PubMed Health⌕ Search

Biomedical subjects

Robin Henderson

Publications and source records attributed to Robin Henderson.

13 recordsLinked to original sources
PubMed2006

A relaxation of the gamma frailty (Burr) model.

Frailty models are used in univariate data to account for individual heterogeneity. In the popular gamma frailty model the marginal hazard has the form of a Burr model. Although the Burr model is very useful and can offer insight on the data, it is far from perfect. The estimation of the covariate effects is linked to the baseline hazard and this makes the model coefficients hard to interpret. At the same time, the frailties are assumed constant over time, while biological reasoning in some cases may indicate that frailties may be time dependent. In this paper we present a relaxation of the Burr model which is based on loosening the link between the estimation of the covariate effects and the baseline hazard. This can be achieved by replacing the cumulative baseline hazard in the Burr model by a set of time functions, and the frailty variance by a vector of coefficients directly estimated from the data using a partial likelihood. We illustrate the similarities of the model with the Burr model and a further extension of the latter, a model with an autoregressive stochastic process for the frailty. We compare the models on simulated data sets with constant and time-dependent frailties and show how the relaxed Burr models performs on two different real data sets. We show that the relaxed Burr model serves as a good approximation to the Burr model when the frailty is constant, and furthermore it gives better results when the frailty is time dependent.

Adolescent↗
PubMed2006

Estimation of optimal dynamic anticoagulation regimes from observational data: a regret-based approach.

A complication of long-term anticoagulation is that the optimal dose level varies not only between patients but over time within patients, in response to short-term changes in lifestyle. Consequently, doseage needs to be adaptive but there are as yet no accepted decision rules. Since anticoagulant use is increasing worldwide there is a need for more objective and routine procedures. In this paper, we describe an analysis of observational longitudinal anticoagulant data, aimed at determining an optimal reactive dose-changing strategy. We use the regret parameterization approach advocated by Murphy (J. R. Stat. Soc. Ser. B 2003; 65:331-366). Practical problems encountered in the implementation of the approach are discussed and illustrated.

Algorithms↗
PubMed2005

Analysis of interval-censored longitudinal data with application to onco-haematology.

The analysis of repeated measurements on a biomarker, either alone or jointly with the analysis of time to the event of interest, is an area of active research. Nevertheless, we are not yet able to deal in complete generality with these complex data, which frequently consist of error-prone, sparse and intermittent values. In many cancer studies, they arise in the framework of clinical trials and thus their relationship with prognosis is a primary focus. In such a setting, the Cox model is regarded as the standard technique for analysis. The aim of this work is to illustrate an alternative approach to the analysis of studies in which the biomarker values are complicated by interval censoring and an event occurs when the biomarker itself passes a certain threshold. We propose a linear mixed model with a Gaussian stochastic process that allows for interval-censored data and can be used both to track the biomarker trajectory and to estimate the probability of event occurrence. It is developed within the classic approach to longitudinal data analysis that was previously adapted for left-censored data, only. We apply this method to a study on the minimal residual disease (MRD) in childhood leukaemia. MRD is an interval-censored measurement of residual leukaemic cells that was scheduled at 9 time-points during treatment. The aim is to investigate the relationship between MRD and the disease process. Relapse, the event of interest, may conveniently be represented as MRD over a pre-defined threshold. Our focus is on modelling the probability of relapse conditional on MRD observed prior to it. Results show that the approach is promising as it allows proper description of the data, while maintaining flexibility of modelling, feasibility of computations and interpretability of results.

Biomarkers↗
PubMed2005

Goodness of fit of relative survival models.

Additive regression models are preferred over multiplicative models in the analysis of relative survival data. Such preferences are mainly grounded in practical experience with mostly cancer registries data, where the basic assumption of the additivity of hazards is more likely to be met. Also, the interpretation of coefficients is more meaningful in additive than in multiplicative models. Nonetheless, the question of goodness of fit of the assumed model must still be addressed, and while there is an abundance of methods to check the goodness of fit of multiplicative models, the respective arsenal for additive models is almost empty. We propose here a variety of procedures for testing the null hypothesis of a good fit. These are based on partial residuals defined similarly to Schoenfeld residuals familiar for Cox model diagnostics. The tests have appropriate sizes under the null hypothesis, and good power under different alternatives. We investigate their performance through simulations and apply the methods to data from a study into survival of colon cancer patients.

Adolescent↗
PubMed2005

Small sample bias in the gamma frailty model for univariate survival.

The gamma frailty model is a natural extension of the Cox proportional hazards model in survival analysis. Because the frailties are unobserved, an E-M approach is often used for estimation. Such an approach is shown to lead to finite sample underestimation of the frailty variance, with the corresponding regression parameters also being underestimated as a result. For the univariate case, we investigate the source of the bias with simulation studies and a complete enumeration. The rank-based E-M approach, we note, only identifies frailty through the order in which failures occur; additional frailty which is evident in the survival times is ignored, and as a result the frailty variance is underestimated. An adaption of the standard E-M approach is suggested, whereby the non-parametric Breslow estimate is replaced by a local likelihood formulation for the baseline hazard which allows the survival times themselves to enter the model. Simulations demonstrate that this approach substantially reduces the bias, even at small sample sizes. The method developed is applied to survival data from the North West Regional Leukaemia Register.

Acute Disease↗
PubMed2004

Modelling converging hazards in survival analysis.

The Cox proportional hazards model has become the standard model for survival analysis. It is often seen as the null model in that "... explicit excuses are now needed to use different models" (Keiding, Proceedings of the XIXth International Biometric Conference, Cape Town, 1998). However, converging hazards also occur frequently in survival analysis. The Burr model, which may be derived as the marginal from a gamma frailty model, is one commonly used tool to model converging hazards. We outline this approach and introduce a mixed model which extends the Burr model and allows for both proportional and converging hazards. Although a semi-parametric model in its own right, we demonstrate how the mixed model can be derived via a gamma frailty interpretation, suggesting an E-M fitting procedure. We illustrate the modelling techniques using data on survival of hospice patients.

Algorithms↗
PubMed2003

Intervention effects in observational survival studies with an application in total hip replacements.

Time to revision is a common and clinically relevant endpoint for studies of patients with total hip replacement. Because failures occur rarely within the first years after replacement, new surgical techniques and materials are often implemented without evidence of their effectiveness from randomized trials. Observational data may be available but this relies on the use of historical controls which has been heavily criticized. Instead the use of changepoint methods has been suggested to detect changes caused by successfully implemented interventions. In the setting of a proportional hazards model we develop a semi-parametric changepoint method to detect changes in baseline hazard. The procedure is motivated by and applied to a clinical study in patients with total hip replacements, where the effect of a new cement type is of interest. Power properties of the proposed method are investigated.

Adult↗
PubMed2003

Survival of hemodialysis patients: modeling differences in risk of dialysis centers.

OBJECTIVE: Dialysis is the most common renal replacement therapy for patients with end stage renal disease. This paper considers survival of dialysis patients, aiming to assess quality of renal replacement therapy at dialysis centers in Rio de Janeiro, Brazil, and to investigate differences in survival between health facilities. METHODS: A Cox proportional hazards model, allowing for time-varying covariates and prevalent data, was the basic method used to analyze the survival of 11,579 patients on hemodialysis in 67 health facilities in Rio de Janeiro State from January 1998 until August 2001, using data obtained from routine information systems. A frailty random effects model was applied to investigate differences in mortality between health centers not explained by measured characteristics. RESULTS: The individual variables associated with the outcome were age and underlying disease, with diabetes being the main isolated risk factor. Considering covariates of the health unit, two factors were associated with performance: bigger units had on average better survival times than smaller ones and units which offered cyclic peritoneal dialysis performed less well than those that did not. There were significant frailty effects among centers, with relative risks varying between 0.24 and 3.15, and an estimated variance of 0.43. CONCLUSIONS: Routine assessment based on health registries of the outcome of any high technology medical treatment is extremely important in maintaining quality of care and in estimating the impact of changes in therapies, units, and patient profiles. The frailty model allowed estimation of variation in risk between centers not attributable to any measured covariates. This can be used to guide more specific investigation and changes in health policies related to renal transplant therapies.

Ambulatory Care Facilities↗
PubMed2003

Diagnostics for joint longitudinal and dropout time modeling.

We present a variety of informal graphical procedures for diagnostic assessment of joint models for longitudinal and dropout time data. A random effects approach for Gaussian responses and proportional hazards dropout time is assumed. We consider preliminary assessment of dropout classification categories based on residuals following a standard longitudinal data analysis with no allowance for informative dropout. Residual properties conditional upon dropout information are discussed and case influence is considered. The proposed methods do not require computationally intensive methods over and above those used to fit the proposed model. A longitudinal trial into the treatment of schizophrenia is used to illustrate the suggestions.

Humans↗
PubMed2003

The use of bovine porous bone mineral in combination with collagen membrane or autologous fibrinogen/fibronectin system for ridge preservation following tooth extraction.

Preservation of the alveolar ridge following tooth extraction is desirable since it facilitates placement of endosseous implants and may improve the adverse esthetics often associated with fixed partial dentures. The purpose of this study was to compare the clinical effectiveness of bovine porous bone mineral (BPBM) used as a graft material combined with either guided tissue regeneration (GTR) or with the autologous fibrinogen/fibronectin system (AFFS) in preserving alveolar ridges following tooth extraction. Twenty-six patients who required extraction of two or more anterior or bicuspid teeth participated in a split-mouth design study. Following tooth extraction and elevation of a buccal full thickness flap, sockets were filled with bovine porous bone mineral which was then covered with either a collagen membrane or mixed and covered with an AFFS system. An acrylic stent served as a reference point for measurements. Primary flap closure was achieved in all surgical sites, and reentry surgery was performed at 6 months. Reentry surgery showed that BPBM/GTR sites presented with [1] significantly more internal socket bone fill (6.04 +/- 0.21 mm vs. 4.98 +/- 0.26 mm), [2] less, although not statistically significant, resorption of alveolar bone height (0.23 +/- 0.28 mm vs. 0.3 +/- 0.21 mm), and [3] significantly less horizontal resorption of the alveolar bony ridge as compared to BPBM/AFFS (1.06 +/- 0.28 mm vs. 2.60 +/- 0.25 mm). This study suggests that treatment of extraction sockets with a combination of bovine porous bone mineral and guided tissue regeneration is of slightly more benefit in preserving alveolar ridge dimensions following tooth extraction than treatment with a combination of bovine porous bone mineral and the autologous fibrinogen/fibronectin system.

Adult↗
PubMed2002

Identification and efficacy of longitudinal markers for survival.

Methods for the combined analysis of survival time and longitudinal biomarker data have been developed in recent years, with most emphasis on modelling and estimation. This paper focuses on the use of longitudinal marker trajectories as individual-level surrogates for survival. A score test for association which requires only standard methods for implementation is derived for the initial identification of candidate biomarkers. Methods for assessing efficacy of markers are discussed and a measure contrasting conditional and marginal distributions is proposed. An application using prothrombin index as biomarker for survival of liver cirrhosis patients is included.

Journal Article↗
PubMedDate not supplied

Bilateral implantation of low-profile interbody fusion cages: subsidence, lordosis, and fusion analysis.

BACKGROUND CONTEXT: The use of interbody fusion cages as a treatment for degenerative disc disease has become widespread. Low-profile cages have been developed to allow a closer fit when implanting bilateral cages in patients with smaller vertebral bodies. Some surgeons feel the open design also allows better bone contact and visualization. This is particularly true when two low-profile cages are used adjacent to one another. Because of the open design of low-profile interbody fusion cages, there has been concern regarding such issues as subsidence, lordosis and fusion rates. PURPOSE: This retrospective review of paired bilateral reduced profile interbody fusion cages was completed to assess changes in subsidence, lordosis and fusion. As a secondary goal, patient outcomes were measured to determine overall health since surgery and the patient's satisfaction with the spine surgery, in an attempt to assess the effect of the outcome variables cited supra. STUDY DESIGN: This was a retrospective evaluation of patients who underwent anterior lumbar interbody fusion with low-profile interbody fusion cages. PATIENT SAMPLE: Seventy-one consecutive patients who underwent bilateral implantation of low-profile interbody fusion cages were evaluated. OUTCOME MEASURES: A patient self-evaluation, which included a Short Form (SF)-36 and questions regarding patient satisfaction were administered to patients who were at least 1 year postoperative. Subsidence and lordosis measurements were completed. Fusion was assessed by the operating surgeon. METHODS: Low-profile interbody fusion cages (BAK/Proximity, Centerpulse Spine-Tech, Inc., Minneapolis, MN) were implanted bilaterally in at least one level from L3-L4 to L5-S1. Most patients had degenerative disc disease with leg and back pain that was not responsive to conservative treatment and demonstrated segmental instability or collapse. A small percentage of patients had either a degenerative spondylolisthesis (7.0%) or an isthmic spondylolisthesis (4.2%). Autograft harvested from the iliac crest was used in all cases. Demographic, surgical and follow-up data were retrospectively collected from patient charts. A clinical outcome questionnaire that included an SF-36 as well as questions regarding patient satisfaction was either mailed to each patient who was at least 1 year postsurgery or given to patients to complete at their 1-year visit. Patients were routinely followed radiographically before surgery, immediately after surgery and at 3, 6, 12 and 24 months after surgery. Fusion was assessed by the operating surgeon using lateral radiographs often in conjunction with a thin-slice computed tomography (CT) scan. Criteria for a successful fusion were lack of motion, anterior bridging bone and lack of lucencies on flexion/extension X-rays and/or contiguous bone through the cage using a thin-cut sagittal CT scan. Lateral X-rays on each patient were also measured for subsidence and lordosis changes. RESULTS: A total of 71 patients (45 men, 26 women) with a mean age of 43.4 years (range, 25 to 74) were evaluated. Thirty-six percent of the patients were smokers, and 96% were worker's compensation patients. Thirty-two percent of the patients had previous lumbar surgery. A total of 100 operative levels were evaluated. There were 45 one-level, 23 two-level and three three-level cases. Forty-nine percent were level L5-S1, 43% were L4-L5 and 8% were L3-L4. The mean duration of symptoms was 31.5 months. Mean surgical time, mean blood loss and mean hospital stay were 139 minutes, 186 cc and 3.34 days, respectively. There were no intraoperative or postoperative complications attributable to the construct and no cases of cage migration or collapse. Patients who were at least 1-year postsurgery and had follow-up X-rays or had undergone a CT scan at this time point were evaluated for fusion status. Sixty-three patients were assessed for fusion. Fifty-four (86%) of these patients were determined to have a solid fusion. Mean time to fusion was 10 months. Fusion was assessed as solid only if all operative levels were fully fused. Mean subsidence of the anterior region was 1.97 mm, whereas the mean subsidence of the posterior region was 0.82 mm. Lordosis was unchanged at all surgical levels with mean lordosis in L3-L4 decreasing only slightly from 13 degrees before surgery to 12 degrees after surgery. L4-L5 and L5-S1 showed only slight increases in lordosis changing from 17 to 18 degrees at L4-L5 and from 17 to 19 degrees at L5-S1. These changes were not statistically significant. The clinical outcome questionnaires had a return rate of 68%. Of the 48 patients who completed the questionnaire, 75% responded that they were happy with the surgical results and would definitely recommend the surgery to a friend. Sixty-seven percent agreed that surgery met their expectations or that surgery improved their condition enough that they would go through it again for the same outcome. The results of the SF-36 portion of the survey revealed that the physical and mental composite scores were within normal range of the US population that has experienced back pain or sciatica. CONCLUSION: Bilateral implantation of low-profile cages in this patient population led to satisfactory outcomes. Subsidence and changes in lordosis were minimal. Fusion rates were good, especially for one-level cases. Patient satisfaction was relatively high, considering the population consisted of 96% worker's compensation cases. With proper surgical technique, bilateral low-profile cages can be used effectively to treat patients with degenerative disc disease.

Health Status↗