Search

Objectives: To describe survival of 1226 Scottish adults with amyotrophic lateral sclerosis/motor neurone disease (ALS/MND). Methods: Ten year, prospective, population based disease register. Cox time dependent proportional hazards modelling for multivariate survival analyses. Results: Median survival from onset was 25 months (interquartile range 16–34 months). In multivariate models we found an increased hazard with more recently diagnosed cases—that is, there was an unexpected decline in survival over the 10 year period (hazard ratio (HR) 1.06 (95% CI 1.04 to 1.09). Positive effects on survival were demonstrated for longer time from onset to diagnosis (HR 0.38 (95% CI 0.33 to 0.42), assessment by a neurological specialist (HR 0.56 (95% CI 0.40 to 0.77), and treatment with riluzole (HR 0.24 (95% CI 0.14 to 0.42). Poor prognosis was associated with bulbar onset (HR 1.25 (95% CI 1.09 to 1.46) and a mixed lower and upper motor neurone syndrome (HR 1.23 (95% CI 1.01–1.49) and increasing age. Conclusions: We found an unexpected decline in survival over the 10 year period, despite controlling for potential confounding variables. We would be cautious about overinterpreting these observations and suggest that further research is required to confirm or refute these findings.

In 2000, the UK Departments of Health recommended influenza immunization to employees directly involved in patient care. Uptake of this immunization had tended to be variable and usually low.To assess personal and organizational factors associated with influenza immunization uptake among Health Care Workers (HCWs).A cross-sectional survey of all HCWs within the Health and Social Care Trusts in Northern Ireland and a parallel-group study of nursing staff within Elderly Care using self-administered questionnaires.Of 203 nurses working in elderly care units 76(37%) were immunized and 127(63%) declined. Almost 70% of those not immunized perceived themselves to be 'healthy' and gave this reason for declining immunization. Nurses were more likely to be immunized by a factor of four if they believed there was benefit for healthy HCWs, three if they felt at-risk of contracting influenza and nine on a recommendation from the occupational health (OH) unit. Fifteen OH units participated in a survey of HCWs at the time of immunization. Five thousand two hundred and thirty (9.7%) HCWs were immunized. Increased uptake was correlated with immunization in area of work (r=0.74, P=0.02) and when provided out of hours (r=0.83; P0.001) and by a factor of two with individual targeting of availability (P0.001) and when individuals had been previously immunized (P0.001).Uptake of influenza immunization is low. Attitudes to one's health and to the value of influenza immunization affect the uptake as does the delivery of the immunization programme.

Recent evidence suggests potential associations between birthweight and disease in later life. For resource or other reasons recorded birthweight may be unavailable to researchers who have access to uniquely relevant outcome data. The present study examined the validity of parental recall of birthweight. Parents of 1015 males and females aged 12 and 15 years participating in the Young Hearts Study (a cluster random sample of 1015 males and females aged 12 and 15 years from post-primary schools in Northern Ireland) completed a questionnaire which included a question about their child's birthweight. The answer provided was compared with recorded birthweight obtained from archived computerised child health records with a cut-off point for inaccurate reporting set at +/- 227 g (1/2 lb). The influence of social class and weight at birth on accuracy of recall was also determined. A total of 84.8% of parents accurately recalled their child's birthweight to within 227 g. Parents from non-manual occupation social classes recalled birthweight more accurately than those from manual occupation social classes (88.0 vs. 82.6% accurate: chi2 = 4.81, p = 0.03). Parents of low birthweight infants tended to recall their birthweight less accurately than parents of normal weight infants: 76.1% accurate compared to 86.1% accurate: chi2 = 3.54, p = 0.06. Parents of high birthweight infants recalled their birthweight less accurately than parents of normal weight infants: 78.5% accurate: chi2 = 3.94, p = 0.05. In conclusion, parentally recalled birthweight may be a suitable proxy for recorded birthweight for population based research into disease in childhood and adolescence.

Although Northern Ireland leads the world in terms of coronary heart disease mortality, the prevalence of coronary risk factors in the childhood population is largely unknown. The aims of the present study were to assess the feasibility of collecting this data and to undertake a preliminary investigation of risk factor status. One hundred and three out of 154 first year pupils (67%), mean age 12.3 years from a Belfast secondary school had their height, weight, pubertal status and blood pressure estimated. Blood samples were analysed for serum cholesterol and thiocyanate in 99 children. 31% had elevated cholesterol, 13% had raised blood pressure, 13% were overweight and 5% were identified as smokers. There appeared to be relatively high risk factor levels with at least one risk factor present in 45% of children. No difficulties arose in compliance with the test protocol suggesting that coronary profiling in children on a large-scale is feasible. On the basis of these results, we feel that coronary risk factor status in Northern Irish children needs further evaluation.

A method of estimation of the coefficients of a linear regression model is described as invariant if the basic regression results obtained by the method are unaltered by a location/scale transformation of the data matrix. A necessary and sufficient condition for a method to be invariant is presented. Finally specific methods of estimation are examined for invariance.

On propose une methode basee sur la combinaison de series infinies et de methodes d'integration numerique

An evaluation of the measurement of maternal plasma alpha-fetoprotein (AFP) levels as a screening test for fetal neural tube defects is reported. AFP levels were estimated on 1844 consecutive patients at the time of their 1st hospital antenatal visits. 28 pregnancies had an abnormal outcome: 19 patients had twins 3 had an anencephalic fetus and 6 had a fetus with a spina bifida (4 of these had an open neural tube effect). The screening test at 15 to 19 weeks gestation indicated a sensitivity of 50% taking the cut-off point as the 95th centile of the plasma AFP levels in pregnancy.

The determination of plotting positions on probability graph paper so that the associated weighted least squares estimators of the scale parameter and the percentiles of a continuous distribution have certain properties is discussed. Necessary and sufficient conditions are given for an invariant optimal plot for percentile estimation. Also discussed is the derivation of ordered plotting positions.

In most applications of ordinary ridge regression the usual practice is to estimate the constant term in the model by least squares. This contrasts with the theoretical development of ridge regression in which all regression coefficients are treated in the same way. A variety of results are given which support the use of the response sample mean to estimate the constant term when the predictor data matrix is in correlation form. This conclusion also applies to generalized ridge regression.

The cardiospecific isoenzyme MB of creatine kinase (CKMB) has previously been shown to be of value in the diagnosis of myocardial infarction (MI). We studied 418 admissions to a coronary care unit (CCU) with suspected MI and calculated the sensitivity, specificity and positive and negative predictive values for several CKMB test functions. Several functions performed better than any combination of the other enzymes in common use. 97% of patients achieving a CKMB activity of at least 15 U/l did so between 6 and 30 h following the onset of symptoms. The present study confirms that the use of the CKMB isoenzyme leads to an earlier and more accurate diagnosis or exclusion of MI compared to the "cardiac enzyme series". The timing of blood sampling for CKMB estimation is also discussed.

Standardized radiological pelvimetry was used to examine 242 male and 314 female adults attending the out-patient departments of the Royal Victoria Hospital, Belfast, Seven indices of pelvic size and shape were measured from X-rays on each individual together with social and biological factors including age, height and year of birth. The aims were to quantify any differences in pelvic anatomy between the sexes and to measure correlations between these pelvic indices and selected biosocial factors. Pelvic indices for men and women of similar stature were significantly different, with the exception of the brim index, and these anatomical changes were correlated with year of birth. The similarity of the findings in both sexes suggests common environmental factors such as nutrition are the most likely causes, rather than factors specifically associated with pregnancy and childbirth.