Biomedical subjects

K B Dear

Publications and source records attributed to K B Dear.

At least 19 recordsLinked to original sources

PubMed2001

Testing for the presence of cured patients: a simulation study.

An important but difficult problem in clinical trials is to determine the presence of cured patients when long-term survivors are observed. The likelihood ratio test has been studied for this purpose in the gamma mixture model. However, its asymptotic null distribution is not readily available due to a violation of boundary conditions in the standard asymptotic theory. In this paper, a simulation study is employed to examine a proposed asymptotic null distribution of the likelihood ratio test. We find that the distribution can also be used to approximate the asymptotic null distribution of the likelihood ratio test in the Weibull and log-normal mixture models when the censoring rate is not too light. However, the simulation study also shows that null distribution of the likelihood ratio test deviates significantly from the suggested distribution under moderate sample sizes when the censoring rate is small or the hazard rate is large. Consequently caution is needed in this case to determine the presence of cured patients. Finally, the results are used to confirm the presence of cured patients in a leukaemia study.

Bone Marrow Transplantation↗

PubMed2000

A nonparametric mixture model for cure rate estimation.

Nonparametric methods have attracted less attention than their parametric counterparts for cure rate analysis. In this paper, we study a general nonparametric mixture model. The proportional hazards assumption is employed in modeling the effect of covariates on the failure time of patients who are not cured. The EM algorithm, the marginal likelihood approach, and multiple imputations are employed to estimate parameters of interest in the model. This model extends models and improves estimation methods proposed by other researchers. It also extends Cox's proportional hazards regression model by allowing a proportion of event-free patients and investigating covariate effects on that proportion. The model and its estimation method are investigated by simulations. An application to breast cancer data, including comparisons with previous analyses using a parametric model and an existing nonparametric model by other researchers, confirms the conclusions from the parametric model but not those from the existing nonparametric model.

Algorithms↗

PubMed2000

Analysis of variation in success rates in conjunctival autografting for primary and recurrent pterygium.

AIMS: To evaluate the success rates of conjunctival autografting for primary and recurrent pterygium performed in a tertiary ophthalmic centre. METHODS: The outcome of 139 cases with primary pterygia and 64 cases with recurrent pterygia who underwent excision with conjunctival autografting was retrospectively reviewed. Outcome was evaluated in terms of recurrence of pterygia onto the cornea. The recurrence rates were determined using Weibull survival functions, in a mixture model that included a component allowing for cure. The suitability of this model was verified using Turnbull's non-parametric method for interval censored data (1974). Estimated recurrence free probabilities were based on the fitted Weibull survival curves. RESULTS: Mean follow up was 8.4 months in the primary group, and 9.5 months for the recurrent group. 29 out of 139 cases of primary pterygia recurred (20.8%) while 20 out of 64 cases in the recurrent group (31.2%) recurred. Recurrence rates varied widely among surgeons, ranging from 5% to 82%. Recurrence rates were inversely related to previous experience in performing conjunctival grafting. The recurrence free probability was 84% at 3 months, 73% at 1 year for primary pterygia, and 80% at 3 months, 67% at 1 year for recurrent pterygia. There was no statistical difference in recurrence rates between primary and recurrent groups (p= 0.80). CONCLUSION: The success of conjunctival autografting for pterygium in this series varies widely, and may be related to a significant learning curve or differing surgical techniques for this procedure. This may account for the wide variation in reported success of this procedure in the ophthalmic literature.

Adult↗

PubMed2000

The familial incidence of symptoms of scotopic sensitivity/Irlen syndrome: comparison of referred and mass-screened groups.

The familial incidence of Scotopic Sensitivity/Irlen Syndrome was investigated in two samples. One sample involved parents and siblings of 126 children identified with symptoms who had been referred for screening. The other sample involved parents and siblings of 33 children who had been identified with symptoms through mass screening of all children in Grades 3 to 6 at two local schools. Two different samples were taken to investigate the possibility of parental referral bias. Familial incidence may be inflated in a referred sample because some parents may be aware of their own symptoms and actively seek assistance. For the sample of children referred for screening, there was an 81% chance of either one or both parents showing similar symptoms and a 76% chance of siblings being similarly affected. For the sample of children identified through school screening, there was an 85%, chance of either one or both parents showing similar symptoms and a 54% chance of siblings being similarly affected. The data confirm previous estimates of incidence and suggest that Scotopic Sensitivity/Irlen Syndrome may be a genetically-based deficit in visual processing.

Adult↗

PubMed1998

A generalized F mixture model for cure rate estimation.

Cure rate estimation is an important issue in clinical trials for diseases such as lymphoma and breast cancer and mixture models are the main statistical methods. In the last decade, mixture models under different distributions, such as exponential, Weibull, log-normal and Gompertz, have been discussed and used. However, these models involve stronger distributional assumptions than is desirable and inferences may not be robust to departures from these assumptions. In this paper, a mixture model is proposed using the generalized F distribution family. Although this family is seldom used because of computational difficulties, it has the advantage of being very flexible and including many commonly used distributions as special cases. The generalised F mixture model can relax the usual stronger distributional assumptions and allow the analyst to uncover structure in the data that might otherwise have been missed. This is illustrated by fitting the model to data from large-scale clinical trials with long follow-up of lymphoma patients. Computational problems with the model and model selection methods are discussed. Comparison of maximum likelihood estimates with those obtained from mixture models under other distributions are included.

Confidence Intervals↗

PubMed1998

Tests for homogeneity of the risk difference when data are sparse.

Test statistics for the homogeneity of the risk difference for a series of 2 x 2 tables when the data are sparse is proposed. A weighted least squares statistic is commonly used to test for equality of the risk difference over the tables; however, when the data are sparse, this statistic can have anticonservative Type I error rates. Simulation is used to compare the proposed test statistics to the weighted least squares statistic. The weighted least squares statistic has the most anticonservative Type I error rates of all the statistics compared. We suggest the use of one of our proposed test statistics instead of the weighted least squares statistic.

Antineoplastic Agents↗

PubMed1997

Estimating correlations from epidemiological data in the presence of measurement error.

Analysis of a major multi-site epidemiologic study of heart disease has required estimation of the pairwise correlation of several measurements across subpopulations. Because the measurements from each subpopulation were subject to sampling variability, the Pearson product moment estimator of these correlations produces biased estimates. This paper proposes a model that takes into account within and between sub-population variation, provides algorithms for obtaining maximum likelihood estimates of these correlations and discusses several approaches for obtaining interval estimates.

Adult↗

PubMed1997

Effect of pterygium morphology on pterygium recurrence in a controlled trial comparing conjunctival autografting with bare sclera excision.

OBJECTIVES: To compare success rates of conjunctival autografting and bare sclera excision for primary and recurrent pterygium in the tropics and to evaluate risk factors for pterygium recurrence. METHODS: A prospective, controlled clinical trial was performed in which 123 primary and 34 recurrent pterygia, matched for age and pterygium morphology, were randomized in 2 separate studies to receive either bare sclera excision or conjunctival autograft. The surgical procedures were performed by one surgeon and reviewed at 1, 3, 6, and 12 months after surgery by an independent observer. Pterygium morphology was clinically graded as atrophic, intermediate, or fleshy according to an assessment of pterygium translucency. Risk factors were assessed using likelihood ratio tests. Weibull curves were used to estimate recurrence rates allowing for the interval censoring. RESULTS: In the group with primary pterygium (mean follow-up, 15.1 months), 38 (61%) of the 62 cases of bare sclera excision (heretofore referred to as the bare sclera group) had pterygium recur in contrast with 1 (2%) of the 61 cases of conjunctival autograft (heretofore referred to as the conjunctival autograph group) (P<.001, likelihood ratio X2 test). Nontranslucency, or fleshiness of the pterygium, and not age was a significant risk factor for recurrence in the bare sclera group (P<.001, likelihood ratio X2 test). In the group with recurrent pterygium (mean follow-up, 13.2 months), 14 (82%) of the 17 bare sclera group had pterygium recur, while no recurrences occurred among 17 cases in the conjunctival autograft group. Nontranslucency was again a highly significant factor for recurrence (P<.001, likelihood ratio X2 test). CONCLUSIONS: Pterygium recurrence is related to pterygium morphology and fleshiness of the pterygium is a significant risk factor for recurrence if bare sclera excision is performed. Conjunctival autografting for primary and recurrent pterygium is effective in reducing pterygium recurrence compared with bare sclera excision.

Adult↗

PubMed1997

A phase III double-blind randomised study of rectal sucralfate suspension in the prevention of acute radiation proctitis.

BACKGROUND AND PURPOSE: A limited number of studies have suggested that oral sucralfate reduces the acute and late gastro-intestinal side-effects of pelvic radiotherapy and sucralfate enemas ameliorate symptoms of chronic proctitis. Sucralfate may act via local bFGF at the mucosal level in promoting angiogenesis and reducing epithelial associated microvascular injury. This multi-institutional study was designed to test the hypothesis that sucralfate given as an enema would have a significant protective effect against acute radiation induced rectal injury by direct application to the mucosa. MATERIALS AND METHODS: Eighty-six patients having radiotherapy for localised carcinoma of the prostate were randomised in a double-blind placebo-controlled study to receive either 15 ml of placebo suspension or 3 g of sucralfate in 15 ml suspension, given as a once daily enema during and for 2 weeks following radiotherapy. Assessment was based on the EORTC/RTOG acute toxicity criteria and a patient self-assessment diary. RESULTS: There was no significant difference between placebo and sucralfate for peak incidences of EORTC/RTOG proctitis. For the placebo and sucralfate arms 95 and 88% (difference 7 +/- 11%) suffered some degree of proctitis, with 71 and 61% (difference 10 +/- 19%) reaching grade 2, respectively. The median period to onset of grade 2 proctitis was 33.5 and 36 days, with the median duration being 9.5 and 15 days, respectively, again these difference being non-significant. Thirty-five and 37% of patients rated the effect of radiotherapy on bowel habit as 'a lot' with a moderate or severe effect on normal daily living in 52 and 49%, respectively. CONCLUSION: This study suggests that sucralfate given as a once daily enema does not substantially reduce the incidence of symptoms associated with acute radiation proctitis and its routine clinical use cannot be recommended. This cohort of patients will be followed to determine if any difference develops in relation to late toxicity.

Acute Disease↗

PubMed1996

The follicular non-Hodgkin's lymphomas--I. The possibility of cure.

The follicular lymphomas pursue an indolent course in many patients. Long-term follow-up in large series is therefore necessary to establish whether cure is taking place, and if so, at what stage in the dissemination of the disease process it becomes unlikely. The time to, and site of relapse, together with its impact on survival has been studied in 398 patients entered into the British National Lymphoma Investigation limited and disseminated disease trials between 1974 and 1980. Relapse data were compared with various models to obtain maximum likelihood estimates of the proportions permanently remaining relapse-free following treatment. Long-term relapse-free survival was observed in 54.8 +/- 14.9% (95% CI) of patients at 15 years with Ann Arbor stage I disease, 29.2 +/- 13.6% in patients with stage II disease, 18.1 +/- 6.6% with stage III and 13.0 +/- 5.9% with IV disease. Relapse time-course data for all trial arms conform closely to lognormal distributions allowing maximum likelihood estimates of proportions remaining permanently relapse-free to be derived. Using this methodology, over a quarter of patients treated with involved radiotherapy alone or radiotherapy plus 6 months of chlorambucil in the limited disease (Ann Arbor stage I and II) trial are unlikely to relapse at any time in the future. Over 10% of patients treated in the disseminated disease trials with disease classified as Ann Arbor stage III are also statistically unlikely to relapse. The finding that a proportion of patients is statistically unlikely to experience a clinically obvious relapse is consistent with clinical cure. It is especially interesting that a small proportion of patients with disseminated disease and treated by chemotherapy have fallen into this category, but additional data are required to know at what point statistical cure becomes unlikely. Whether "clinical cure" is the same as "pathological cure" in this disease remains uncertain.

Antineoplastic Combined Chemotherapy Protocols↗

PubMed1996

The follicular non-Hodgkin's lymphomas--II. Prognostic factors: what do they mean?

The Ann Arbor staging classification has long been recognised to have shortcomings when used to stage the follicular lymphomas. To date, the identification of important prognostic variables has not succeeded in producing a superior staging classification that reflects the stages of dissemination of these processes in a way that can be used in the testing of new therapeutic strategies. A fresh look is taken at these factors. Data from 398 patients entered into the British National Lymphoma Investigation trials between 1974 and 1980, were analysed to evaluate the performance of the Ann Arbor staging classification. Multiple regression and proportional hazards techniques were used to determine what factors independently influence response to initial treatment, the durability of that response and ultimate survival, and to isolate factors that relate to disease progression from those that have other mechanisms of action. The Ann Arbor staging classification fared poorly, minimally separating relapse-free and cause-specific survival probabilities in patients with the largest staging groupings, III and IV. Significant prognostic heterogeneity was seen in both of these stage groupings, with 22% of patients with stage IV disease on the basis of marrow involvement having slightly better outcomes than patients with stage III disease. Significant differences in outcome were also observed between patients of different age and sex in each Ann Arbor stage grouping. Increasing number of lymph node regions involved, constitutional symptoms, the presence of splenomegaly and increasing age were observed to have powerfully independent adverse influence on probability of complete response to treatment and cause-specific survival. The evolution of the follicular lymphomas is reflected at the clinical level by an increase in the number of lymph node regions involved and splenomegaly. Simple classifications based on simple counts of lymph node regions involved and splenomegaly are more successful than the Ann Arbor staging classification in subdividing the series into patient subgroups that, regardless of gender or age, experience significantly different probabilities of responding completely to therapy and, as a consequence, relapse-free and cause-specific survival expectations. The definition of poor prognosis in subgroups may be of value in selecting patients for newer and more intensive therapeutic approaches.

Age Factors↗

PubMed1996

The familial incidence of symptoms of Scotopic Sensitivity/Irlen syndrome.

The familial incidence of Scotopic Sensitivity/Irlen Syndrome was investigated using parents of 751 children identified with symptoms. Children were identified by methods independent of their parents' symptoms or lack of symptoms. For these children, there was an 84% chance of either one or both parents showing similar symptoms, with similar numbers of mothers identified with symptoms as fathers. The data suggest that Scotopic Sensitivity/Irlen Syndrome may be a genetically based deficit in visual processing, but the simplest genetic models do not appear to fit.

Adolescent↗

PubMed1995

Acute reaction parameters for human oropharyngeal mucosa.

The purpose of this study was to determine the influence of changes in dose rate over the range 0.8-240 Gy/h on acute oropharyngeal mucosal reactions in human subjects, and to estimate the values of the important parameters that influence these reactions. Sixty-one patients requiring radiotherapy to palliate incurable head and neck cancer were treated on a telecaesium unit, using opposing lateral portals to total midline doses, varying between 30 and 42 Gy in 10 daily fractions over 2 weeks, at dose rates of 0.8, 1.8, 3.0 and 240 Gy/h according to a central composite study design. The severity and time course of reactions were charted at least twice weekly for each patient, using the EORTC/RTOG acute mucosal reaction grading system. Duration of reaction at each grade was observed to provide a more sensitive reflection of effect than the proportion of patients reaching any particular reaction grade. Analysis of duration by direct and indirect methods suggest alpha/beta ratios in the range 7-10 Gy and half-time (t1/2) values in the range 0.27-0.5 h, if mono-exponential repair kinetics are assumed. The t1/2 values are short and raise the question as to whether the repair kinetics of this tissue are well described by a mono-exponential function. Further prospective studies involving multiple daily fraction treatment regimes delivered at high dose rate, in which interfraction interval is deliberately varied, are needed to find out whether the parameters derived from this project are applicable to fractionated treatment courses at high dose rate.

Carcinoma, Squamous Cell↗

PubMed1995

Factors influencing the degree of erythematous skin reactions in humans.

Dose-response relationships have been studied using an ordinal visual scale and reflectance spectrophotometry data from 123 treatment sites on 110 patients treated with 10 dose fractions over 12-14 days. Dose rates varied between 3 and 240 Gy/h and total doses of between 25 and 41 Gy were given using teletherapy apparatus. We found qualitative scoring of erythematous skin reactions to be subject to considerable inter- and intra-observer variation. Reflectance spectrophotometry provided more reproducible information, some of which was undetectable by naked eye. Baseline erythema readings were significantly higher in male patients and at anatomical sites of previous heavy UV exposure. In addition, a pronounced decline in erythema readings during the second week of therapy and 'reciprocal vicinity' (abscopal) effects adjacent to the field, undetected by the eye, were observed in a subset of patients. Meaningful dose-response relationships could be derived only from reflectance data with peak change from the pretreatment baseline measure providing the best discrimination. Peak erythema measures following treatment were found to depend on the age and gender of the patient as well as the treatment site and its baseline erythema measurement. This was independent of the total dose administered or the instantaneous dose rate at which it was delivered. The rate of erythema development was also dose rate dependent but only weakly dependent on the biological dose intensity (Gy equiv./day) of the treatment course. The data raise the question of whether irradiation-induced erythema is exclusively a secondary phenomenon occurring as a result of basal cell killing. The short repair half time value of 0.06 h obtained by direct analysis is perplexing and may reflect a dose rate-dependent physiological vasodilatory response to irradiation and/or a multi-component cellular repair process.

Aged↗

PubMed1994

Iterative generalized least squares for meta-analysis of survival data at multiple times.

A method is presented for joint analysis of survival proportions reported at multiple times in published studies to be combined in a meta-analysis. Generalized least squares is used to fit linear models including between-trial and within-trial covariates, using current fitted values iteratively to derive correlations between times within studies. Multi-arm studies and nonrandomized historical controls can be included with no special handling. The method is illustrated with data from two previously published meta-analyses. In one, an early treatment difference is detected that was not apparent in the original analysis.

Antineoplastic Agents↗

PubMed1994

Jackknife estimators of variance for parameter estimates from estimating equations with applications to clustered survival data.

An estimate of a parameter vector beta is often obtained by setting a "score" vector equal to zero and solving for beta. Estimating equations of this type include maximum likelihood, quasi-likelihood (McCullagh, 1983, Annals of Statistics 11, 59-67), and generalized estimating equations (Liang and Zeger, 1986, Biometrika 73, 13-22). White (1982, Econometrica 50, 1-26) proposed a variance estimator for beta that is robust to model misspecification. We show that a "one-step" jackknife estimator of variance is asymptotically equivalent to the variance estimator proposed by White. The one-step variance estimator may be preferred when the appropriate computer packages are not available to compute White's estimator directly. This jackknife estimator is very useful in our example with clustered survival data.

Analysis of Variance↗

PubMed1993

Meta-analysis of radiation therapy with and without adjuvant chemotherapy for malignant gliomas in adults.

BACKGROUND: The value of chemotherapy after standard postoperative external beam radiation in the treatment of malignant gliomas remains controversial. Despite recent recommendations from the Brain Tumor Cooperative Group that chemotherapy should be considered part of the standard treatment of patients with high-grade astrocytomas, several recent trials have questioned the efficacy of this approach. METHODS: Using results from 16 randomized clinical trials involving more than 3000 patients, the authors compared the survival rates of patients who received radiation alone or radiation with chemotherapy. The combined data were analyzed using the statistical method of meta-analysis as described by DerSimonian and Laird. RESULTS: The estimated increase in survival for patients treated with combination radiation and chemotherapy was 10.1% at 1 year (95% confidence interval, 6.8, 13.3%) and 8.6% at 2 years (5.2, 12.0%). These absolute increases in survival (treated-control [TC]) in patients treated with chemotherapy represent relative increases (T-C)/C of 23.4% at 1 year (15.8, 30.9%) and 52.4% at 2 years (31.7, 73.2%). This survival advantage is conferred by several different chemotherapeutic agents. When the prognostic variables of age and histology are factored into the analysis, however, the data suggest that the survival benefit from chemotherapy occurs earlier in patients with anaplastic astrocytoma (AA) than in patients with glioblastoma. CONCLUSIONS: The authors concluded that chemotherapy is advantageous for patients with malignant gliomas and should be considered part of the standard therapeutic regimen. Additional randomized trials using optimal radiation and chemotherapy may still be needed to precisely define which subgroups of patients, based on prognostic variables, will benefit most from chemotherapy after radiation.

Adult↗

PubMed1993

Reporting of age data in clinical trials of arthritis. Deficiencies and solutions.

OBJECTIVE: We studied recent nonsteroidal anti-inflammatory drug (NSAID) randomized control trials of arthritis to identify the age and number of older people (> or = 65 years) and to document the way information on age was presented. We hypothesized that older people, who are most likely to take NSAIDs are underrepresented and underreported. STUDY SELECTION AND DATA EXTRACTION: All NSAID articles (n = 1008) in MEDLINE between September 1987 and May 1990 were identified. Eighty-three trials employing NSAIDs in a randomized control trial of arthritis reported in 73 articles were identified and studied in detail for age-related information. RESULTS: A total of 9664 subjects with a female-to-male ratio of 2.3:1 were enrolled. Forty-four trials studied osteoarthritis (53.0%), 37 studied rheumatoid arthritis (44.6%), and two studied both conditions (2.4%). More than half of the studies reviewed included people 65 years of age or older, only 207 people in this older age group could be identified (2.1%). While there was inclusion of the 'young-old' (65 to 74 years of age), only 14 of the 9664 people studied were between 75 and 84 years of age, and no one 85 years or older could be identified. The inclusion of the young-old is documented by the weighted mean age that ranged from 59.6 to 64.9 years for patients with osteoarthritis (mean, 62.9; SD, 1.67) and from 47.4 to 53.0 years (mean, 49.9; SD, 2.16) for those with rheumatoid arthritis. CONCLUSION: We demonstrate that older people, who represent a high proportion of the population treated with NSAIDs in practice, are generally omitted from drug trials. Recommendations designed to improve the reporting of age information to make clinical trials more informative and applicable to older people are presented.

Aged↗