Estimation of competing risks with general missing pattern in failure types.

Anup Dewanji; Debasis Sengupta

doi:10.1111/j.0006-341x.2003.00122.x

Home/Archive/PubMed

PubMed · 14969486

Estimation of competing risks with general missing pattern in failure types.

Anup Dewanji·Debasis Sengupta

View original source ↗Export metadata ↓

Abstract

In competing risks data, missing failure types (causes) is a very common phenomenon. In this work, we consider a general missing pattern in which, if a failure type is not observed, one observes a set of possible types containing the true type, along with the failure time. We first consider maximum likelihood estimation with missing-at-random assumption via the expectation maximization (EM) algorithm. We then propose a Nelson-Aalen type estimator for situations when certain information on the conditional probability of the true type given a set of possible failure types is available from the experimentalists. This is based on a least-squares type method using the relationships between hazards for different types and hazards for different combinations of missing types. We conduct a simulation study to investigate the performance of this method, which indicates that bias may be small, even for high proportion of missing data, for sufficiently large number of observations. The estimates are somewhat sensitive to misspecification of the conditional probabilities of the true types when the missing proportion is high. We also consider an example from an animal experiment to illustrate our methodology.

Explore related subjects

Administration, Oral Algorithms Animals Biometry Carcinogens Male Models, Statistical Neoplasms, Experimental Nitrosamines Proportional Hazards Models Rats Risk Assessment Treatment Failure Water Supply Journal Article

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Anup Dewanji, Debasis Sengupta. 2003. Estimation of competing risks with general missing pattern in failure types.. https://doi.org/10.1111/j.0006-341x.2003.00122.x

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2007

Successful treatment with voriconazole of Aspergillus brain abscess in a boy with medulloblastoma.

Invasive aspergillosis is an increasing problem in immuno-incompetent patients after prolonged steroid therapy, cancer radio-chemotherapy, and bone marrow or solid organ transplantation. Cerebral aspergillosis is a well-described complication of the invasive aspergillosis but only in rare cases, the brain is the sole site of infection. Despite increasing availability of antifungal drugs, the prognosis of cerebral aspergillosis is poor. We report on an 11-year-old boy with medulloblastoma in the area of the fourth ventricle. Following tumor surgery and radio-chemotherapy, several abscess-like structures occurred in the operating field. After incomplete abscess, resection histology and culture confirmed a localized Aspergillus fumigatus infection. The initial treatment of the Aspergillus fumigatus infection with conventional amphotericin B failed, and treatment with the triazole voriconazole was started. Intravenous treatment with voriconazole resulted in a reduction of the Aspergillus fumigatus abscess. After switching to oral ambulatory therapy, the Aspergillus fumigatus abscess increased in size. To improve treatment, voriconazole dosage was adapted to reach drug concentrations in cerebrospinal fluid (CSF) above the minimal fungicidal concentration and plasma specimens. During the concentration-controlled voriconazole therapy for a period of 18 months, a complete response was achieved.

Administration, Oral↗

PubMed2007

Temozolomide and intravenous irinotecan for treatment of advanced Ewing sarcoma.

BACKGROUND: Preclinical models show sequence-dependent synergy with the combination of temozolomide and irinotecan, and a Phase I trial has demonstrated the combination to be tolerable and active in children with relapsed solid tumors. Because responses were seen in patients with Ewing sarcoma (ES) on that trial, additional patients were treated with this combination following study completion. PROCEDURE: We reviewed data from all ES patients treated with temozolomide and irinotecan at four institutions, including seven patients treated on the above Phase I trial. RESULTS: Sixteen patients received a total of 95 courses, with a median of five courses per patient. All patients had either progressive disease (PD) during initial therapy (n = 5) or relapse within 2 years of diagnosis (n = 11). Twelve patients had metastatic disease at diagnosis, including 5 with bone and/or marrow metastases. Patients received oral temozolomide 100 mg/m(2)/day on days 1-5 plus intravenous irinotecan 10-20 mg/m(2)/day on days 1-5 and 8-12, with courses repeated every 21-28 days. We observed 1 complete, 3 partial, and 3 minor responses in 14 evaluable patients, with a median duration of response of 30 weeks. Planned 21-day courses were tolerable and no more toxic than 28-day courses. Myelosuppression was minimal despite heavy pretreatment. Grade 3-4 diarrhea occurred in 11% of courses and was related to higher irinotecan doses. Over 600 irinotecan doses were administered uneventfully at home. CONCLUSIONS: Temozolomide and protracted intravenous irinotecan given in 21-day courses was tolerable and active in patients with advanced ES. Home administration of irinotecan with temozolomide was safe and is reasonable palliative therapy. A formal Phase II study using a uniform dose and schedule is warranted to better define activity.

Administration, Oral↗

PubMed2007

Medical and non-medical barriers to outpatient treatment of fever and neutropenia in children with cancer.

BACKGROUND: A number of clinical trials have employed clinical criteria that can identify pediatric patients at low-risk for complicated episodes of fever and neutropenia (F&N) and have successfully treated low-risk patients in the outpatient setting. Despite this, inpatient management remains the standard of care. This trial tested the hypothesis that a strategy of initial hospitalization followed by continuation of therapy in the outpatient setting could be practically implemented in the majority of episodes. PROCEDURE: Patients presenting with F&N were initially evaluated to determine if they had high-risk clinical criteria that would exclude them from this approach. Eligible patients were then hospitalized and treated with iv antibiotics. On subsequent days the attending physician determined whether the patient had exhibited improvement and could continue therapy in the outpatient setting with oral antibiotics. Outpatients were seen three times weekly and continued antibiotics until recovery from F&N. RESULTS: Outpatient oral antibiotic therapy was practically implemented in less than one-quarter of episodes of pediatric F&N. Forty-nine percent of episodes were excluded from study by medical and social protocol exclusion criteria. One hundred five episodes were enrolled and among these 59 episodes included outpatient management. Common barriers to outpatient care included serious medical comorbidities, non-medical barriers including language and distance of residence from the medical center, and lack of interest. The average duration of outpatient care was 3.6 days following an average of 3.5 days of hospitalization. Ninety percent did not require rehospitalization. They experienced no complications. CONCLUSIONS: In only a minority of episodes can outpatient antibiotic management be implemented. Medical comorbidities and social barriers can make the transition to outpatient care difficult. However, initial hospitalization followed by oral antibiotic outpatient management appears safe and effective for low-risk patients who exhibit good responses to initial antibiotic therapy in hospital.

Administration, Oral↗