PubMed Health⌕ Search

Biomedical subjects

James A Hanley

Publications and source records attributed to James A Hanley.

At least 19 recordsLinked to original sources
PubMed2006

Do OSCAR winners live longer than less successful peers? A reanalysis of the evidence.

In an article published in Annals of Internal Medicine in 2001, Redelmeier and Singh reported that Academy Award-winning actors and actresses lived almost 4 years longer than their less successful peers. However, the statistical method used to derive this statistically significant difference gave winners an unfair advantage because it credited an Oscar winner's years of life before winning toward survival subsequent to winning. When the authors of the current article reanalyzed the data using methods that avoided this "immortal time" bias, the survival advantage was closer to 1 year and was not statistically significant. The type of bias in Redelmeier and Singh's study is not limited to longevity comparisons of persons who reach different ranks within their profession; it can, and often does, occur in nonexperimental studies of life- or time-extending benefits of medical interventions. The current authors suggest ways in which researchers and readers may avoid and recognize this bias.

Achievement↗
PubMed2006

The effect of omitted covariates on confidence interval and study power in binary outcome analysis: a simulation study.

BACKGROUND/OBJECTIVES: The consequence of omitted but balanced covariates on odds ratio point estimation is well-known in the literature. When exposure or intervention has a non-null effect on disease outcome, omitted covariates lead to underestimation of the effect of exposure or intervention. However, the effect of omitted covariates on confidence interval and study power is unknown. STUDY DESIGN AND SETTING: A simulation study is carried out to assess the effect of omitted covariates on confidence interval and study power for a plausible range of scenarios. Coverage probability and study power are assessed systematically over a range of study size, type of omitted covariate and magnitude of effect. A real-life example using a randomised experiment on flies' sexuality is provided. RESULTS: When a balanced covariate is omitted, coverage probability was lowered by 2.9-80%. Likewise study power was reduced by as much as 58%. The impact becomes substantial when the covariate is continuous, has large variability and has a larger effect than the effect of exposure or intervention. The result from a real-life example concurs with the simulation finding. CONCLUSION: Omitting an important balanced covariate lowers both coverage probability and study power. This implies the need for thoughtful consideration of important covariates at the design as well as the analysis stages of a study.

Animals↗
PubMed2006

Creating non-parametric bootstrap samples using Poisson frequencies.

This article describes how, in the high-level software packages used by non-statisticians, approximate non-parametric bootstrap samples can be created and analyzed without physically creating new data sets, or resorting to complex programming. The comparable performance of this shortcut method, which uses Poisson rather than multinomial frequencies for the numbers of copies of each observation, is demonstrated theoretically by evaluating the bootstrap variance in an example where the classic estimator of the sampling variance of the statistic of interest has a known closed form. For sample sizes of 50 or more, bootstrap standard errors obtained by this shortcut method exceeded those obtained by the standard version by less than 1%. The proposed method is also evaluated in two worked examples, involving statistics whose sampling distribution is more complex. The second of these is also used to illustrate when one can and cannot use non-parametric bootstrap samples.

Computational Biology↗
PubMed2006

Identification in administrative databases of women dying of breast cancer.

PURPOSE: Palliative care is an essential component of cancer care, and population-based research is needed to monitor its impact. Administrative databases are the cornerstone of health services research. Their limitation is that cause of death is not sufficient to readily classify decedents as terminally ill for the study of the health services they received at the end of life. The study purpose is to develop and test the validity of an algorithm allowing the classification of the decedents as dying of breast cancer (BC), using administrative data. METHODS: Validation was carried out through a chart review of 119 BC decedents extracted from hospital-based databases. This algorithm was applied to 3,384 deceased women with BC representative of the whole population. The effect of the classification by the algorithm was illustrated by the shift in the distributions of age and place of death. RESULTS: The validation showed a sensitivity of 95%, a specificity of 89%, a positive predictive value of 98%, and negative predictive value of 77% for the classification of women dying of BC. Of the 3,384 decedents, 2,293 were classified as dying of, and 1,091 as not dying of BC. Women dying of BC were younger, died less often at home (6.9% v 17.9%), and in chronic care institutions (4.1% v 14.8%), and more often in acute-care beds (69.9% v 57.1%). CONCLUSION: This novel way to classify decedents is conceptually based and empirically validated through chart review and impact on distribution of age and place of death.

Adult↗
PubMed2006

An 'unconditional-like' structure for the conditional estimator of odds ratio from 2 x 2 tables.

In the estimation of the odds ratio (OR), the conditional maximum-likelihood estimate (cMLE) is preferred to the more readily computed unconditional one (uMLE). However, the exact cMLE does not have a closed form to help divine it from the uMLE or to understand in what circumstances the difference between the two is appreciable. Here, the cMLE is shown to have the same 'ratio of cross-products' structure as its unconditional counterpart, but with two of the cell frequencies augmented, so as to shrink the unconditional estimator towards unity. The augmentation involves a factor, similar to the finite population correction, derived from the minimum of the marginal totals.

Algorithms↗
PubMed2006

Balance self-efficacy and its relevance to physical function and perceived health status after stroke.

OBJECTIVES: To estimate the level of balance self-efficacy among community-dwelling subjects with stroke and to determine the relative importance of balance self-efficacy compared with functional walking capacity in predicting physical function and perceived health status. DESIGN: Secondary analysis of baseline, postintervention, and 6-month follow-up data from a randomized trial. SETTING: General community. PARTICIPANTS: Ninety-one subjects with a first or recurrent stroke, discharged from rehabilitation therapy with a residual walking deficit. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: The Activities-Specific Balance Confidence (ABC) Scale, Medical Outcomes Study 36-Item Short-Form Health Survey physical function scale, and the EQ-5D visual analog scale of perceived health status. RESULTS: Average balance self-efficacy was 59 out of 100 points on the ABC scale (95% confidence interval, 55-64; n=89). After adjusting for age and sex, functional walking capacity explained 32% and 0% of the respective variability in physical function and perceived health status scores obtained 6 months later. After adjustment for age, sex, and functional walking capacity, balance self-efficacy explained 3% and 19% of variation in 6-month physical function and perceived health status scores, respectively. CONCLUSIONS: Subjects living in the community after stroke experience impaired balance self-efficacy. Enhancing balance self-efficacy in addition to functional walking capacity may lead to greater improvement, primarily in perceived health status, but also in physical function, than the enhancement of functional walking capacity alone.

Activities of Daily Living↗
PubMed2006

Psychometric evaluation of the original and Canadian French version of the activities-specific balance confidence scale among people with stroke.

OBJECTIVE: To evaluate the internal and absolute reliability and construct validity of the Activities-Specific Balance Confidence (ABC) scale and a new Canadian French version (ABC-CF) of it among people with stroke. DESIGN: Cross-sectional data from a randomized controlled trial. SETTING: Community. PARTICIPANTS: Ninety-one people with a residual walking deficit between 57 and 386 days poststroke. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: The ABC and ABC-CF scales, Berg Balance Scale (BBS), comfortable and maximum gait speeds, Timed Up & Go (TUG) test, 6-minute walk test (6MWT), Barthel Index, physical function scale of the Medical Outcomes Study 36-Item Short-Form Health Survey, Geriatric Depression Scale (GDS), and the EQ-5D visual analog scale (EQ VAS). RESULTS: Internal consistency (Cronbach alpha) was .94 and .93 and the standard error of measurement was 5.05 and 5.13 for the ABC (n=51) and the ABC-CF (n=35) scales, respectively. Spearman rho values ranged from .30 to .60 for the ABC scale and from .45 to .68 on the ABC-CF scale for associations with scores on the BBS, comfortable and maximum gait speeds, TUG, 6MWT, Barthel Index, physical function scale, GDS, and EQ VAS. CONCLUSIONS: Evidence of internal and absolute reliability and of construct validity of the ABC and the ABC-CF scales supports their use for cross-sectional measurements of balance self-efficacy among community-dwelling people in the first year poststroke.

Adult↗
PubMed2006

Statistical practice in high-throughput screening data analysis.

High-throughput screening is an early critical step in drug discovery. Its aim is to screen a large number of diverse chemical compounds to identify candidate 'hits' rapidly and accurately. Few statistical tools are currently available, however, to detect quality hits with a high degree of confidence. We examine statistical aspects of data preprocessing and hit identification for primary screens. We focus on concerns related to positional effects of wells within plates, choice of hit threshold and the importance of minimizing false-positive and false-negative rates. We argue that replicate measurements are needed to verify assumptions of current methods and to suggest data analysis strategies when assumptions are not met. The integration of replicates with robust statistical methods in primary screens will facilitate the discovery of reliable hits, ultimately improving the sensitivity and specificity of the screening process.

Biological Assay↗
PubMed2006

Ranking hospitals according to acute myocardial infarction mortality: should transfers be included?

OBJECTIVE: The objective of this population-based observational cohort study was to estimate the extent to which the inclusion/exclusion of transferred patients with acute myocardial infarction (AMI) impacts on hospital performance rankings. SUBJECTS: The authors studied 91,633 adult patients admitted to 116 acute care hospitals in Quebec, Canada, with a primary diagnosis of AMI between 1992 and 1999. MAIN OUTCOME MEASURE: Hospital performance ranks, based on 30-day AMI mortality rates, were estimated with hierarchical models and compared using 3 different methods for handling transferred patients (exclude all transfers; include transfers and assign outcome to the referring hospital; include transfers and assign outcome to the receiving hospital). The explanatory variable of interest was the hospital to which the patient's outcome was attributed. RESULTS: Using the 3 methods, 4 hospitals were ranked "best performers" once, and 1 hospital ranked among the best in 2 of the 3 analyses performed. Nine hospitals were ranked "worst performers" at least once (4 of which ranked among the "worst" once only, 2 ranked among the "worst" twice, and 3 were consistently ranked "worst performers" in all analyses). There was significant variation in mortality rates among hospitals, and the difference in the rates between the highest and lowest ranking hospitals exceeded the clinically relevant benchmark of 1%. CONCLUSIONS: Performance evaluation studies that compare hospital mortality rates typically exclude transferred patients. However, methods used to deal with AMI patient transfers influenced hospital ranks when comparing 30-day mortality rates. Excluding transfers may lead to an inaccurate depiction of the quality of healthcare services in regionalized healthcare systems that call for the timely interhospital transfer of patients with AMI.

Aged↗
PubMed2005

Two-stage case-control studies: precision of parameter estimates and considerations in selecting sample size.

A two-stage case-control design, in which exposure and outcome are determined for a large sample but covariates are measured on only a subsample, may be much less expensive than a one-stage design of comparable power. However, the methods available to plan the sizes of the stage 1 and stage 2 samples, or to project the precision/power provided by a given configuration, are limited to the case of a binary exposure and a single binary confounder. The authors propose a rearrangement of the components in the variance of the estimator of the log-odds ratio. This formulation makes it possible to plan sample sizes/precision by including variance inflation factors to deal with several confounding factors. A practical variance bound is derived for two-stage case-control studies, where confounding variables are binary, while an empirical investigation is used to anticipate the additional sample size requirements when these variables are quantitative. Two methods are suggested for sample size planning based on a quantitative, rather than binary, exposure.

Case-Control Studies↗
PubMed2005

Prostate cancer and the Will Rogers phenomenon.

BACKGROUND: Information on tumor stage and grade are used to assess cancer prognosis and to produce standardized comparisons of end results over time. Changes in the interpretation of classification schemes can alter the apparent distribution of cancer stage or grade in the absence of a true biologic change. Since the introduction of prostate-specific antigen testing, the reported incidence of low-grade prostate cancer has declined. To determine whether this decline is in part a result of Gleason score reclassification during the same time period, we documented the potential impact of reclassification between 1992 and 2002 on clinical outcomes. METHODS: A population-based cohort of 1858 men who were < or = 75 years of age at diagnosis of prostate cancer in 1990-1992 was assembled retrospectively from the Connecticut Tumor Registry. Histology slides of the diagnostic prostate tissue were retrieved and reread in 2002-2004 by an experienced pathologist blinded to the original Gleason score readings. Prostate cancer mortality rates for the cohort calculated using the original Gleason score readings were compared with those calculated using the contemporary Gleason score readings. Statistical tests were two sided. RESULTS: The contemporary Gleason score readings were statistically significantly higher than the original readings (mean score increased from 5.95 to 6.8; difference = 0.85, 95% confidence interval = 0.79 to 0.91; P < .001). Consequently, the Gleason score-standardized contemporary prostate cancer mortality rate (1.50 deaths per 100 person-years) appeared to be 28% lower than standardized historical rates (2.08 deaths per 100 person-years), even though the overall outcome was unchanged. This apparent improvement in mortality held for all Gleason score categories. CONCLUSIONS: In this population, a decline in the reported incidence of low-grade prostate cancers appears to be the result of Gleason score reclassification over the past decade. This reclassification resulted in apparent improvement in clinical outcomes. This finding reflects a statistical artifact known as the Will Rogers phenomenon.

Aged↗
PubMed2005

The structural equation modeling technique did not show a response shift, contrary to the results of the then test and the individualized approaches.

BACKGROUND AND OBJECTIVE: Persons experiencing changes in their physical health may change their values and rerate the importance of basic elements of health-related quality of life (HRQL), a process known as response shift. Developing an estimator of HRQL that differentiates between objective change and response shift is essential for the interpretation of the results. The purpose of the present article was to contrast three methodologic approaches for evaluating response shift to develop a proposed set of HRQL measurement recommendations under circumstances where response shift is expected to occur. METHODS: The three approaches compared were a structural equation modeling (SEM) technique, the then test, and an individualized approach. The data collection procedures for these methods were incorporated into a poststroke randomized controlled trial. RESULTS: The SEM did not show a response shift, contrary to the results of the then test and the individualized approaches. We discuss factors that effect the selection of a methodologic approach including feasibility, subjects' memory and more advanced cognitive tasks, and whether response shift was evaluated at the group or individual level. CONCLUSION: The evaluation of response shift is an integral part of HRQL evaluations, and further comparisons between methodologic approaches are needed.

Activities of Daily Living↗
PubMed2005

Distribution of LDL particle size in a population-based sample of children and adolescents and relationship with other cardiovascular risk factors.

BACKGROUND: Smaller, denser LDL particles are associated with an increased risk for cardiovascular diseases (CVD). In youths, data on the distribution of LDL particle size and on its association with other CVD risk factors are limited. METHODS: We determined LDL peak particle size by nondenaturing 2%-16% gradient gel electrophoresis in a representative sample of 2249 youths 9, 13, and 16 years of age who participated in a school-based survey conducted in 1999 in the province of Quebec, Canada. Standardized clinical measurements and fasting plasma lipid, glucose, and insulin concentrations were available. RESULTS: The LDL peak particle size distribution was gaussian. The 5th, 50th (median), and 95th percentiles by age and sex were 255.5-258.6, 262.1-263.2, and 268.1-269.5 A, respectively. The prevalence of the small, dense LDL phenotype (LDL peak particle size <or=255 A) was 10% in participants with insulin resistance syndrome (IRS), in contrast to 1% in those without IRS. In a multiple regression analysis, the association of LDL size with other CVD risk factors [apolipoprotein B, HDL-cholesterol (HDL-C), triglyceride (TG), and insulin concentrations, and body mass index] was strongest with TG and HDL-C concentrations: a 1 SD increase in log(e)-transformed TG concentration was associated with a 1.2 A reduction in LDL size, and a 1 SD increase in HDL-C was associated with a 1.1 A increase in LDL size. CONCLUSIONS: Although the small, dense LDL phenotype is less prevalent in youths than adults, its prevalence is clearly increased in childhood IRS. Metabolic correlates of LDL size are similar in youths and adults.

Adolescent↗
PubMed2005

20-year outcomes following conservative management of clinically localized prostate cancer.

CONTEXT: The appropriate therapy for men with clinically localized prostate cancer is uncertain. A recent study suggested an increasing prostate cancer mortality rate for men who are alive more than 15 years following diagnosis. OBJECTIVE: To estimate 20-year survival based on a competing risk analysis of men who were diagnosed with clinically localized prostate cancer and treated with observation or androgen withdrawal therapy alone, stratified by age at diagnosis and histological findings. DESIGN, SETTING, AND PATIENTS: A retrospective population-based cohort study using Connecticut Tumor Registry data supplemented by hospital record and histology review of 767 men aged 55 to 74 years with clinically localized prostate cancer diagnosed between January 1, 1971, and December 31, 1984. Patients were treated with either observation or immediate or delayed androgen withdrawal therapy, with a median observation of 24 years. MAIN OUTCOME MEASURES: Probability of mortality from prostate cancer or other competing medical conditions, given a patient's age at diagnosis and tumor grade. RESULTS: The prostate cancer mortality rate was 33 per 1000 person-years during the first 15 years of follow-up (95% confidence interval [CI], 28-38) and 18 per 1000 person-years after 15 years of follow-up (95% CI, 10-29). The mortality rates for these 2 follow-up periods were not statistically different, after adjusting for differences in tumor histology (rate ratio, 1.1; 95% CI, 0.6-1.9). Men with low-grade prostate cancers have a minimal risk of dying from prostate cancer during 20 years of follow-up (Gleason score of 2-4, 6 deaths per 1000 person-years; 95% CI, 2-11). Men with high-grade prostate cancers have a high probability of dying from prostate cancer within 10 years of diagnosis (Gleason score of 8-10, 121 deaths per 1000 person-years; 95% CI, 90-156). Men with Gleason score of 5 or 6 tumors have an intermediate risk of prostate cancer death. CONCLUSION: The annual mortality rate from prostate cancer appears to remain stable after 15 years from diagnosis, which does not support aggressive treatment for localized low-grade prostate cancer.

Aged↗
PubMed2005

Using the Patient Generated Index to evaluate response shift post-stroke.

BACKGROUND: Individualized measures allow for the assessment of health-related quality of life (HRQL) based on areas that are relevant to the person and may prove to be useful for evaluating response shift (reconceptualization of HRQL and change in values). OBJECTIVE: The objective of this study was to assess reconceptualization of HRQL and change of individual values among persons with stroke during the first six months of recovery. METHODS: The data collection for this study was incorporated into a randomized trial of acute post-stroke care. Individualized HRQL was evaluated at 6 and 24 weeks post-stroke using the Patient Generated Index (PGI). At 24 weeks a semi-structured interview was administered to assess whether verbalizations given by subjects indicated that they had experienced a response shift. RESULTS: Ninety two subjects (61%) had complete PGI information at the 6- and 24-week evaluations, and of these, 46 completed the semi-structured interview. Between the 6- and 24-week evaluations, the domains selected were: the same for 10 (11%) subjects, reduced for 27 (29%), expanded for 11 (12%), and completely different for 44 (48%) subjects. Twenty eight percent (n = 13) of subjects recovering from stroke experienced a response shift as evidenced by the semi-structured interviews. CONCLUSION: The PGI provides valuable information regarding changes in person's conceptualization of HRQL and values, but the feasibility of using an individualized measure during the first six months post-stroke is limited by the added complexity of completing and interpreting such a measure.

Aged↗
PubMed2005

Discordance between cross-sectional and longitudinal studies for the effect of dust on COPD: why?

Regression analyses for the effect of an environmental agent on lung function often give discordant results when derived from cross-sectional compared with longitudinal studies. To evaluate why this occurs, a normal population was created by computer, and modeled to simulate functional change during life. Thus, factors known to influence lung function measurement (including those that may cause COPD) were manipulated experimentally so that their contributions to any discordance could be assessed. Regression analyses showed that significant discordance could be induced if the oldest birth cohort failed to reach the same maximal level of function as the youngest (a "cohort effect"). This distorted the cross-sectional (but not longitudinal) estimate for the dominating effect of age and additionally influenced cross-sectional estimates for the effects of partially collinear variables such as cumulative exposure to hazardous environmental dust. Discordance also occurred if regression coefficients became imprecise through random measurement/reporting error, between-subject variability, and differing susceptibility, but then the differences (sometimes marked) between cross-sectional and longitudinal estimates were not significant. We conclude that modeling a population with known characteristics can provide a useful means of demonstrating that cross-sectional versus longitudinal discordance may be fundamental and unavoidable (though explicable), or merely a consequence of imprecision.

Aging↗