILHO, D. C. Cancer Incidence and survival among children and adolescents in Campinas: SP, Brazil. 2015. (Apresentação de Trabalho/Congresso).
77
OLIVEIRA FRIESTINO, J.K.; MOREIRA-FILHO, D. C. Câncer em crianças e adolescentes - Análise de sobrevida e distribuição espacial de casos registrados em um município brasileiro. 2014. (Apresentação de Trabalho/Congresso). OLIVEIRA FRIESTINO, J.K.; MOREIRA-FILHO, D. C. Leucemia Linfóide Aguda em crianças e adolescentes no município de Campinas: estudo da distribuição espacial dos casos. 2014. (Apresentação de Trabalho/Congresso).
OLIVEIRA SILVA, J.K.; MOREIRA-FILHO, D. C. Distribuição Espacial dos
Tumores do Sistema Nervoso Central em crianças e adolescentes no município de Campinas SP. 2012. (Apresentação de Trabalho/Congresso).
OLIVEIRA SILVA, J.K.; MOREIRA-FILHO, D. C. Childhood Cancer and Spatial Distribution in a Brazilian city. 2012. (Apresentação de Trabalho/Outra).
Resumos aprovados em congressos
OLIVEIRA FRIESTINO, J.K.; MENDONÇA, D.; OLIVEIRA, P.; OLIVEIRA, C. MOREIRA-FILHO, D. C. Incidence and Spatial Pattern of Childhood Cancer in Campinas: SP, Brazil. Submetido ao II Congreso Iberoamericano de Epidemiologia y Salud Publica em 2-4 Setembro 2015 em Santiago de Compostela, Espanha. OLIVEIRA FRIESTINO, J.K.; MENDONÇA, D.; OLIVEIRA, P.; ANTUNES, L., BENTO, M.J., MOREIRA-FILHO, D. C. Five-Year Survival of Childhood Cancer: Comparison of Campinas-SP (Brazil) and North Region of Portugal, 1996-2005. 11º Congresso Brasileiro de Saúde Coletiva, 25 Julho a 01 Agosto 2015 in Goiânia, Brasil.
MENDONÇA, D.; OLIVEIRA FRIESTINO, J.K.; OLIVEIRA, P.; OLIVEIRA, C., MOREIRA-FILHO, D. C. Modelos de Análise Estatística Espacial no Estudo da Incidência de Câncer em Crianças e Adolescentes. 11º Congresso Brasileiro de
78
Oliveira Friestino, JK; Mendonça, D; Oliveira, P.; Moreira Filho, DC
Cancer incidence and survival among children and adolescents in Campinas – SP, Brazil
Câncer em crianças e adolescentes: incidência e sobrevivência no município de campinas-SP, Brasil
Cancer incidence and survival among children and adolescents
Jane Kelly Oliveira Friestino1,2, Denisa Mendonça3, Pedro Oliveira3, Djalma de Carvalho Moreira Filho1
1
School of Medical Sciences, Department of Collective Health, University of Campinas - UNICAMP, Campinas, Brazil.
2
CAPES Foundation, Ministry of Education of Brazil, Brasília, Brazil.
3
Institute of Public Health - ISPUP, University of Porto, Porto, Portugal
Corresponding author
Jane Kelly Oliveira Friestino
School of Medical Sciences, Department of Collective Health, University of Campinas (UNICAMP)
Tel.: +55 19 3251-8018; Fax +55 19 3521-8044
R. Vital Brasil n.100, 3rd floor – Cidade Universitária Zeferino Vaz. Campinas - SP, 13.083-970– Brazil.
79
ABSTRACT
We analyse the cancer incidences and survival time of children and adolescents (0-19 years) residents in Campinas, Southeastern Brazil diagnosed in 1996-2005. The follow-up censored date was December 31, 2011. Cancers were classified according to the ICCC-3 Groups. By design, just the 4 most common groups were studied: Group I, Group II, Group III and Group IX. Age-standardized incidence rates were calculated using the world standard population. Kaplan-Meier estimation methods, Log-rank test and Cox regression were used for survival analysis. A spatial Bayesian hierarchical regression model (controlling for data heterogeneity and spatial autocorrelation), assuming that the number of cancer follows a Poisson distribution. A total of 180 cases were diagnosed during the study period. Overall crude incidence rate was 54.2 per million and standardized incidence rates was 59.1 (Group I-28.8, Group II-11.6, Group III-7.5 and Group IX- 4.3). Spatial differences were not found. Significant differences in survival times were found by diagnostic Groups adjusted for age and sex (p=0.001). In conclusion, higher incidence was found in Group-I and 5-year survival is higher in children than in adolescents similar to the findings reported in the literature.
80
RESUMO
Analisamos a incidência de câncer e sobrevivência de crianças e adolescentes (0- 19 anos) residentes em Campinas, região Sudeste do Brasil, diagnosticados entre 1996-2005. A data final do seguimento foi 31 de dezembro de 2011. Os cânceres foram classificados de acordo com os Grupos da CICI-3. Neste estudo, apenas foram considerados os 4 grupos mais incidentes: Grupo I, II; III e IX. As taxas de incidência padronizada foram calculadas utilizando a população mundial padrão. Métodos de estimativa de Kaplan-Meier, teste de log-rank e regressão de Cox foram utilizados para a análise de sobrevivência. Um Modelo Linear Misto Generalizado (GLMM), que inclui termos de efeitos aleatórios, com e sem estrutura espacial, foi construído com base em uma abordagem Bayesiana. Foi assumido que o número de casos incidentes de câncer segue uma distribuição de Poisson. Um total de 180 casos foram diagnosticados durante o período de estudo. A taxa de incidência global foi de 54,2 por milhão, e a incidência padronizada foi de 59,1 (Grupo I-28,8, Grupo II-11,6, Grupo III-7,5 e Grupo IX-4,3). Diferenças nos padrões espaciais não foram encontradas. Diferenças significativas em tempos de sobrevivência foram encontradas por Grupos de diagnóstico ajustados para idade e sexo (p=0,001). Conclui-se que os achados corresponderam aos apresentados pela literatura: incidência mais frequente para esta população foi o Grupo I e a sobrevivência em 5 anos é maior em crianças do que em adolescentes.
81
BACKGROUND
In Brazil, 32.9% of the population is under 19 years old(1) and although cancers among this age group are rare, 11,530 cases among Brazilians were estimated for 2012. Based on the 14 Population-Based Cancer Registries (PBCR) report, the median incidence corresponds to 154.3 per million children(2).
In the year 2011, in Brazil, neoplasms correspond to the second leading cause of death in children only surpassed by external causes(3).
Childhood cancers (age at diagnosis 0–19 years) comprise a variety of malignancies, with incidence varying worldwide by age, sex, ethnicity and geography, which provide insights into cancer etiology(4-6).
In developed countries, over the last four decades, advances in cancer treatment and in diagnostic techniques have resulted in an increased survival for children (ages 0-14 years) and adolescents (ages 15-19 years). It has been reported that in developed countries the 5-year survival increased from less than 30% in the 60s to approximately 80% in the 2000s for all childhood cancers(7, 8).
Campinas, a large Brazilian city, located in the Southwest area in Brazil, has a PBCR operating since 1990. Although population-based studies concerning childhood cancer incidence and survival are important to evaluate local health services, rarely studies are available for the region.
Better understanding of the incidence and spatial patterns of childhood cancer in the local area may help in better targeting health services and provide clues into the deficiencies of the healthcare system(9, 10) .However, innovative
82
analyses of registry data may provide partial information about childhood cancer and this is a gap in the scientific literature regarding this issue.
Therefore, this study analyses spatial patterns of cancer incidences and survival among children (0–14 years) and adolescents (15–19 years) residents in Campinas diagnosed in the period 1996–2005 and mortality between 1996 and 2011.
83
MATERIAL AND METHODS
Observational study of longitudinal cohorts retrospective of children and adolescents (0 - 19 years) in Campinas, São Paulo, Brazil, in the period 1996- 2005.
Campinas is an industrial city with approximately 1.1 million inhabitants(11), located in the State of São Paulo. About 30% of the population living in Campinas- SP is aged less than 19 years. During the study period there was an increase of 8.34% in the population under 19 years(11).
In 2000, Campinas was divided by the Municipal Health Department (MHD) into 47 Health Care Unit Areas (HCUA) which are the geographic units used in the spatial analysis. The population size under 19 years of these areas ranges from 1,099 to 7,036 with a median of 3,429.
The main health cancer care for children and adolescents occurs in the High Care Unit complexity in Paediatric Oncology, and according to assessment by Population-based cancer registry of Campinas, this unit has an 85% coverage of incident cases in the city for this age group(12).
Although Campinas has a PBCR, no data were available for the time period considered in this research. Therefore, this study was conducted in the mentioned Campinas referral service of High Care Unit complexity in Paediatric Oncology. This study used all data collection steps already validated for PBCRs and the main procedure for gathering data was active notification of cases.
The present study was conducted in accordance with Resolution 196/96 of the Brazilian National Health Council and was approved by the Dr. Domingos A.
84
Boldrini Children’s Centre for Hematologic Research Human Research Ethics Committee, number 006/2011 (Attach 5).
Case definition
Data were retrieved of all children and adolescents residing in Campinas- SP, aged 0–19 years, diagnosed between 1st
of January 1996 to 31st December 2005. Confirmed diagnosis of cancer was obtained from medical records, using the date of diagnoses of first primary tumour (Attach 6).
To avoid the inclusion of cases that did not belong to the population at risk, a data check was performed manually, comparing cases home address location to the covered geographical area.
Cancers were classified according to the International Classification for Childhood Cancer, third edition (ICCC - 3)(13). Benign tumours were excluded. By design, just the most common 4 groups were studied: Group I – Leukemias, Group II - Lymphomas and reticuloendothelial neoplasm, Group III - CNS and miscellaneous intracranial and intraspinal neoplasm and Group IX - Soft tissue and other extra osseous sarcomas.
Cancer Incidence
The person-years for the period under analysis were obtained by population estimates from the Brazilian Institute of Geography and Statistics (IBGE)(11). Crude and standardized incidence rates, calculated per 1,000,000 inhabitants, refer to the risk of new cases. The population denominators used for calculation of the incidence rates were based on the population census in 2000 and estimated projected numbers for the remaining years(11, 14). Age-specific incidence rates per
85
million inhabitants were calculated for children and adolescents(14) in the corresponding age-group, resident in Campinas city. Age-standardized incidence rates were calculated by the direct method, using the World standard population(15).
Spatial Analysis
The digital maps of administrative limits were produced by IBGE. We used the HCUA that corresponds to the area of coverage of health centres in the city and their respective population.
The measure used in disease mapping is the Standardized Incidence Ratio (SMR). In order to control for data heterogeneity and spatial autocorrelation between areas a statistical model using a spatial Bayesian hierarchical regression was implemented. The estimated Relative Risk (RR) and the 95% Credible Interval were obtained from described Poisson Model implemented in a Bayesian framework (16, 17) (Attach 7).
The analyses were performed using the softwares: SPSS, Geoda, Winbugs (WinBUGS14, Cambridge, UK)(18) and R version 2.14.1 (Project for Statistical Computing)(19), using R2WingBUGS package to connect the last two tools. WinBUGS uses Gibbs sampling, a specific Markov Chain Monte Carlo (MCMC) method, to produce samples from the posterior distribution of each parameter. ArcGis was used to perform the maps.
86 Survival Analysis
Mortality data from 1996 to 2011 were confirmed, by examining the medical records linked to the Campinas’s Mortality Registry and Health Mortality Information System, Department of Data and Information Technology (DATASUS)
(20)
, the information and statistics office of the Unified National Health System (SUS), representing the Ministry of Health(20).
The survival time was calculated as the difference between the date of diagnosis and the date of death or loss to follow-up. The analysis was performed with all cases followed until December 31, 2011. The follow-up for vital status was collected from the hospital medical records.
To calculate the lower and upper limits of the 95% confidence interval for a proportion, according to methods described by E. B. Wilson, the procedure with a correction for continuity was used(21).
Survival in each group was analysed using Kaplan-Meier estimation methods. Log rank test was used to compare the survival curves(22). The 5-year survival was estimated by Kaplan Meier method.
Cox regression models were used for the uni and multiple analyses, and assumptions of proportionality of hazards was verified by inspection of the log- hazard plots versus time(23).
87
RESULTS
A total of 180 cases were diagnosed, 154-children and 26-adolescents. The absolute frequency, distribution of tumour cases by diagnostic groups and the incidence rates per million by age stratified into two groups, 0–14 and 15-19 years, and diagnostic groups are shown in Table 1.
Incidence
The most frequent neoplasms were: Acute lymphoblastic (ALL) (children: 37.7%; adolescents: 15.4%), Acute non-lymphoblastic (children: 13%; adolescents: 23.1%), Astrocytomas (children: 11%) and Hodgkin lymphomas (adolescents: 19.2%) (Table 1).
Among children, the most frequent tumours were in Group – I which accounted for 51.9% of the overall number incident neoplasm, followed by the Group - III (22.7%), and the Group - II (18.1%). Among adolescents, the diagnostic Group – II represented 42.3% of the diagnoses, followed by the Group – I (38.4%), and Group - III (11.5%). Pronounced differences in the male/female ratio were observed for Group II.
Overall crude incidence rates (cases per million and 95% CI) was 54.2 (46.2 – 62.1); 64.2 in children (54.0 – 74.3), and 28.2 in adolescents (17.3 – 39.0). The annual overall standardized incidence rates (cases per million and 95% CI) were: 0-19 years, 56.5 (48.1 – 64.9); Group I - 28.8 (22.7 – 34.8), Group II - 11.6 (7.85 – 15.2), Group III – 11.9 (8.08 – 15.7) and Group IX - 4.3 (1.94 – 6.74). (Table 1).
88
Table 1. Frequency, Sex Ratio, and Incidence Rates per Million Inhabitants (1996-2005)
Diagnostic group Age (Years) Crude Incidence
Adjusted Incidence M/F* ratio 0 - 14 n(%) 15 - 19 n(%) Total n(%) 0 - 14 (106) 15 - 19 (106) Total (106) Total (106) I. Leukemias 1.3 80(52.0) 10(38.5) 90(49.9) 33.3 10.8 27.1 28.8
Ia. Acute lymphoblastic 1.5 58(37.7) 4(15.4) 62(34.4) 24.2 4.3 18.7 20.0 Ib. Acute non-
lymphoblastic 1.2 20(13.0) 6(23.1) 26(14.4) 8.3 6.5 7.8 8.1 Ic,d,e. Unspecified and
other specified leukemias 0.0 2(1.3) 0.0 2(1.1) 0.8 0.0 0.6 0.7
II.Lymphomas and reticuloendothelial
neoplasm 2.9 28(18.1) 11(42.2) 39(21.7) 11.7 11.9 11.7 11.6
IIa. Hodgkin lymphomas 1.6 8(5.2) 5(19.2) 13(7.2) 3.3 5.4 3.9 3.7 IIb. Non-Hodgkin
lymphomas except Burkitt
lymphoma 4.0 12(7.8) 3(11.5) 15(8.3) 5.0 3.3 4.5 4.3 IIc. Burkitt lymphoma 4.0 7(4.5) 3(11.5) 10(5.6) 2.9 3.3 3.0 3.1 IId,e. Unspecified lymphomas and miscellaneous lymphoreticular neoplasm --- 1(0.6) 0.0 1(0.6) 0.4 0.0 0.3 0.4 III. CNS and miscellaneous intracranial and intraspinal neoplasm 0.7 35(22.6) 3(11.5) 38(21.1) 14.6 3.3 11.4 11.9
IIIa. Ependymomas and
choroid plexus tumor 0.1 8(5.2) 0.0 8(4.4) 3.3 0.0 2.4 2.5 IIIb. Astrocytomas 0.8 17(11.0) 1(3.8) 18(10.0) 7.1 1.1 5.4 5.8 IIIc. Intracranial and
intraspinal embryonal
tumors 2.0 7(4.5) 2(7.7) 9(5.0) 2.9 2.2 2.7 2.6 IIId,e,f. Other specified
and unspecified intrarcranial and
intraspinal neoplasms 0.5 3(1.9) 0.0 3(1.7) 1.3 0.0 0.9 1.0
IX. Soft tissue and other extra osseous
sarcomas 2.3 11(7.1) 2(0.7) 13(7.2) 4.6 2.2 3.9 4.3
IXa.
Rhabdomyosarcomas 2.7 9(5.8) 2(0.7) 11(6.1) 3.8 2.2 3.3 3.7 IXd,e. Other specified
and unspecified soft
tissue sarcomas 1.0 2(1.3) 0.0 2(1.1) 0.8 0.0 0.6 0.7
Total 1.4 154(100) 26(100) 180(100) 64.2 28.2 54.2 56.5
89
The annual average age-standardized incidence rates of cancer among children and adolescents for the period, according to HCUA, is shown in Figure 1, where there is an accentuated geographic pattern, with the lowest incidence rate in the North and Eastern of Campinas.
Figure 1. Annual average age-standardized incidence rates of cancer among children and
90
The spatial distribution incidence rates found after use of Empirical Bayes approach, are shown in Figure 2, observing a less fragmented identification of possible more severe clusters for the occurrence of childhood cancer. Identify areas where they are concentrated neighbourhoods with higher detection rates: the neighbourhoods of South and West area of the city. When considering the rates estimated using the Empirical Bayesian, the values were between 1 and 159 new cases per million inhabitants.
91
Figure 2. Empirical Bayes of annual average age-standardized incidence rates of cancer
among children and adolescents for period 1996-2005 according to Health Centre Unit Area (Campinas-SP)
Figure 3 shows the map of the SMR by sex and age group, verifying that their spatial distribution HCUA is different from the distribution of the annual average rate observed in Figure 1. These differences reflect variation in the distribution of population by age and sex in each HCUA of Campinas. The highest
92
values of SMR are concentrated in the southwest and northwest regions of the municipality
Figure 3. Standardized Incidence Ratio of cancer among children and adolescents for
period 1996-2005 according to Health Centre Unit Area (Campinas-SP)
The relative risks (RR) were no statistically significant in HCUA. Figure 4 e 5 shows the RR of childhood cancer attributed to unobserved spatial effect.
93
Although some HCUA of the South and West present higher crude incidence rates, taking into account the spatial heterogeneity and the spatial autocorrelation no statistically significant differences in the SMR were observed among HCA. This may be related with unobserved variables, and future investigation needs to be performed.
Figure 4. Estimated Relative Risk of cancer incidence among children and adolescents for
94
Figure 5. Estimated Relative Risk and 95% Credible Intervals of cancer incidence among
children and adolescents for period 1996-2005 according to Health Centre Unit Area (Campinas-SP)
95 Survival
The median follow-up time in the cohort was 7.55 years (percentile 25:1.32; percentile 75: 11.54). Death was observed for 68 cases (37.8 %). Figure 6 shows the cumulative survival of children and adolescents by ICCC-3 and age groups. Statistically significant differences in the probability of survival were observed according to diagnostic Group (p = 0.001), but not to age groups (p = 0.071). Group IX and Group III showed lowest survival rates with median estimated survival 2.89 years and 2.91 years, respectively. Survival is much higher in Group II and Group I (87.1% and 63.1% survived 6 years) (Figure 6).
96
Figure 6. Survival rates of children and adolescents by groups and age, A (0 – 14 years),
B (15 – 19 years) between 1996-2005 in Campinas – SP, Brazil.
97
Overall 5-year survival was 65.6% (95% CI: 57.4% - 72.9%) among children and 53.8% (95% CI: 33.7% - 72.8%) among adolescents (Table 2). The 5-year survival was higher than 80% for Group II, both in children and adolescents. The most favourable prognosis was observed for those with Non-Hodgkin lymphomas except Burkitt lymphoma, with 91.7 % of children and 100% of adolescents alive 5 years after the diagnosis. The 5-year survival was lower than 40% among children for the diagnostic Group - IX, and among adolescents for the diagnostic Group III.
98
Table 2. Five-Year Survival (1996-2005)
Diagnostic group Age (0 - 14 years) Age (15-19 years) Total (0 - 19 years) Nr. of cases Survival* (%) (CI95%) Nr. of cases Survival* (%) (CI95%) Nr. of cases Survival* (%) CI I. Leukemias 80 67.5 56.0 - 77.3 10 50 23.6 - 76.3 90 64.5 53.5 - 74.0 Ia. Acute lymphoblastic 58 79.3 4 33.3 62 77.2 Ib. Acute non-
lymphoblastic 20 40.0 6 50 26 42.3 Ic,d,e. Unspecified
and other specified
leukemias 2 0 0 _ 2 0 II.Lymphomas and reticuloendothelial neoplasm 28 92.9 75.0 - 98.7 11 72.7 39.3 - 92.6 39 87.1 71.7 - 95.1 IIa. Hodgkin lymphomas 8 87.5 5 100 13 92.3 IIb. Non-Hodgkin lymphomas except Burkitt lymphoma 12 91.7 3 100 15 93.3 IIc. Burkitt lymphoma 7 100 3 33.3 10 80 IId,e. Unspecified lymphomas and miscellaneous lymphoreticular neoplasm 1 100 0 _ 1 100 III. CNS and miscellaneous intracranial and intr.neoplasm 35 48.6 31.7 - 65.7 3 33.3 1.7 - 87.4 38 47.4 31.3 - 63.9 IIIa. Ependymomas and choroid plexus
tumor 8 62.5 0 8 62.5 IIIb. Astrocytomas 17 58.8 1 100 18 61.1 IIIc. Intracranial and
intraspinal embryonal tumors 7 14.3 2 0 9 11.1 IIId,e,f. Other specified and unspecified intrarcranial and intraspinal neoplasms 3 33.3 2 _ 5 33.3
IX. Soft tissue and other extra osseous sarcomas
11 36.4 12.3 - 68.3 2 50 2.6 - 97.3 13 38.5 15.1 - 67.7
IXa.
Rhabdomyosarcomas 9 33.3 2 50 11 36.4 IXd,e. Other specified
and unspecified soft
tissue sarcomas 2 50.0 0 _ 2 50
Total 154 65.6 57.4 - 72.9 26 53.8 33.7 - 72.8 180 63.9 56.3 - 70.8
99
Table 3 shows the main results of Cox Regression models. The probabilities of overall survival for girls, adolescents, mother’s and father’s profession not specified, private health service, white ethnic group, were higher than those for other categories of these covariates (p > 0.05).
Significant differences in survival time were found in terms of diagnostic Groups adjusted for age and sex (p=0.001), with lower survival in the Group IX
(HRadj =6.38- Group IX vs Group II) followed by Group III (HRadj =6.27 Group III vs Group II)
100
Table 3. Cox Regression Analysis for Survival among Children and adolescents
Unadjusted hazard ratio Cox adjusted* hazard ratio (n= 180) Covariates HR (exp(b1)) 95% CI p-value HR (exp(b1)) 95% CI p-value ICCC -3 groups < 0.001 0.001 II. Lymphomas 1 1
IX: Soft tissue 6.22 2.031 19.045 6.388 2.084 19.580 III. CNS Tumors 5.372 2.220 14.271 6.271 2.293 17.153 I. Leukemias 3.408 1.333 8.717 3.851 1.480 10.020 Sex 0.86 0.529 Male 1 1 Female 1.041 0.644 1.685 0.851 0.498 1.381 Age group 0.518 0.154 0-14 1 1 15-19 1.237 0.648 2.362 1.620 0.790 32.010 Mother's profession 0.57