This thesis presents the results of a study undertaken to assess the contribution of medical factors, especially those of genetic origin, to mild mental retardation. Such knowledge would permit appraisal of the potential role of genetic counselling in reducing the prevalence of mild retardation. Medical histories of 169 Southampton schoolchildren attending schools for the mildly mentally retarded were studied. Each child had a clinical examination, and chromosomal and biochemical analysis. Consideration was given to both those medical features commonly thought to be significant, and to those whose relevance is less certain. Non-medical factors such as parental education and social background were also taken into account. Medical factors of recognised significance were present in 71 children (42%). These were prenatal in 22, perinatal in 41, and postnatal in eight. Factors of possible, but less certain, significance were found in a further 63 children (37%). In 86 families (51%) there was a history of serious learning difficulties in both parents. The prevalence of both types of medical factors was higher in children whose parents had had no educational problems. There were, however, 25 children (15%) whose parents had had no learning difficulties, and in whom medical factors were absent or minimal. Genetic abnormalities were present in 19 children (11.2%), but were largely sporadic. Hence the scope for reduction in the prevalence of mild mental retardation by genetic counselling would appear to be very limited.

The present study compared the families of 27 Anorexia Nervosa patients with the families of 29 Cystic Fibrosis (CF) patients and a third group of 31 well controls. The individual child in each household was aged between 14 and 24. Assessment was carried out at baseline and at an eight month follow up. Data was collected via three modalities: (1) Self report questionnaires (2) the Camberwell Family which was rated for Expressed Emotion (EE) and (3) Observation of a family problem solving task which was rated for evidence of Minuchin's "psychosomatic family" dimensions. The self report measures of family functioning for the most part did not distinguish between the 3 groups. There were no differences between groups in levels of EE criticism. AN and CF households showed more emotional over-involvement (EOI) than did the well households and EOI correlated with severity of illness suggesting that this family characteristic is an illness related phenomenon. Mothers of both CF and AN patients showed more emotional disturbance than did mothers of well young people. There were more enmeshed and overprotective households in the AN group than in the CF and well groups. AN families were also less successful problem solvers than the other two groups giving partial support to Minuchin's conceptualization. The robust psychological health of CF patients found in previous research was replicated here and was in marked contract to the AN sufferers. No conclusive relationship was found between change in health status and change in household climate between baseline and follow up. This could in part be explained by design limitations. AN families were more likely to feel that they had been blamed for their child's illness than were CF families. A brief qualitative account of the AN and CF families in the study showed that they had had very different experiences of getting appropriate support.

For a variety of socio-economic reasons an epidemic of injection drug use (IDU) involving heroin occurred during the early 1980's in Edinburgh: one third were female, most were young, unemployed and living on large council estates. At the peak of this IDU epidemic, HIV arrived and rapidly spread through this community. By July 1989 over 1000 individuals or 0.1% of the population of Lothian (750,000) had been recognised to have been infected with HIV, the majority via IDU. This is of the same order as the worst affected region in England (North West Thames). The majority of these individuals however live in the City of Edinburgh with a population of only around 300,000 (1981 census). Consequently a realistic prevalence for this population is actually 0.3% or 3 times the worst affected English region. Thus this new area of medicine has considerable relevance for future medicine in Edinburgh and Scotland. The thesis describes the disease, the epidemiology of Injection Drug Use and IDU related HIV and the early epidemic in Edinburgh. It also describes the clinical services that were developed at the City Hospital in Edinburgh and the problems that this new service encountered. The provision of health care for this difficult patient population facilitated a variety of research projects. The thesis describes some of the results of these projects particularly those concerned with natural history, clinical presentation and use of antiretroviral therapy in IDU related HIV. Lastly the factors found to affect the transmission of HIV to the heterosexual partners of the patients are presented together with their relevance for other populations.

Heart muscle disease was found in 14.2% of HIV patients and took three principal forms - dilated cardiomyopathy, borderline left ventricular dysfunction and isolated right ventricular dilation. Dilated cardiomyopathy was associated significantly with a very low CD4 count indicative of late stage HIV disease. It was invariably irreversible. In contrast, some patients with borderline left ventricular dysfunction and isolated right ventricular dilation subsequently reverted to normal. The latter was usually related to pressure or volume overload of the right ventricle rather than to a primary myopathic process. Survival curves were calculated and these showed that HIV patients with dilated cardiomyopathy met a significantly earlier death from an AIDS related condition than those from all the other groups, even after accounting for their low CD4 count. This remained true when patients with dilated cardiomyopathy were matched individually with a group of patients identical in every respect except for the presence of cardiac disease. Heart muscle disease in HIV infection is common and takes a number of forms. Dilated cardiomyopathy occurs in late stage disease, is invariably irreversible and is associated with a particularly poor prognosis. This is in contrast to borderline left ventricular dysfunction and isolated right ventricular dilation which occur at an earlier stage of HIV infection, are potentially reversible and do not carry adverse prognostic implications. Neither infection with Toxoplasma gondii and cytomegalovirus nor treatment with zidovudine appear to have a primary role in the development of heart muscle disease. Although HIV is often found within the myocardium, it does not appear to replicate within this tissue. Low serum selenium concentrations are widespread in HIV patients but do not correlate with cardiac dysfunction.

The study reports nosocomial infection rate of 7 infected patients per 100 patients admitted. As a percentage of the total, the most prevalent organisms are: coagulase negative staphylocci (32%), Pseudomonas aeruginosa (23%), Candida species (20%) and Staphylococcus aureus (9&37); the commonest sites of infection are: blood stream (36%), skin/eye drain site (22&37); respiratory tract (16&37); wound (15%) and urinary tract (9%). Severity of the underlying illness of admission, as measured by the PRISM scoring system, predicts a population at risk of developing nosocomial infection. Patients with admission PRISM scores of ≥10 are significantly more likely to acquire infection than those with scores < 10 (10.8% vs 3.6%, p < 0.001) andthis association holds through age, clinical speciality and length of stay. The sensitivity, specificity, positive and negative predictive values of a PRISM score ≥10 are 75%, 53%, 11% and 97% respectively. In post operative cardiac surgery patients non-wound infections account for 72&37 of the total nosocomial infections. With regard to wound infection; the most prevalent pathogenic organisms vary depending on whether surgery is closed ie. non bypass (Staph. aureus and coagulase negative staphylocci) or open ie. bypass (coagulase negative staphylococci, P. aeruginosa, Candida species and Staph. aureus). Risk factors for the acquisition of infection relate to specific operative procedures and to surgical technique particularly the presence of an open sternotomy wound in the post operative period. The system of infection surveillance recently introduced in the PICU (the Infection Control Sentinel Sheet; ICSS) compares favourably with daily bedside examination of patients plus daily review of in-patient charts. The ICSS, which requires only 20 minutes of surveillance time per day, detects 87&37 of nosocomially infected patients; 85&37 of infections at the three standard sites (blood, wound and urine); and 72% of infections at all of the 11 sites surveyed. Certain adverse effects of nosocomial infection are reported to occur in up to 40% of infected patients. A crude costing study of intravenous antibiotic required for treatment of nosocomial infection suggests that the minimum cost was Can 15,000 (approx <i>7,500). In conclusion, following a resume of the results of the individual studies, areas, where future research efforts might be focussed, are identified.

Epidemiological studies of many groups of miners, of various ores, have shown that exposure to radon daughters is associated with increased lung cancer risk. These results, together with surveys of indoor radon levels, suggest that indoor exposure might present a hazard for the general population. However, the magnitude of risk at the comparatively low levels found indoors is uncertain, since the exposure of mining groups, on which risk estimates are based, is in general higher. Surveys of radon gas in British coalmines have shown levels of a similar order of magnitude to those indoors, and it therefore appeared that coalminers might be a suitable occupational group in which to study the effects of low-level exposure to radon daughters. The present study group consists of 19418 male industrial coalmine workers who attended either or both of two medical surveys carried out at 10 British collieries during the 1950s and early 1960s. Deaths in this group up to 31 December 1989 were notified by the Office of Population Censuses and Surveys. Cumulative exposures to radon and thoron daughters acquired during working time were calculated for a subgroup of 14956 men using data from two sources: a set of 146 measurements of daughter levels made at the 10 collieries during the 1970s, and extensive information on time worked underground and on the surface, gathered as part of the British National Coal Board's Pneumoconiosis Field Research. An overlapping subgroup of 14145 men were categorized by smoking habit. Death rates for two causes - lung cancer and stomach cancer - were tabulated by colliery, calendar time period, age, smoking category, and lagged cumulative exposure to radon or thoron daughters. Relationships between death rates and exposure were investigated by Poisson regression analysis. In a series of matched case-referent studies, relationships between exposure and mortality from these two causes of death, and eight others, were analysed by conditional logistic regression. External comparisons of mortality were made for lung cancer and all-causes only; exposure-response was not examined.

No description available.

Prostate cancer is becoming an increasingly important health problem in Scotland. Its reported incidence increased substantially over the past two decades. Mortality rates also increased but less rapidly than incidence. The aim of the study is to describe the incidence trends in Scotland between 1971 - 1990 and to estimate the artefactual increases due to ageing, improved reporting and increased detection of TURP-detected incidental cases, and the true component in the overall trend. More than 90% of cases with an SMRI for PCa were reported to SCR at the end compared to 86% at the beginning of the study. Clinical PCa constituted the majority of all registered cases, but incidental incidence rate increased dramatically (70%) as opposed to 5.7% increase in the clinical PCa incidence. Analysis of the trend shows that out of (540) total increase in annual registrations between 1982-84 and 1988-90, 24.3% (131 cases) could be explained by improvement in reporting, 34% (184 cases) due to increased detection of incidental cases, 35% (190 cases) due to ageing and 6.5% (35 cases) remain to be explained and may be due to untested components or a genuine increase. More than half of the increase in incidence of PCa is artefactual which is not surprising given the dramatic rise in TURP and the plausibility of improved reporting. The clinical disease also increased, though much slower, reflecting perhaps a real increase in prevalence of risk factors. While the artefactual part is reassuring from the public health point of view, for policy makers and finance managers it is a case for arguing against the relative importance of PCa. This, however, has to be considered in the light of similar artefacts in other cancers and some other diseases. The actual increase is present indicating that the disease is to be taken seriously and ways of reducing its morbidity and consequent fatality need to be explored.

159 women and 153 men with epilepsy treated with AEDs (treated group) along with 36 women and 34 men not receiving any AED therapy (untreated group) were recruited from a specialist epilepsy clinic. In addition 48 healthy women and 45 men were recruited to act as controls. All were administered the SES scales. Three of the scales were used - SES1 examines attitudes to sexual morality, SES 2 openness to psychosexual stimulation, and SES 3 the respondents attraction to their sexual partner. SES 3 scale was administered only to those in a stable heterosexual relationship.

The emergence of the Five Factor model, which posits that personality can be described on the five broad dimensions of neuroticism, extraversion, agreeableness, openness to experience and conscientiousness, offered a way to standardize and amalgamate personality-health research. The aims of this study are: to apply the five-factor model to cardiovascular disease, and to follow up previous research on hostility and anger, using objective measures of clinical and subclinical disease. Prospective analysis of hostility data will help determine the causal directions of the relationship, and the use of standard, reproducible instruments to assess the five factors will help further elucidate the role of personality in cardiovascular disease risk. The study is based on a cohort of 809 men and 783 women, aged 55-74 years, who were randomly sampled from the general population in Edinburgh. Submissiveness was protective of objectively-determined nonfatal myocardial infarction over five years, particularly in women. Traits associated with neuroticism were predictive of subjectively defined angina pectoris. Associations between anger-out and prevalent myocardial infarction in men, and neuroticism and prevalent intermittent claudication in women, were prominent. Different aspects of anger were inconsistently associated with other disease outcomes. Low agreeableness was not independently associated with disease. These results confirm the strength and direction of personality and cardiovascular disease associations previously observed, and can help improve our prediction of risk. They indicate that the five factor model's dimensions may be too heterogeneous for this type of research, and that they should be used alongside narrower measures. However, because of the important interaction between many personal factors, such as life stress, coping and especially socioeconomic status, studies examining only one of these elements may be too restrictive. In addition, the research into the biological mechanisms of the association is important and should continue. Only by integrating the individual strands of research can we understand the complex effects of personality and other non-physical factors on cardiovascular diseases.