Pseudo-observations in survival analysis

Andersen, Per Kragh; Perme, Maja Pohar

doi:10.1177/0962280209105020

Cited by 270 publications

(314 citation statements)

References 22 publications

Supporting

Mentioning

311

Contrasting

Unclassified

Order By: Relevance

“…The variance-covariance matrix of̂ * d can be estimated by a sandwich estimator. 12,13 We omit the details here as the variance estimator for the DR estimator we proposed in the next section does not depend on it.…”

Section: Model-based Direct Adjustmentsmentioning

confidence: 99%

“…based on the OR model (12). Alternatively the Cox model based prediction (11) can also be used for m 1 (t, Z i ,̂).…”

Section: The Proposed Dr Estimatormentioning

confidence: 99%

“…For example, the Cox model can be converted to a generalized linear model for pseudo observations. 12 For this model and a given time t, the complementary log-log link function g(x) = log(−log(x)) leads to …”

Section: Model-based Direct Adjustmentsmentioning

confidence: 99%

“…However, DR estimators can be constructed based on the IPW estimator (8) and model-based prediction using either (11) or (12) such that they are consistent as long as one of the models is correct. Doubly robust estimators for population means are simple to construct and easy to implement.…”

Section: The Proposed Dr Estimatormentioning

confidence: 99%

“…To this end, the pseudo-observation approach 12 is a powerful tool. The pseudo-observation approach constructs "pseudo observations" of the survival function, or its functions, for individual subjects such that they can be used for further analyses without censoring issues.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A simple, doubly robust, efficient estimator for survival functions using pseudo observations

Wang

2017

Pharmaceutical Statistics

View full text Add to dashboard Cite

Survival functions are often estimated by nonparametric estimators such as the Kaplan-Meier estimator. For valid estimation, proper adjustment for confounding factors is needed when treatment assignment may depend on confounding factors. Inverse probability weighting is a commonly used approach, especially when there is a large number of potential confounders to adjust for. Direct adjustment may also be used if the relationship between the time-to-event and all confounders can be modeled. However, either approach requires a correctly specified model for the relationship between confounders and treatment allocation or between confounders and the time-to-event. We propose a pseudo-observation-based doubly robust estimator, which is valid when either the treatment allocation model or the time-to-event model is correctly specified and is generally more efficient than the inverse probability weighting approach.The approach can be easily implemented using standard software. A simulation study was conducted to evaluate this approach under a number of scenarios, and the results are presented and discussed. The results confirm robustness and efficiency of the proposed approach. A real data example is also provided for illustration.

show abstract

Section: Model-based Direct Adjustmentsmentioning

confidence: 99%

“…based on the OR model (12). Alternatively the Cox model based prediction (11) can also be used for m 1 (t, Z i ,̂).…”

Section: The Proposed Dr Estimatormentioning

confidence: 99%

Section: Model-based Direct Adjustmentsmentioning

confidence: 99%

Section: The Proposed Dr Estimatormentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A simple, doubly robust, efficient estimator for survival functions using pseudo observations

Wang

2017

Pharmaceutical Statistics

View full text Add to dashboard Cite

show abstract

State Public Assistance Spending and Survival Among Adults With Cancer

Barnes,

Johnston,

Johnson

et al. 2023

JAMA Netw Open

View full text Add to dashboard Cite

ImportanceSocial determinants of health contribute to disparities in cancer outcomes. State public assistance spending, including Medicaid and cash assistance programs for socioeconomically disadvantaged individuals, may improve access to care; address barriers, such as food and housing insecurity; and lead to improved cancer outcomes for marginalized populations.ObjectiveTo determine whether state-level public assistance spending is associated with overall survival (OS) among individuals with cancer, overall and by race and ethnicity.Design, Setting, and ParticipantsThis cohort study included US adults aged at least 18 years with a new cancer diagnosis from 2007 to 2013, with follow-up through 2019. Data were obtained from the Surveillance, Epidemiology, and End Results program. Data were analyzed from November 18, 2021, to July 6, 2023.ExposureDifferential state-level public assistance spending.Main Outcome and MeasureThe main outcome was 6-year OS. Analyses were adjusted for age, race, ethnicity, sex, metropolitan residence, county-level income, state fixed effects, state-level percentages of residents living in poverty and aged 65 years or older, cancer type, and cancer stage.ResultsA total 2 035 977 individuals with cancer were identified and included in analysis, with 1 005 702 individuals (49.4%) aged 65 years or older and 1 026 309 (50.4%) male. By tertile of public assistance spending, 6-year OS was 55.9% for the lowest tertile, 55.9% for the middle tertile, and 56.6% for the highest tertile. In adjusted analyses, public assistance spending at the state-level was significantly associated with higher 6-year OS (0.09% [95% CI, 0.04%-0.13%] per $100 per capita; P &lt; .001), particularly for non-Hispanic Black individuals (0.29% [95% CI, 0.07%-0.52%] per $100 per capita; P = .01) and non-Hispanic White individuals (0.12% [95% CI, 0.08%-0.16%] per $100 per capita; P &lt; .001). In sensitivity analyses examining the roles of Medicaid spending and Medicaid expansion including additional years of data, non-Medicaid spending was associated with higher 3-year OS among non-Hispanic Black individuals (0.49% [95% CI, 0.26%-0.72%] per $100 per capita when accounting for Medicaid spending; 0.17% [95% CI, 0.02%-0.31%] per $100 per capita Medicaid expansion effects).Conclusions and RelevanceThis cohort study found that state public assistance expenditures, including cash assistance programs and Medicaid, were associated with improved survival for individuals with cancer. State investment in public assistance programs may represent an important avenue to improve cancer outcomes through addressing social determinants of health and should be a topic of further investigation.

show abstract

Factors Associated With Successful Discontinuation of Immune Suppression After Allogeneic Hematopoietic Cell Transplantation

et al. 2020

View full text Add to dashboard Cite

IMPORTANCE Immune suppression discontinuation is routinely attempted after allogeneic hematopoietic cell transplantation (HCT) and under current practices may lead to graft-vs-host disease (GVHD)-associated morbidity and death. However, the likelihood and predictive factors associated with successful immune suppression discontinuation after HCT are poorly understood.OBJECTIVES To examine factors associated with successful immune suppression discontinuation and risk for immune suppression discontinuation failure under conventional HCT approaches and develop a practical tool to estimate successful immune suppression discontinuation likelihood at the clinical point of care. DESIGN, SETTING, AND PARTICIPANTS Using long-term follow-up data from 2 national Blood and Marrow Transplant Clinical Trial Network studies (N = 827), a multistate model was developed to investigate the probability and variables associated with immune suppression discontinuation success. The study began in July 2015, and analyses were completed in August 2019. MAIN OUTCOMES AND MEASURES Immune suppression discontinuation and immune suppression discontinuation failure. RESULTSOf the 827 patients included in the analysis, 456 were men (55.1%). Median age at transplant was 44 (range, <1-67) years. With median follow-up of 72 (range, 11-124) months, 20.0% of the patients were alive and not receiving immune suppression at 5 years. Older recipient age (adjusted odds ratio [aOR] of >50 vs <30 years, 0.27, 99% CI, 0.14-0.50; P < .001), mismatched unrelated donor (aOR, mismatched unrelated vs matched related, 0.37; 99% CI, 0.14-0.97; P = .008), peripheral blood graft (aOR of peripheral blood graft vs bone marrow, 0.46; 99% CI, 0.26-0.82; P < .001), and advanced stage disease (aOR of advanced vs early disease, 0.45; 99% CI, 0.23-0.86, P = 0.002), were significantly associated with decreased odds of immune suppression discontinuation. Failed attempts at immune suppression discontinuation (127 patients [37.1% of total immune suppression discontinuation events]) resulting in GVHD were significantly associated with use of peripheral blood stem cells (HR, 2.62; 99% CI, 1.30-5.29; P < .001), prior GVHD, and earlier immune suppression discontinuation attempts. Earlier immune suppression discontinuation was not associated with protection from cancer relapse after HCT (adjusted hazard ratio for discontinuation vs not, 1.95; 99% CI, 0.88-4.31; P = .03). Dynamic prediction models were developed to provide future immune suppression discontinuation probability according to individual patient characteristics.CONCLUSIONS AND RELEVANCE Successful immune suppression discontinuation is uncommon in the setting of peripheral blood stem cell grafts. The data suggest earlier attempts at ISD conferred no long-term benefit, given frequent ISD failure, limited subsequent success after initial failed ISD attempt, and no evidence of relapse reduction. Using a risk model-based clinical application, physicians may be able to identify individual patients' probability of successful immu...

show abstract

Pseudo-observations in survival analysis

Cited by 270 publications

References 22 publications

A simple, doubly robust, efficient estimator for survival functions using pseudo observations

A simple, doubly robust, efficient estimator for survival functions using pseudo observations

State Public Assistance Spending and Survival Among Adults With Cancer

Factors Associated With Successful Discontinuation of Immune Suppression After Allogeneic Hematopoietic Cell Transplantation

Contact Info

Product

Resources

About