Biomedical subjects

M G Kenward

Publications and source records attributed to M G Kenward.

17 recordsLinked to original sources

PubMed2000

Obesity, increased linear growth, and risk of type 1 diabetes in children.

OBJECTIVE: The purpose of the present study was to evaluate the effect of obesity and linear growth on the risk of developing type 1 diabetes in children. RESEARCH DESIGN AND METHODS: The study population consists of all diabetic children <15 years of age diagnosed from September 1986 to April 1989 in Finland and their birth date- and sex-matched population-based control subjects. Growth data were obtained from well-baby clinics and school health care units for 586 diabetic and 571 control subjects, resulting in a total of 18,823 paired weight-height observations. RESULTS: Both boys and girls who developed type 1 diabetes were heavier and taller throughout childhood than control children. A 10% unit increment in relative weight was associated with a 50-60% increase in the risk of type 1 diabetes before 3 years of age and a 20-40% increase from 3 to 10 years of age. The increase in risk of type 1 diabetes for 1 SD score increment in relative height was 20-30%. Obesity (relative weight > 120%) after 3 years of age was associated with a more than twofold risk of developing type 1 diabetes. CONCLUSIONS: The present observation that obesity and rapid linear growth are risk factors for type 1 diabetes in children indicates that the increase in the prevalence of obesity and secular growth that has occurred in most industrialized countries over the last decades may be involved in the increase in type 1 diabetes incidence simultaneously observed in many countries.

Aging↗

PubMed1999

Parametric models for incomplete continuous and categorical longitudinal data.

This paper reviews models for incomplete continuous and categorical longitudinal data. In terms of Rubin's classification of missing value processes we are specifically concerned with the problem of nonrandom missingness. A distinction is drawn between the classes of selection and pattern-mixture models and, using several examples, these approaches are compared and contrasted. The central roles of identifiability and sensitivity are emphasized throughout.

Data Interpretation, Statistical↗

PubMed1999

Infant feeding, early weight gain, and risk of type 1 diabetes. Childhood Diabetes in Finland (DiMe) Study Group.

OBJECTIVE: To evaluate whether the increased risk of type 1 diabetes conferred by an early introduction of cow's milk supplements can be mediated by accelerated growth in formula-fed infants. RESEARCH DESIGN AND METHODS: All children < or = 14 years of age who were diagnosed with type 1 diabetes from September 1986 to April 1989 were invited to participate in the study. Birth date- and sex-matched control children were randomly selected from the Finnish Population Registry. At least three weight measurements from the first year of life were obtained for 435 full-term diabetic subjects and 386 control subjects from well-baby clinics and school health care units. RESULTS: Increase in body weight was greater in the diabetic girls than in the control girls, and the difference increased from 111 g (95% CI 0-218, P = 0.04) at 1 month of age to 286 g (95% CI 123-450, P = 0.0006) at 7 months. For boys, the difference in weight between the diabetic subjects and the control subjects remained stable during infancy (difference 95 g, 95% CI-2-205, P = 0.09). Increased weight was associated on average with a 1.5-fold risk of type 1 diabetes. Early introduction of formula feeding (< 3 vs. > or = 3 months) was also associated with an increased risk of type 1 diabetes after adjustment for the individual weight gain curve (adjusted odds ratio 1.53, 95% CI 1.1-2.2). No evidence for interaction was observed. CONCLUSIONS: These observations indicate that an early exposure to cow's milk formula-feeding and rapid growth in infancy are independent risk factors of childhood type 1 diabetes.

Adolescent↗

PubMed1999

Analysis of incomplete public health data.

The problem of dealing with missing values is common throughout statistics and is very prominent with epidemiologic data in the broad sense. Not only do data collection procedures break down, but subjects may be lost to follow up, or simply withdraw their consent without further providing a reason for doing so. In this paper, we review a framework for handling incomplete studies, and then concentrate on a specific case. It comes from a complex health interview survey, conducted in Belgium in 1997, where different types of missingness arise at various levels of the hierarchical sampling procedure.

Belgium↗

PubMed1998

Selection models for repeated measurements with non-random dropout: an illustration of sensitivity.

The outcome-based selection model of Diggle and Kenward for repeated measurements with non-random dropout is applied to a very simple example concerning the occurrence of mastitis in dairy cows, in which the occurrence of mastitis can be modelled as a dropout process. It is shown through sensitivity analysis how the conclusions concerning the dropout mechanism depend crucially on untestable distributional assumptions. This example is exceptional in that from a simple plot of the data two outlying observations can be identified that are the source of the apparent evidence for non-random dropout and also provide an explanation of the behaviour of the sensitivity analysis. It is concluded that a plausible model for the data does not require the assumption of non-random dropout.

Animals↗

PubMed1998

Achilles tendon rupture and sciatica: a possible correlation.

The association between Achilles tendon rupture and sciatica was investigated by questionnaire in 138 patients who underwent repair of an Achilles tendon rupture, and in a group of individuals nominated by the patients, matched for age, sex, and occupation. A total of 102 patients (74%) and 128 peer nominated controls (71%) replied to the questionnaire. Of the 102 respondent patients, 18 had an officebased job, 47 were involved in skilled nonmanual work, and 16 were retired. Back pain had been experienced by 63 of the patients who replied to the questionnaire, and by 91 (75%) of the individuals in the control group (difference not significant). In about 30% of both groups, the pain confined them to bed for at least two days, and resulted in absence from work. Thirteen of the patients and 16 of the controls had undergone thoracic, lumbar, or sacral radiography. One individual in each group had received surgery for back pain. However, 35 of 102 patients had experienced sciatic pain before Achilles tendon rupture. Pain of a similar nature had been experienced by only 15 individuals in the control group (12%) (p < 0.001). Using this study design, we found a highly significant association between Achilles tendon rupture and sciatica. We propose that this association could be due to impaired afferent signals from the lower leg, or to similar collagen or vascular anomalies of the vertebral disc and the Achilles tendon.

Achilles Tendon↗

PubMed1997

Assessment of late results of surgery in talipes equino-varus: a reliability study.

UNLABELLED: The variability of a method for clinical and functional assessment of the long-term results of surgical correction of idiopathic congenital talipes equino varus was studied in ten boys and four girls (average age: 19.1 (SD 2.3 years); 22 affected feet) with radiographical evidence of fusion of the foot and ankle ossification centres. Patients were measured twice, 1 week-1 month apart, by the same investigator and were assessed twice on each visit. Assessment included anthropometry, functional assessment and subjective functional evaluation. Calf circumference, skinfold thickness, foot length and width were highly reproducible. Foot length was not significantly influenced by the operation whereas calf circumference, skinfold thickness and foot width were. Although highly reproducible, hopping was not significantly affected by operation. Active and passive range of motion were significantly different. Each was highly reproducible and both were significantly affected by the operation. Patients reported a high and highly reproducible (within two points) functional level. CONCLUSION: An assistant-administered functional questionnaire together with measurement of active and passive range of motion allows easy, valid and reproducible assessment of long-term results of surgery for idiopathic congenital talipes equino-varus correction. A significant effect of the number and type of operation was evidenced.

Adolescent↗

PubMed1997

Small sample inference for fixed effects from restricted maximum likelihood.

Restricted maximum likelihood (REML) is now well established as a method for estimating the parameters of the general Gaussian linear model with a structured covariance matrix, in particular for mixed linear models. Conventionally, estimates of precision and inference for fixed effects are based on their asymptotic distribution, which is known to be inadequate for some small-sample problems. In this paper, we present a scaled Wald statistic, together with an F approximation to its sampling distribution, that is shown to perform well in a range of small sample settings. The statistic uses an adjusted estimator of the covariance matrix that has reduced small sample bias. This approach has the advantage that it reproduces both the statistics and F distributions in those settings where the latter is exact, namely for Hotelling T2 type statistics and for analysis of variance F-ratios. The performance of the modified statistics is assessed through simulation studies of four different REML analyses and the methods are illustrated using three examples.

Analysis of Variance↗

PubMed1994

The analysis of binary and categorical data from crossover trials.

A review is presented of methods for the analysis of discrete data from crossover trials. The definition and interpretation of the model for the data is used as a central theme. Distinctions are drawn between different types of model, particularly marginal and subject specific. It is seen how much recent methodology for analysing correlated categorical data can be applied successfully to the crossover setting. The current accessibility of each method is considered and the different approaches are illustrated and compared using two examples.

Bias↗

PubMed1994

An application of maximum likelihood and generalized estimating equations to the analysis of ordinal data from a longitudinal study with cases missing at random.

Data are analysed from a longitudinal psychiatric study in which there are no dropouts that do not occur completely at random. A marginal proportional odds model is fitted that relates the response (severity of side effects) to various covariates. Two methods of estimation are used: generalized estimating equations (GEE) and maximum likelihood (ML). Both the complete set of data and the data from only those subjects completing the study are analysed. For the completers-only data, the GEE and ML analyses produce very similar results. These results differ considerably from those obtained from the analyses of the full data set. There are also marked differences between the results obtained from the GEE and ML analysis of the full data set. The occurrence of such differences is consistent with the presence of a non-completely-random dropout process and it can be concluded in this example that both the analyses of the completers only and the GEE analysis of the full data set produce misleading conclusions about the relationships between the response and covariates.

Age Factors↗

PubMed1993

Effect of the prenatal maternal environment on the control of breathing during non-rapid-eye-movement sleep in the developing lamb.

This study examines the effects of altering the prenatal maternal metabolic and hormonal environment via chronic cold exposure of under-fed ewes on developmental changes in breathing control of developing lambs. Breathing frequency and timing were measured during non rapid-eye-movement (non-REM) sleep in lambs born from either shorn or unshorn ewes after being maintained for at least one hour at warm (28-19 degrees C) and cool (14-5 degrees C) ambient temperatures at 1, 4, 14 and 30 days of age. Breathing frequency and oxygen consumption were significantly higher in 1 day old lambs born from shorn ewes compared with those lambs born from unshorn ewes, at both warm and cool ambient temperatures. In the shorn group breathing frequency decreased between 1 and 4 days of age and continued decreasing upto 30 days of age, during which period inspiratory and to a greater extent expiratory time, lengthened. Laryngeal "braking" of expiratory airflow was observed in more than 50% of lambs born from shorn ewes during non-REM sleep in the warm at 4, 14 and 30 days of age, and in the cold at 14 and 30 days of age. In contrast, lambs born from unshorn ewes showed no change in breathing frequency between 1 and 4 days of age, but a decrease was observed between 4 and 14 days of age, whilst laryngeal "braking" of expiratory airflow was rarely observed at any age.(ABSTRACT TRUNCATED AT 250 WORDS)

Animals↗

PubMed1992

Alternative approaches to the analysis of binary and categorical repeated measurements.

The purpose of this paper is to describe, illustrate, and compare a number of different approaches to the analysis of repeated binary and categorical data. These approaches include empirical generalized least squares and generalized estimating equations, as well as traditional log-linear modeling methods. It is shown that the interpretation of the parameters in the various models depends critically on the type of model fitted. In particular, we contrast the population-averaged and subject-specific models. Two example data sets are used to illustrate the approaches, and throughout we concentrate on methods that can be easily implemented.

Clinical Trials as Topic↗

PubMed1991

The analysis of categorical data from cross-over trials using a latent variable model.

A latent variable model for categorical data is described. The nuisance parameters in the model can be eliminated from the analysis through the use of a conditional likelihood and this leads to an analysis based on a log-linear model. It is shown how the structure of a cross-over trial allows considerable simplification in the formulation of this model and routine application of well-known statistical packages. The method is illustrated by data from a three-period cross-over trial on the relief of primary dysmenorrhea using the GLIM and SAS packages.

Analysis of Variance↗

PubMed1987

The analysis of data from 2 x 2 cross-over trials with baseline measurements.

An account is given of the analysis of data from 2 x 2 cross-over trials which include baseline measurements. We show that most of the previously proposed methods can be incorporated into a general framework of least-squares estimation with a simple linear model. A simple analysis based on ordinary least-squares estimators is described which can be used with either two-sample t-tests and confidence intervals or with the corresponding non-parametric procedures. It is shown how the use of generalized least-squares estimators is equivalent to the use of covariance adjustment. These methods require no assumptions about the covariance structure of the measurements from each subject. The results of assessing the covariance structure present in examples of data from a number of trials are summarized. These results suggest that previously proposed simple covariance structures are unlikely to be appropriate in general.

Analysis of Variance↗

PubMed1985

The use of fitted higher-order polynomial coefficients as covariates in the analysis of growth curves.

For orthogonal polynomials fitted to repeated measurements, a computer simulation study is used to investigate the effect of selecting higher-order coefficients as covariates in order to minimise the estimated variance of lower-order coefficients. Under the assumption that the repeated measurements have certain autoregressive covariance structures, it is seen that the gain in precision due to covariance adjustment can be largely illusory and that the resulting estimates of variances can grossly underestimate the true variances. The consequences of using all higher-order coefficients as covariates is also examined and seen to produce a gain in precision in certain circumstances.

Aging↗

PubMedDate not supplied

Missing data perspectives of the fluvoxamine data set: a review.

Fitting models to incomplete categorical data requires more care than fitting models to the complete data counterparts, not only in the setting of missing data that are non-randomly missing, but even in the familiar missing at random setting. Various aspects of this point of view have been considered in the literature. We review it using data from a multi-centre trial on the relief of psychiatric symptoms. First, it is shown how the usual expected information matrix (referred to as naive information) is biased even under a missing at random mechanism. Second, issues that arise under non-random missingness assumptions are illustrated. It is argued that at least some of these problems can be avoided using contextual information.

Data Interpretation, Statistical↗

PubMedDate not supplied

Modelling binary data from a three-period cross-over trial.

A new method of analysing binary data from a three-treatment, three-period cross-over trial is described. This method is based on a log-linear model and mirrors the analysis of continuous data. It is an extension of the method we introduced recently for the analysis of binary data from a two-treatment, two-period cross-over trial. We illustrate our method using data from a trial which compared two analgesics and a placebo for the relief of primary dysmenorrhea.

Analgesics↗