RESUMO
Composite end points are common primary outcomes in clinical trials. Their main benefit of utilizing a composite outcome is increasing the number of primary outcome events, meaning fewer participants are required to deliver an adequately powered trial. By combining multiple important end points in the primary outcome rather than having to select only 1, composite end points potentially make clinically meaningful benefits easier to detect and avoid ranking outcomes hierarchically. However, there are a number of important considerations when designing and interpreting clinical trials that utilize composite end points. In this Statistical Primer, issues with composite end points such as competing events, halo effect, risk of bias, time-to-event limitations and the win ratio are discussed in the context of real world clinical trials.
RESUMO
INTRODUCTION: In some studies, the peritoneal solute transfer rate (PSTR) through the peritoneal membrane has been related to an increased risk of mortality. It has been observed in the literature that those patients with rapid diffusion of solutes through the peritoneal membrane (high/fast transfer) and probably those with high average transfer characterized by the Peritoneal Equilibrium Test (PET) are associated with higher mortality compared to those patients who have a slow transfer rate. However, some authors have not documented this fact. In the present study, we want to evaluate the (etiological) relationship between the characteristics of peritoneal membrane transfer and mortality and survival of the technique in an incident population on peritoneal dialysis in RTS Colombia during the years 2007-2017 using a competing risk model. MATERIALS AND METHODS: A retrospective cohort study was carried out at RTS Colombia in the period between 2007 and 2017. In total, there were 8170 incident patients older than 18 years, who had a Peritoneal Equilibration Test (PET) between 28 and 180 days from the start of therapy. Demographic, clinical, and laboratory variables were evaluated. The (etiological) relationship between the type of peritoneal solute transfer rate at the start of therapy and overall mortality and technique survival were analyzed using a competing risk model (cause-specific proportional hazard model described by Royston-Lambert). RESULTS: Patients were classified into four categories based on the PET result: Slow/Low transfer (16.0%), low average (35.4%), high average (32.9%), and High/Fast transfer (15.7%). During follow-up, with a median of 730 days, 3025 (37.02%) patients died, 1079 (13.2%) were transferred to hemodialysis and 661 (8.1%) were transplanted. In the analysis of competing risks, adjusted for age, sex, presence of DM, HTA, body mass index, residual function, albumin, hemoglobin, phosphorus, and modality of PD at the start of therapy, we found cause-specific HR (HRce) for high/fast transfer was 1.13 (95% CI 0.98-1.30) pâ¯=â¯0.078, high average 1.08 (95% CI 0.96-1.22) pâ¯=â¯0.195, low average 1.09 (95% CI 0.96-1.22) pâ¯=â¯0.156 compared to the low/slow transfer rate. For technique survival, cause-specific HR for high/rapid transfer of 1.22 (95% CI 0.98-1.52) pâ¯=â¯0.66, high average HR was 1.10 (95% CI 0.91-1.33) pâ¯=â¯0.296, low average HR of 1.03 (95% CI 0.85-1.24) pâ¯=â¯0.733 compared with the low/slow transfer rate, adjusted for age, sex, DM, HTA, BMI, residual renal function, albumin, phosphorus, hemoglobin, and PD modality at start of therapy. Non-significant differences. CONCLUSIONS: When evaluating the etiological relationship between the type of peritoneal solute transfer rate and overall mortality and survival of the technique using a competing risk model, we found no etiological relationship between the characteristics of peritoneal membrane transfer according to the classification given by Twardowski assessed at the start of peritoneal dialysis therapy and overall mortality or technique survival in adjusted models. The analysis will then be made from the prognostic model with the purpose of predicting the risk of mortality and survival of the technique using the risk subdistribution model (Fine & Gray).
Assuntos
Diálise Peritoneal , Insuficiência Renal Crônica , Humanos , Colômbia/epidemiologia , Estudos Retrospectivos , Masculino , Feminino , Diálise Peritoneal/mortalidade , Pessoa de Meia-Idade , Insuficiência Renal Crônica/terapia , Insuficiência Renal Crônica/mortalidade , Adulto , Fatores de Tempo , Idoso , Peritônio/metabolismo , Taxa de Sobrevida , Soluções para Diálise/químicaRESUMO
The COVID-19 pandemic has had overwhelming global impacts with deleterious social, economic, and health consequences. To assess the COVID-19 death toll, researchers have estimated declines in 2020 life expectancy at birth (e0). When data are available only for COVID-19 deaths, but not for deaths from other causes, the risks of dying from COVID-19 are typically assumed to be independent of those from other causes. In this research note, we explore the soundness of this assumption using data from the United States and Brazil, the countries with the largest number of reported COVID-19 deaths. We use three methods: one estimates the difference between 2019 and 2020 life tables and therefore does not require the assumption of independence, and the other two assume independence to simulate scenarios in which COVID-19 mortality is added to 2019 death rates or is eliminated from 2020 rates. Our results reveal that COVID-19 is not independent of other causes of death. The assumption of independence can lead to either an overestimate (Brazil) or an underestimate (United States) of the decline in e0, depending on how the number of other reported causes of death changed in 2020.
Assuntos
COVID-19 , Causas de Morte , COVID-19/complicações , COVID-19/mortalidade , Estados Unidos/epidemiologia , Brasil/epidemiologia , Humanos , Masculino , Feminino , Recém-Nascido , Lactente , Pré-Escolar , Criança , Adolescente , Adulto Jovem , Adulto , Pessoa de Meia-Idade , Idoso , Idoso de 80 Anos ou mais , Neoplasias/complicações , Neoplasias/mortalidade , Cardiopatias/complicações , Cardiopatias/mortalidade , Diabetes Mellitus/mortalidade , Complicações do Diabetes/mortalidade , Causas de Morte/tendências , Tábuas de Vida , Expectativa de Vida/tendênciasRESUMO
As lesões por luxação representam o grupo de lesões traumáticas dento-alveolares (LTDA) caracterizadas pelo dano simultâneo ao feixe vásculo-nervoso periapical e estruturas de sustentação dentária, em diferentes graus de complexidade, dependendo da força e direção do impacto. A patogenia das alterações pulpares decorrentes das luxações é resultado da extensão da lesão do feixe vásculo-nervoso apical e de sua capacidade de reparo, que determinam os seguintes desfechos para a polpa: manutenção da vitalidade pulpar, manutenção da vitalidade pulpar com obliteração da cavidade pulpar (OCP) ou a necrose pulpar. Entretanto, o processo de reparo pulpar pode envolver fenômenos intermediários cujos sinais e sintomas se modificam ao longo do período de acompanhamento até um diagnóstico definitivo. Sendo assim, do ponto de vista clinico, o tempo até a observação destas respostas, bem como seus fatores determinantes, representa uma informação tão importante quanto o próprio desfecho, pois tem influência direta na tomada de decisão sobre as condutas clinicas mais adequadas. A presente pesquisa consistiu num estudo clínico longitudinal retrospectivo com o objetivo de avaliar o prognóstico pulpar de dentes permanentes portadores de lesões por luxação, seu padrão cronológico e fatores determinantes, entre os pacientes atendidos na Clínica de Traumatismos Dentários da Faculdade de Odontologia da Universidade Federal de Minas Gerais no período de 2014 a 2022. Foram analisados prontuários de 224 pacientes com 427 dentes permanentes portadores de luxações, acompanhados por um período mediano de 1,4 anos (variação de 11 dias a 6,9 anos). O padrão de resposta pulpar foi classificado como manutenção da vitalidade pulpar, OCP e necrose pulpar através da avaliação clínica e radiográfica. Uma análise de sobrevivência utilizando-se o modelo de riscos competitivos foi realizada e curvas de função de incidência acumulada (FIA) foram construídas para se avaliar as taxas de cada um dos desfechos pulpares utilizando-se o estimador de Aalen-Johansen. O efeito de variáveis independentes, incluindo sexo, idade do paciente no momento do trauma, estágio de desenvolvimento radicular, grupo dentário, tipo de luxação, presença de fratura coronária concomitante e prescrição de antibioticoterapia sistêmica (ATS), foi avaliado utilizando-se um modelo multivariado de Cox causa-específico. Os dados foram analisados no programa estatístico R (versão 4.0.4, Viena, Áustria, 2021). Os resultados demonstraram manutenção da vitalidade pulpar em 38,2% da amostra (163 dentes), necrose pulpar em 28,1% (120 dentes) e OCP em 12,9% da amostra (55 dentes). O tempo mediano para o diagnóstico de necrose pulpar foi de 3 meses, enquanto o tempo para vitalidade pulpar foi de 8 meses e para OCP foi de 16 meses A probabilidade de manutenção de vitalidade pulpar diminuiu significativamente com a ocorrência de fraturas coronárias de esmalte e dentina concomitantes (HR 0,38 IC 95% [0,2 0,8] p = 0,006). A taxa de necrose pulpar foi significativamente menor em dentes com forame apical amplo (HR 0,62 IC 95% [0,4 0,96] p = 0,03) mas aumentou significativamente em dentes com luxações associadas às fraturas coronárias (HR 4,0 IC 95% [2,6 - 6,1] p = 0,001) e em dentes portadores de luxações intrusivas (HR 2,3 IC 95% [1,2 - 4,1] p = 0,007). Dentes portadores de luxações laterais ou extrusivas (HR 3,0 IC 95% [1,3 6,9] p = 0,001) ou com forame amplo (HR 2,4 IC 95% [1,2 4,7] p=0.01) apresentaram as maiores taxas de OCP.
Luxation injuries are a group of traumatic dental injuries (TDI) that involve damage to both the pulp and periodontium, with different degrees of complexity. Pulp damage results from injury to the apical neuro-vascular bundle and, depending on the repair potential, may result in the following outcomes: maintenance of pulp vitality, maintenance of pulp vitality with pulp canal obliteration (PCO) or pulp necrosis. However, the pulp healing process may involve intermediate and reversible phenomena whose signs and symptoms mimic pulp necrosis, rendering the timing of these events critical for decision-making during the follow-up period. The present study consisted in a retrospective cohort to evaluate the pulpal prognosis of luxated permanent teeth, its chronological pattern and predictive factors, among patients treated at the Dental Trauma Clinic of the Federal University of Minas Gerais, from 2014 to 2021. Records of 224 patients with 427 permanent teeth with luxations, followed up for a median period of 1.4 years (ranging from 11 days to 6.9 years) were analysed. Pulp outcomes - maintenance of pulp vitality, PCO and pulp necrosis were defined trough clinical and radiographic evaluation. A competing risk survival analysis was performed and cumulative incidence function (CIF) curves were build using the Aalen-Johansen estimator to assess the rates of each of the pulp outcomes during the follow-up period. The effect of independent variables including patient's gender and age at the time of trauma, tooth group and stage of root development, type of luxation, concomitant injuries, systemic antibiotic therapy (SAT) prescription, was evaluated using a cause-specific Cox regression model. Data were analysed in the R statistical program (version 4.0.4, Vienna, Austria, 2021). The results showed maintenance of pulp vitality in 38.2% of the sample (163 teeth), pulp necrosis in 28.1% (120 teeth) and OCP in 12.9% of the sample (55 teeth). The median time for the diagnosis of pulp necrosis was 3 months, while the time for pulp vitality was 8 months and for PCO was 16 months. The hazards of pulp vitality significantly decreased with the occurrence of concomitant coronal fractures (HR 0 .38 95% CI [0.2 0.8] p = 0.006). The hazards of pulp necrosis was significantly lower in teeth with a wide apical foramen (HR 0.62 95% CI [0.4 0.96] p = 0.03) but increased significantly in teeth with luxations associated with crown fractures (HR 4.0 CI 95% [2.6 - 6.1] p = 0.001) and in teeth with intrusive displacement (HR 2.3 CI 95% [1.2 - 4.1] p = 0.007). Teeth with lateral or extrusive luxations (HR 3.0 CI 95% [1.3 6.9] p = 0.001) or with a wide foramen (HR 2.4 CI 95% [1.2 4.7] p =0.01) showed the highest rates of PCO.
Assuntos
Avulsão Dentária , Cicatrização , Traumatismos Dentários , Medição de Risco , Polpa DentáriaRESUMO
BACKGROUND: Breast cancer (BC) prevalence steadily increases in older patients but their treatment is based on the geriatric evaluations of comorbidities and potential benefits proved in clinical trials with younger patients. The lack of better decision-making tools urges to promote the search for new prognostic markers. The association between inflammation, aging and cancer may be crucial for better treatment selection. We sought to analyze its impact on the survival of older BC patients, evaluating the interaction with age and comorbidities. METHODS: We evaluated the relationship between inflammatory biomarkers at BC diagnosis (circulating blood cell counts and inflammatory indexes) and BC-related and not related mortality rate, evaluating the influence of comorbidities and age through the competitive risks assessment. RESULTS: We analyzed 148 consecutive BC patients aged ≥ 70 years old, diagnosed with BC and regional lymph node metastases. After the median follow-up of 51.5 months, 59 patients died (28 due to breast cancer progression and 31 because of other causes). Increased levels of circulating monocytes, neutrophils and neutrophil-to-lymphocytes ratio and decreased level of eosinophils and eosinophil multiple by neutrophils-to-lymphocytes ratio were associated with higher probability of BC-related death but not with death related to other causes. CONCLUSION: Our data suggest a role of inflammatory parameters as a possible prognostic tool in therapeutic decision-making process in older patients with BC, as increased level of inflammation was associated with cancer-specific mortality. Prospective studies may give the possibility of refining the geriatric evaluation for BC treatment in elderly.
Assuntos
Neoplasias da Mama , Idoso , Mama/patologia , Neoplasias da Mama/patologia , Feminino , Humanos , Inflamação/patologia , Prognóstico , Estudos ProspectivosRESUMO
RATIONALE & OBJECTIVE: Technique failure in peritoneal dialysis (PD) remains one of the most critical challenges of this therapy and is associated with a significant increase in costs and morbidity. Our objective was to estimate the frequency of PD technique failure and identify factors associated with technique failure. STUDY DESIGN: A retrospective multicenter observational cohort study. SETTING & PARTICIPANTS: All adult patients initiating PD between January 1, 2010, and December 31, 2015, with follow-up until December 31, 2018, at the Renal Therapy Services network in Colombia. EXPOSURE & PREDICTORS: PD modality (continuous ambulatory PD and automated PD) and demographic and clinical characteristics. OUTCOMES: Technique failure, defined as a switch to hemodialysis lasting at least 30 days. ANALYTICAL APPROACH: Sociodemographic and clinical characteristics of all patients were summarized descriptively according to modality. We estimated the cumulative incidence of technique failure, and a flexible parametric survival model with competing risks was used to evaluate factors associated with this outcome. RESULTS: Among 6,452 patients meeting inclusion criteria, 67% were treated with continuous ambulatory PD. The cumulative incidence of technique failure within 1 year of PD initiation adjusting for competing risks was 6.9% (95% CI, 6.3%-7.6%); within 2 years, technique failure was 13.5% (95% CI, 12.6%-14.4%); and within 3 years, 19.6% (95% CI, 18.5%-20.7%). Female sex, larger center size, and higher Kt/V were associated with lower risk for modality change, whereas diabetes, history of major abdominal surgery, catheter implant technique (laparotomy and percutaneous techniques), obesity, and peritonitis were associated with a higher likelihood of technique failure. LIMITATIONS: Variables of distance to the center, use of icodextrin, and measures of outcomes reported by patients were not included. CONCLUSIONS: Technique failure is relatively uncommon in Colombia; catheter-related problems are the most frequent cause of technique failure. Best practices in catheter insertion could minimize the risk for this outcome.
RESUMO
We introduce a numerically tractable formulation of Bayesian joint models for longitudinal and survival data. The longitudinal process is modeled using generalized linear mixed models, while the survival process is modeled using a parametric general hazard structure. The two processes are linked by sharing fixed and random effects, separating the effects that play a role at the time scale from those that affect the hazard scale. This strategy allows for the inclusion of nonlinear and time-dependent effects while avoiding the need for numerical integration, which facilitates the implementation of the proposed joint model. We explore the use of flexible parametric distributions for modeling the baseline hazard function which can capture the basic shapes of interest in practice. We discuss prior elicitation based on the interpretation of the parameters. We present an extensive simulation study, where we analyze the inferential properties of the proposed models, and illustrate the trade-off between flexibility, sample size, and censoring. We also apply our proposal to two real data applications in order to demonstrate the adaptability of our formulation both in univariate time-to-event data and in a competing risks framework. The methodology is implemented in rstan.
Assuntos
Modelos Estatísticos , Teorema de Bayes , Simulação por Computador , Humanos , Modelos Lineares , Estudos LongitudinaisRESUMO
INTRODUCTION: Both cytomegalovirus (CMV) infection and CMV disease have been linked with several long-term indirect effects in kidney transplant recipients. Research questions: We conducted a retrospective study to assess the association between cytomegalovirus disease and risks of death, shortterm cardiovascular events and graft loss in a cohort of renal transplant recipients. DESIGN: The associations between CMV disease and death and cardiovascular events were determined using Cox regression models, while the association between viral disease and graft loss risk was analyzed through a competing risks regression according to the Fine and Gray method. Death with a functioning graft was considered as a competing risk event. RESULTS: A total of 865 consecutive renal transplant recipients were included. The prevalence of seropositive donor/seronegative recipient (D+/R-) group was 89.9% with the remaining patients classified as seropositive recipient (R+). After median follow-up time of 24.4 months, CMV disease was not a risk factor for all-causes mortality (HR = 1.75; 95% CI 0.94-3.25), early cardiovascular events (HR = 0.54; 95% CI 0.16-1.82) or graft loss (subhazard ratio [the HR adjusted for competing risk of death with functioning graft] = 0.99; 95% CI 0.53-1.84). CONCLUSIONS: In this cohort with high prevalence of CMV IgG antibodies, we found no association between cytomegalovirus disease and risk of death or graft loss. The relationship between CMV and cardiovascular disease remains to be unraveled and probably corresponds to a multifactorial phenomenon involving individual risk factors and the immune response to infection rather than the virus effect itself.
Assuntos
Doenças Cardiovasculares , Infecções por Citomegalovirus , Transplante de Rim , Antivirais/uso terapêutico , Doenças Cardiovasculares/epidemiologia , Citomegalovirus , Infecções por Citomegalovirus/tratamento farmacológico , Infecções por Citomegalovirus/epidemiologia , Sobrevivência de Enxerto , Humanos , Imunoglobulina G , Estudos Retrospectivos , Fatores de Risco , Estudos Soroepidemiológicos , TransplantadosRESUMO
BACKGROUND/AIM: Tooth displacement during avulsion causes total rupture of the pulp's neurovascular supply. Revascularization and pulp healing may occur in immature teeth, which gives rise to the recommendation that root canal treatment may not be required. The aim of this study was to evaluate the prognostic factors for the pulp's response after replantation of young permanent teeth. METHODS: Records from 117 patients with 133 replanted permanent immature teeth were reviewed, and pulp outcomes were classified as healing (hard tissue deposition on the dentinal walls followed by narrowing of the pulp lumen or ingrowth of bone-like tissue inside the pulp canal) or non-healing (pulp necrosis with infection). The effect of clinical and demographic co-variates on the hazards of both outcomes was assessed performing a competing risk model. RESULTS: Pulp necrosis with infection was diagnosed in 78.2% of the teeth, and healing was observed in 12.8% of the teeth. A total of 12 teeth (9.0%) were censored due to prophylactic removal of the pulp or severe external root resorption caused by eruption of adjacent canines. The cs-Cox model demonstrated that the hazards of pulp healing increased in teeth with extra-alveolar periods <15 min (csHR: 7.83, 95% CI 1.76-34.80, p = .01), while the hazards of pulp necrosis with infection decreased (csHR: 0.31, 95% CI: 0.10-0.92, p = .04). Teeth replanted with Moorrees' stages 4 and 5 of root development had higher hazards of pulp necrosis with infection than teeth with stage 2 of root development (csHR: 2.23, 95% CI 1.11-4.50, p = .03; csHR: 2.89, 95% CI: 1.40-5.95; p = .01). CONCLUSIONS: Pulp healing rarely occurred after replantation of young permanent teeth being associated with short extra-alveolar periods <15 min. Early stages of root development decreased the hazards of pulp necrosis with infection.
Assuntos
Reabsorção da Raiz , Avulsão Dentária , Polpa Dentária , Necrose da Polpa Dentária/etiologia , Dentição Permanente , Humanos , Medição de Risco , Reabsorção da Raiz/etiologia , Reimplante DentárioRESUMO
OBJECTIVE: To determine risk factors for waitlist mortality in children with biliary atresia listed for liver transplantation. STUDY DESIGN: There were 2704 children with biliary atresia (<12 years of age) listed for a first liver transplant (2002-2018) in the United Network for Organ Sharing database. Fine-Gray regression models for competing risks analysis (main risk = waitlist mortality/delisting owing to too sick; competing risk = liver transplantation) were implemented to identify risk factors for waitlist mortality. RESULTS: The median waitlist time was 83 days (IQR, 34-191). The cumulative incidence of waitlist mortality was 5.2%. In multivariable analysis (n = 2253), increasing bilirubin level (P < .001), portal vein thrombosis (P = .03), and ventilator dependence (P < .001) at listing were associated with a higher risk, whereas weight ≥10 kg at listing (P = .009) was associated with a lower risk of waitlist mortality. When ascites at listing was included in multivariable analysis (n = 1376), it was associated with a higher risk for the composite outcome (P = .03). Encephalopathy at listing was not associated with waitlist mortality (n = 1376; P = .15). CONCLUSIONS: These parameters can be used to more objectively prioritize children with biliary atresia awaiting liver transplantation and identify children with biliary atresia-related end-stage liver disease at high-risk of mortality.
Assuntos
Atresia Biliar/cirurgia , Transplante de Fígado , Listas de Espera/mortalidade , Atresia Biliar/diagnóstico , Atresia Biliar/mortalidade , Criança , Pré-Escolar , Feminino , Humanos , Lactente , Masculino , Prognóstico , Estudos Retrospectivos , Fatores de Risco , Índice de Gravidade de Doença , Taxa de Sobrevida/tendências , Estados Unidos/epidemiologiaRESUMO
Semi-competing risks data include the time to a nonterminating event and the time to a terminating event, while competing risks data include the time to more than one terminating event. Our work is motivated by a prostate cancer study, which has one nonterminating event and two terminating events with both semi-competing risks and competing risks present as well as two censoring times. In this paper, we propose a new multi-risks survival (MRS) model for this type of data. In addition, the proposed MRS model can accommodate noninformative right-censoring times for nonterminating and terminating events. Properties of the proposed MRS model are examined in detail. Theoretical and empirical results show that the estimates of the cumulative incidence function for a nonterminating event may be biased if the information on a terminating event is ignored. A Markov chain Monte Carlo sampling algorithm is also developed. Our methodology is further assessed using simulations and also an analysis of the real data from a prostate cancer study. As a result, a prostate-specific antigen velocity greater than 2.0 ng/mL per year and higher biopsy Gleason scores are positively associated with a shorter time to death due to prostate cancer.
Assuntos
Algoritmos , Teorema de Bayes , Humanos , Incidência , Masculino , Cadeias de Markov , Análise de SobrevidaRESUMO
The clinical picture of patients with sickle cell anemia (SCA) is associated with several complications some of which could be fatal. The objective of this study is to analyze the causes of death and the effect of sex and age on survival of Brazilian patients with SCA. Data of patients with SCA who were seen and followed at HEMORIO for 15 years were retrospectively collected and analyzed. Statistical modeling was performed using survival analysis in the presence of competing risks estimating the covariate effects on a sub-distribution hazard function. Eight models were implemented, one for each cause of death. The cause-specific cumulative incidence function was also estimated. Males were most vulnerable for death from chronic organ damage (p = 0.0005) while females were most vulnerable for infection (p=0.03). Age was significantly associated (p ≤ 0.05) with death due to acute chest syndrome (ACS), infection, and death during crisis. The lower survival was related to death from infection, followed by death due to ACS. The independent variables age and sex were significantly associated with ACS, infection, chronic organ damage and death during crisis. These data could help Brazilian authorities strengthen public policies to protect this vulnerable population.
RESUMO
In this paper, we introduce a new model for recurrent event data characterized by a baseline rate function fully parametric, which is based on the exponential-Poisson distribution. The model arises from a latent competing risk scenario, in the sense that there is no information about which cause was responsible for the event occurrence. Then, the time of each recurrence is given by the minimum lifetime value among all latent causes. The new model has a particular case, which is the classical homogeneous Poisson process. The properties of the proposed model are discussed, including its hazard rate function, survival function, and ordinary moments. The inferential procedure is based on the maximum likelihood approach. We consider an important issue of model selection between the proposed model and its particular case by the likelihood ratio test and score test. Goodness of fit of the recurrent event models is assessed using Cox-Snell residuals. A simulation study evaluates the performance of the estimation procedure in the presence of a small and moderate sample sizes. Applications on two real data sets are provided to illustrate the proposed methodology. One of them, first analyzed by our team of researchers, considers the data concerning the recurrence of malaria, which is an infectious disease caused by a protozoan parasite that infects red blood cells.
Assuntos
Biometria/métodos , Malária/epidemiologia , Modelos Estatísticos , Brasil/epidemiologia , Humanos , Funções Verossimilhança , Distribuição de Poisson , Probabilidade , RecidivaRESUMO
A new flexible cure rate survival model is developed where the initial number of competing causes of the event of interest (say lesions or altered cells) follow a compound negative binomial (NB) distribution. This model provides a realistic interpretation of the biological mechanism of the event of interest as it models a destructive process of the initial competing risk factors and records only the damaged portion of the original number of risk factors. Besides, it also accounts for the underlying mechanisms that leads to cure through various latent activation schemes. Our method of estimation exploits maximum likelihood (ML) tools. The methodology is illustrated on a real data set on malignant melanoma, and the finite sample behavior of parameter estimates are explored through simulation studies.
RESUMO
Objetivo: Estimar o impacto na esperança de vida após a eliminação dos principais grupos de causas de morte no Nordeste brasileiro em 2000. Métodos: Dados sobre os óbitos registrados para os estados do Nordeste do Brasil em 2000 foram extraídos do Sistema de Informações sobre Mortalidade. Foram avaliadas as coberturas dos registros de óbitos menores de dez anos usando procedimentos indiretos. Empregou-se o método de Ledermann de redistribuição das causas de morte mal definidas, segundo o sexo. Construíram-se tábuas de vida de múltiplo decremento para os principais grupos de causas de morte, cujos impactos dos decrementos foram avaliados pela esperança de vida. Resultados: A eliminação total dos principais grupos de causas na mortalidade geral proporcionou para homens e mulheres, respectivamente, os seguintes ganhos na esperança de vida do Nordeste: doenças infecciosas e parasitárias (1,0 e 0,5 anos), neoplasmas malignos (1,4 e 1,5 anos), doenças do aparelho circulatório (4,0 e 4,0 anos), doenças do aparelho respiratório (1,1 e 0,8 anos) e causas externas (2,9 e 0,3 anos). Conclusões: Os níveis das coberturas dos óbitos foram aceitáveis, assim como as redistribuições das causas mal definidas para os estados. Considerando que as estatísticas vitais do Nordeste são pouco utilizadas, devido aos argumentos de que sua qualidade limita a produção de indicadores confiáveis da mortalidade, sua recuperação de forma direta possibilita resultados consistentes.
Assuntos
Masculino , Feminino , Humanos , Causas de Morte , Expectativa de Vida , Mortalidade , Tábuas de Vida , Estudos TransversaisRESUMO
Objetivo: Estimar la probabilidad de recurrencia en pacientes con cáncer de lengua según la edad, el estado ganglionar patológico y el tipo de tratamiento de los pacientes considerando la muerte antes de la recurrencia como evento competitivo.Material y métodos: Serie de casos retrospectiva de 290 pacientes con cáncer de lengua, con tratamiento en el INEN, entre los años 1977 y 2000. Se excluyeron del estudio 29 pacientes tratados solo con radioterapia. De los 261 pacientes, 31 solo tuvieron tratamiento del tumor primario. Resultados: La recurrencia fue 36,8%, la recurrencia local fue la más frecuente. La incidencia acumulada de la recurrencia a los 5 años, con el método de Kaplan Meier, se estimó en 44,7%, y según el análisis de riesgos competitivos en 42,4%. En el análisis univariado considerando riesgos competitivos, las incidencias acumuladas fueron: 56,8 y 38,9% en pacientes menores o iguales que 45 y mayores que 45 años respectivamente (p = 0,1556), 29,4% y50,5% en pacientes con ganglios patológicos negativos y positivos, respectivamente (p = 0,0002), y 37,5% en pacientes con cirugía, y 47,4% con radioterapia (p = 0,03). En el análisis multivariado, con regresión de riesgos competitivos, no se encontró diferencia entre los tipos de tratamiento sobre la recurrencia (RR = 1,146, p = 0,620). Conclusiones: La pequeña diferencia entre los resultados del método Kaplan Meier y el que toma en cuenta eventos competitivos se debe a la baja tasadel evento competitivo, más aún por que se trata de muerte no relacionada con la enfermedad. La tasa de recurrencia fuesimilar al reportado en la literatura. Sólo se encontró diferencia significativa en la tasa de recurrencia en el grupo con compromiso ganglionar positivo. Aunque la comparación con los métodos estándar, Kaplan Meier y Regresión de Cox, muestra resultados similares se deben tener en cuenta los eventos competitivos.
Objective: To estimate the probability of recurrence of tongue cancer, according age, stage pathological of lymph node and treatment type, considering death before recurrence as competing event. Materials and methods: Retrospective cases series study of 290 patients with tongue cancer, who were treated at INEN between 1977 and 2000. Twenty-nine patients who were treated only with radiotherapy were excluded. Of the remaining 261 patients, 31 were treated only at the primary tumor site. Results: Recurrence was 36.8%, local recurrence was most frequent. Five year cumulative incidence of recurrence, using method of Kaplan Meier, was estimated at 44.7%; and according to the analysis of competing risks in 42.4%. In univariate analysis, using competitive risks analysis, estimated cumulative incidences were 56.8 y 38.9% in patients less than or equal to 45 years and older than 45 years of age, respectively (p=0.1556); 29.4 y 50.5% in patients with negative andpositive pathological node, respectively (p=0.0002); 37.5% in patients with surgery, and 47.4% with surgery combined with radiotherapy, (p=0.03). In multivariate analysis, using competing risks regression, no difference was found between the types of treatment over recurrence (RR=1.146; p=0.620). Conclusions: The difference in results using Kaplan Meiermethod, and competing risks analysis was small, because there was a low rate of competitive event, and even more so when the cause of death was not related to the disease at hand. Recurrence rate was similar to those reported in other studies. Only significant difference was found in teh recurrence in the group with lymph node positive. Although the comparison with standard survival analysis methods shows similar results, competitive events should be taken into account.