The literature on alcohol education for young people suggests that few interventions have been effective in reducing misuse of alcohol amongst this population. However in a high proportion of such interventions there is an absence of any firm theoretical foundation and of systematic evaluation. This research therefore set out to conduct a controlled prospective study of the effectiveness of a school-based alcohol education programme for students aged 12 to 13 years. The theoretical basis of the proposed intervention is discussed in this thesis with reference to theories of substance use and models of health related behaviour. The research had three phases: (1) pre-intervention survey of alcohol-related knowledge, attitudes and behaviour; (2) development and teaching of a short alcohol education package and (3) post-intervention survey. Five experienced health/social education teachers were involved in the development of the alcohol education materials. Nine participating schools were selected from three regions of Britain - Highland in Scotland, Berkshire in England and Dyfed in Wales. The pre-intervention survey was completed by 1586 respondents, and the post-intervention survey by 1350 of the original study group. The findings from the pre-intervention survey reinforce those from other studies suggesting that the majority of 12 to 13 year olds have limited experience of alcohol, and that this is most likely to occur with parents in the family home. A small percentage, however, reported experience of negative consequences of alcohol consumption and intoxication. The main focus of the study, the evaluation of programme effectiveness, was assessed quantitatively in terms of the shift in alcohol-related knowledge, attitudes and behaviour reported pre- and post-intervention. The principal finding was that the young people exposed to the alcohol education showed a significantly greater increase in alcohol-related knowledge than did the controls. There were no statistically significant differences between control and intervention groups in either attitudes or behaviours. A consistent pattern however did emerge for alcohol-related behaviours, with controls more likely than the intervention group to have increased (a) the recency of their drinking (as indicated at the time of survey) (b) the quantities of beer and wine drunk on the last occasion and (c) the maximum quantity of alcohol drunk in one session. It may be concluded that the approach to alcohol education for young people adopted in this study had some impact on the target population in the intended direction. The implications of this outcome for future initiatives are discussed.

In this thesis the literature on alcohol treatment, evaluation and related issues are reviewed from a historical perspective. Conclusions were reached about both the efficacy of treatment and treatment evaluations. The possibility of a new source model of treatment was considered. Patient data on outcome was analysed on a before and after basis. It was found that the successful outcome group exhibited much the same characteristics as the rest of the sample at phase one, such that it was not possible to extract a discrete profile of completers. It was further noted that the successful outcome group displayed similar characteristics to other groups reported in the literature, social class being a prominent exception. Process data were analysed to observe the pattern of change over the study period. This pattern tended to conform to common sense views of the recovery process held by clinic staff. The pattern was predominantly linear with a disruption occurring around the three to four months period, thereafter the scores continued in a linear fashion. Correlations amongst the scores were noted. The concept of affect balance was introduced as a clinical, methodological and theoretical construct that has value for treatment evaluations. A thirteen life domain scale was constructed to operationalize the concept of affect balance and its properties were discussed.

The objectives of the thesis were to answer the following questions: 1. Are genetic and environmental factors associated with carriage among Greek children and young adults similar to those found in northwest Europe? 2. Are strains with serogroups, serotypes and subtypes associated with disease in northwest Europe isolated from patients and carriers in Greece? 3. Are the antibiotic sensitivities of meningococci isolated from patients and carriers in Greece similar to those observed in northwest Europe? 4. Are the genetic clones associated with disease in northwest Europe present among meningococcal strains in Greece and other Balkan countries? Among military recruits and primary and secondary school children active smoking or exposure to cigarette smoke were significantly associated with carriage of meningococci. In both populations the rate of carriage was higher among the 15-19 year age range; similar results were reported for studies in Britain and the Faroe islands. The results obtained from the epidemiological studies (recruits and schoolchildren) showed that viral upper respiratory tract infection in general is not a predisposing factor for colonisation and that specific viral infections (<I>e.g.</I>, RSV and influenza) need to be investigated. Lower socio-economic group was not associated with carriage in Greek school children but with smoking habits of members of the house closely involved in child care. While the subtype reagents were able to differentiate strains from both patients and carriers, the serotype antibodies did not react with the majority of strains. The serogroup, serotype and subtype combinations associated with outbreaks in northwest Europe were not found among over 500 isolates examined. The studies on meningococcal strains isolated from patients in Greece and Romania were the first to identify significant phenotypic and genetic differences between meningococcal strains isolated in northwest and southeast European countries.

This volume contains 136 copies of articles, chapters and reviews published between 1958 and 1992, set out under a series of headings which cover the major areas of laboratory and clinical research undertaken by me and my associates in that period. Whenever I am listed as first author, all or the greater part of the design and conduct of the work was undertaken by me. Whenever one of my associates or research students is listed as first author, I was intimately involved in the concept, design and conduct of the studies reported in any publication which bears my name. Where original reports are not available, the permission of publishers has been sought for inclusion of photocopies within the thesis and any necessary royalties have been paid. Where such permission was not obtained the articles or chapters have been omitted from the thesis but are listed by title in Appendix 1. Articles in press at the time of submission are listed in Appendix 2. Within each broad area of activity, the publications are grouped in chronological sequence under subheadings which concentrate on particular directions in research or on particular techniques or models. The major components are in the broad field of transplantation immunology, in the related fields of pregnancy and tumour immunology, followed by studies in tissue preservation, the clinical aspects of organ transplantation and dialysis, the surgery of hypertension, some aspects of general surgery and a modest contribution to surgical education. Short summaries of the findings in groups of articles, or in some instances single articles, are presented to highlight particular features of the work and its relevance.

This thesis describes the results of a longitudinal study of regular illicit drug users in Edinburgh, Scotland. Subjects were recruited using the method of 'snowballing'. An initial group of 115 regular users of illicit substances was interviewed using a semi-structured schedule. It was found that multiple substance use was the norm with many individuals reporting adverse consequences resulting from their use of licit and illicit drugs. Concern about HIV infection affected the drug taking behaviour of injectors, but in contrast, had little influence upon the sexual behaviour of both injectors and non-injectors. Sixty-three percent (n = 72) of respondents were reinterviewed approximately 18 months after the initial interview, in order to assess patterns of behaviour change. It was hypothesised that level of involvement in drug-using lifestyles at first interview and repondents' cognitions about their current and future use would be predictive of behaviour change at the time of follow-up. Drug involvement variables, such as length of drug using career, opiate use and having an income from drug dealing, discriminated significantly between users who 'reduced' in terms of the nature and level of their use, and those who 'progressed' or remained 'static'. Cognitions about use, such as perceptions of being addicted and desire to stop, discriminated 'statics' from the 'reducers' and 'progressors'. Results from this study show that drug use behaviour change has multiple predictors at personal, social and environmental levels. These predictors are as varied and complex as those of drug initiation suggested in the retrospective data. Further studies of predictors of change should facilitate the identification of those new users who are 'at risk' of progression, which would have important implications for both primary and secondary prevention.

No description available.

This thesis examines methodological and theoretical issues related to health outcome measures. The International Classification of Functioning, Disability and Health (ICF) has become widely adopted as the theoretical framework of choice. The evaluation and assessment of any theoretical model depends on appropriate measures. It is important to establish that each measure is assessing only the theoretical construct of interest and not other constructs; otherwise, observed relationships between constructs may be misleading. It was shown that existing osteoarthritis measures and items were contaminated with other ICF constructs based on classifications by expert judges. However, a pool of uncontaminated items was identified by the expert judges. These items formed the basis of new measures of Impairment (I), Activity Limitation (A) and Participation Restriction (P). The items were examined using responses from patients having a hip or knee replacement. Item analysis was carried out combining item response theory and classical test theory methods, and led to the removal of misfitting items. Using all the uncontaminated items, patients' responses to the items did not result in statistically separable measures. However, statistically independent I, A and P measures were established using a reduced item set, and these measures appeared to provide a more accurate estimation of the ICF pathways than pre-existing measures. Overall, this thesis highlights the importance of developing health outcome measures that are theoretically-based and both uncontaminated and statistically independent.

The aim of this study was to assess whether pre-natal growth restriction due to intra-uterine factors influences risk factors for coronary heart disease in adult life. 131 pairs of same-six (60 MZ and 71 DZ) twins aged 19-50 years and 89 singleton controls matched for sex, gestational age, maternal age and parity were recruited from a local obstetric database. Measurements were made of height, weight, waist circumference, body fat and blood pressure, and a fasting blood sample was taken for measurements of cholesterol, triglycerides, apolipoproteins, fibrinogen, glucose and insulin. Smoking, diet, physical activity and medical details were obtained by questionnaire, and an ECG was recorded. The birth weight difference between twins ranged from 0-1840g, with a mean within-pair difference of 337g. The overall results suggest that twins do not have an increased risk of coronary heart disease despite their considerably smaller size at birth and substantial catch-up growth. There was no evidence of differences between MZ and DZ twins in either unpaired or paired analysis of birth weight and adult risk factors apart from total cholesterol which was significantly negatively associated with birth weight in DZ twin pairs. Birth weight was inversely associated with cardiovascular risk factors in controls, whereas the associations were inconsistent in twins. The relationship between birth weight and cardiovascular risk factors was stronger in controls than in twins, which indicates that the effect of birth weight in twins is different to that in singletons. The results of this study suggest that growth restriction in mid to late gestation, as a result of inter-uterine factors, does not lead to an increased risk of coronary heart disease in adult twins up to 50 years. This is possibly due to factors influencing birth weight in twins, which do not operate in singletons.

The statistical analysis of repeated measures or longitudinal data always requires the accommodation of the covariance structure of the repeated measurements at some stage in the analysis. The general linear mixed model is often used for such analyses, and allows for the specification of both a mean model and a covariance structure. Often the covariance structure itself is not of direct interest, but only a means to producing valid inferences about the response. This thesis considers methods for the analysis of repeated measurements which arise from very small samples. In Part 1, existing methods of analysis are shown to be inadequate for very small samples. More precisely, statistical measures of goodness of fit are not necessarily the right measure of the appropriateness of a covariance structure and inferences based on conventional Wald type procedures (with small sample adjustments) do not approximate sufficiently well their nominal properties when data are unbalanced or incomplete. In Part 2, adaptive-estimation techniques are considered for the sample covariance matrix which smooth between unstructured and structured forms; 'direct' smoothing, a weighted average of the unstructured and structured estimates, and an estimate chosen via penalised likelihood. Whilst attractive in principle, these approaches are shown to have little success in practice, being critically dependent on the 'correct' choice of smoothing structure. Part 3 considers methods which are less dependent on the covariance structure. A generalisation of a small sample adjustment to the empirical sandwich estimator is developed which accounts for its inherent bias and increased variance. This has nominal properties but lacks power. Also, a modification to Box's correction, an ANOVA F-statistic which accounts for departures from independence, is given which has both nominal properties and acceptable power. Finally, Part 4 recommends the adoption of the modified Box statistic for repeated measurements data where the sample size is very small.

Two studies were undertaken. Firstly, congenital heart defect (CHD) data from the Saudi Arabian Congenital Heart Defects registry (CHD registry) were compared to data published by the Baltimore-Washington Infant Survey (BWIS) group and the European Surveillance of Congenital Anomalies registry (EUROCAT). Distributions of CHD diagnoses within the Saudi Arabian dataset (Riyadh region and Saudi Arabia as a whole) were similar to those from these more comprehensive efforts, providing evidence for the completeness and accuracy of the CHD registry, for Riyadh region in particular. Secondly, an unmatched case-control study of risk factors for all structural congenital heart defects in children resident in Riyadh, Saudi Arabia was undertaken. The primary exposure of interest was consanguinity up to and including third cousins. Incident cases were identified from the CHD Registry from June 1, 2002 to December 31, 2004. Controls were obtained from the Well Baby Clinic, Riyadh Anned Forces (Military) Hospital. Using a detailed and reverse translated questionnaire, a face to face interview was conducted with 235 case and 247 control mothers by research assistants fluent in the local dialect. Mothers were asked to consider their exposure to risk factors within the period of 3 months prior to and 3 months post conception. Consanguinity was collected by phylogram method. The majority of mothers were interviewed when the infant was less than one year of age. Analyses were conducted using four different case groups: all cases, isolated cardiac cases, and embryological earliest and latest cases. Twenty five percent of cases and controls were first cousins or closer. Sixteen percent of cases versus 13 percent of controls were first cousins once removed or equivalent and 12 percent of both cases and controls were second 01' third cousins. Consanguinity was not found to increase the risk of CHD in this population. The adjusted odds ratio for all cases was 1.0 (CI9S=O.7-1.7) and for isolated cardiac cases it was 1.2 (CI9S=O.7-2.0). Statistically significant associations were found for other exposures such as previous pregnancy losses, maternal age, multiplicity, maternal use of hair dyes and pesticides sprayed in the house, confirming findings from previous studies. It is unlikely that the findings for consanguinity can be explained by misclassification of exposure or, in the analysis of all cases, low statistical power.