Since type B adverse drug reactions tend to be rare and serious, they tend to be treated by tertiary-care specialists; and since they are commonly iatrogenic, the specialists should be concerned to document carefully not only the case per se but also the drug use history, leading to practice data of good research quality. The specialist should also be concerned to submit the data record to a central facility that would supply the probabilities, evidence-based, that a recent drug use by a patient caused the adverse event. Continual and systematic accumulation of these data records at the central facility--using the same logistic and organizational framework for each of different type B events--provides for both the numerator and denominator series for etiologic research. Since the targeted events are quite rare, the catchment population of the "registry" would have to be very large, international in scope, especially if the system is to provide for rapid resolution of crises arising from novel suspicions of type B effects with respect to newly marketed drugs.

The Newfoundland Department of Health administers several rubella vaccination programs for the prevention of congenital rubella syndrome (CRS). This study examined the effectiveness of these programs by reviewing rubella susceptibility in the population at risk for CRS, assessing the predictive value of a rubella vaccination record, and evaluating the effectiveness of the postnatal rubella vaccination program. / From 1989 to 1993, rubella susceptibility in women aged 15 to 44 averaged 4.6% overall, but was significantly higher in women aged 15 to 19 years, averaging 14%. The positive predictive value of a rubella vaccination record was 92% overall, but it differed by type of vaccine product and vaccine viral strain; 99% for any monovalent rubella vaccine, compared to 81% for recipients of HPV-77 DE-5 strain MR (measles rubella) or MMR (measles mumps rubella) vaccine. The postnatal rubella vaccination program failed to provide testing for 13% of pregnant women in the province in 1992, and 10% of susceptible women in 1992 were not subsequently vaccinated. / These results suggest that women of childbearing age in Newfoundland remain at risk of having children with CRS. The rubella vaccination record is not adequate proof of immunity for some of these women, and the postnatal vaccination program requires some improvement in order to prevent cases of CRS in the future.

The Brazilian Wilms' Tumor Study Group carried out a case-control study of Wilms' tumour (WT) to identify risk factors for the development of WT. Two controls per case, matched for sex, age (${ pm12}$ months), hospital and interviewer, were chosen. Controls were children admitted to hospital for conditions other than neoplasms. Pediatricians interviewed parents of 109 cases and 218 controls. / Increasing levels of parental education and family income were associated with the diagnosis of WT, probably the result of selecting a hospitalized control group. The positive association between high family income and WT was strongest among the cases diagnosed early, and absent among the cases diagnosed late. / Parental occupational exposure to pesticides or herbicides before the pregnancies of the subjects was identified as a risk factor, with a significant trend for increasing risk with increasing cumulative frequency of use. / Also identified as risk factors were maternal consumption during the pregnancy of coffee in excess of 2 cups/day and of dipyrone, an analgesic (both among the lowest income group), and of alcoholic beverages (among the most educated).

Objective. To investigate the independent association of changes in sexual and drug injection risk behaviour on Human Immunodeficiency Virus (HIV) seroconversion among Montreal injection drug users (IDUs). / Design. Subjects were selected from the St. Luc cohort, a prospective study that has recruited IDUs since September 1988. Eligible subjects must have enrolled in the cohort after December 31, 1991 and completed at least two follow-up visits. HIV serostatus testing was performed at baseline and semi-annually. Questionnaires on injection and sexual risk behaviour were administered at each visit. Cox regression with time-dependent covariates was used in the analysis. / Results. The risk of HIV infection appears to be dependent upon the consistency of risk behaviour practiced. Chronic IDUs, who made no attempt to change risk behaviour, were at high risk of HIV infection. Episodic IDUs, who attempted to practice low risk behaviour but experienced frequent relapse to high risk behaviour, were also at considerable risk of HIV infection. Episodic IDUs demonstrated a higher risk of HIV infection than chronic IDUs for certain behaviours. Stable IDUs, who had managed to maintain low risk behaviour, were at minimal risk of HIV infection. / Conclusion. Despite widespread reported behaviour change among Montreal IDUs, a substantial amount of high risk sexual and injection behaviour continues. Programs should be developed to address IDUs in each stage of change: IDUs who have no desire to change their behaviours, IDUs who require assistance maintaining behaviour changes and IDUs who have adopted low risk practises and are approaching long term maintenance of behaviour change.

The prognosis of Systemic Lupus Erythematosus (SLE) has improved markedly over recent decades, however, little research has focused on the improvement of SLE patient's quality of life. The main objective of this cross-sectional study was to evaluate the relationship between disease activity, cumulative damage and self-reported quality of life in 54 patients with SLE. / Disease activity was measured by the SLE Disease Activity Index (SLEDAI), Systemic Lupus Activity Measure (SLAM-2) and cumulative damage by the Systemic Lupus International Cooperating Clinics/ACR damage index (DI). Quality of life was assessed by the Medical Outcome Survey Short Form 36 (SF-36) and the Euroqol (EQ-5D) self-report questionnaires. Multiple linear regression was used to identify significant predictors of patients' self reported health status. Cumulative damage was found to be associated with physical function, physical health and social functioning (SF-36); disease activity was found to have a significant association with general health (SF-36) and a weaker association on overall health status as evaluated through the 'thermometer' rating scale of the EQ-5D. Patients' ratings of ability with usual activities was strongly related to overall physical health (SF-36) as well as the physical functioning and general health subscales of the SF-36. In addition, patients' ratings of anxiety and depression were strongly related to overall mental health status (SF-36). / In conclusion, physical health of SLE patients was associated with disease activity, disease damage, capacity for usual activity, and mobility.

A survey was planned to explore how estrogen receptor (ER) status of breast cancer was related to dietary and reproductive factors, in postmenopausal patients from Toronto. Unforeseeable circumstances created major delays and, even after enhancements of design, the number of subjects who could be included was seriously reduced. As statistical power had thus become undesirably low, emphasis is placed on the realities of epidemiologic research of this nature, i.e. how inevitable difficulties arise, have to be identified and, at the very least, mitigated. Despite small numbers, the odds of positive ER status were found to be low for patients with many pregnancies and high for those with one or two pregnancies, but intermediate for cases who had never been pregnant. This very strong association underlines the weaknesses of those with measured dietary intakes, which appear unlikely to be of major relative importance.

The inter-relationship between depression, medication use and cognitive decline in older persons has potentially important clinical and public health implications, yet research findings on the nature of this relationship remain inconclusive. This thesis presents a systematic investigation into this topic in a sample of 281 medical inpatients aged 65 and over, who were followed for up to 12 months after admission. / In the first three chapters, the concept, population burden and measurement of depression and cognitive function in the elderly population are described. The relevant literature is reviewed, and the rationale and approaches of this thesis are presented. / In the fourth chapter (1st manuscript), the short-term temporal relationship between depression and cognitive functioning was explored using an interviewer-rated depression severity scale. Based on competing mixed effects models under alternative temporal assumptions, the severity of depression symptoms appeared to have a concurrent rather than prospective relationship with cognitive functioning. / In the fifth chapter (2nd manuscript), diagnostic criteria were used to define depression. After adjusting for covariates, both major and minor depression were significantly predictive of subsequent cognitive decline, and the strength of the association appeared to increase with the duration of "exposure". / In the sixth chapter (3rd manuscript), using a provincial prescription database, the effects of medication exposure on cognitive function were evaluated. Antidepressant use was not associated with cognitive decline in general, but interacted with depression diagnoses. In exploratory analyses, antidepressant use appeared to be associated with improved cognitive function over time in the minor depression group, independent of comorbid diseases, current depression symptoms and concomitant medications. Both major and minor depression were independently predictive of subsequent cognitive decline, especially in those not prescribed antidepressants. / In summary, this thesis demonstrates that, in this sample of older medical inpatients, both major and minor depression are independent risk factors for 12-month cognitive decline. The potentially beneficial effects of antidepressants for patients with minor depression should be investigated.

BACKGROUND Keeping pace with the rapidly evolving demands of infectious disease monitoring requires constant advances in surveillance methodology and infrastructure. A promising new method is syndromic surveillance, where health department staff, assisted by automated data acquisition and statistical alerts, monitor health indicators in near real-time. Several syndromic surveillance systems use diagnoses in administrative databases. However, physician claim diagnoses are not audited, and the effect of diagnostic coding variation on surveillance case definitions is not known. Furthermore, syndromic surveillance systems are limited by high false-positive (FP) rates. Almost no effort has been made to reduce FP rates by improving the positive predictive value (PPV) of surveilled data. OBJECTIVES 1) To evaluate the feasibility of identifying syndrome cases using diagnoses in physician claims. 2) To assess the accuracy of syndrome definitions based on diagnoses in physician claims. 3) To identify physician, patient, encounter and billing characteristics associated with the PPV of syndrome definitions. METHODS & RESULTS STUDY 1: We focused on a subset of diagnoses from a single syndrome (respiratory). We compared cases and non-cases identified from physician claims to medical charts. A convenience sample of 9 Montreal-area family physicians participated. 3,526 visits among 729 patients were abstracted from medical charts and linked to physician claims. The sensitivity and PPV of physician claims for identifying respiratory infections were 0.49, 95%CI (0.45, 0.53) and 0.93, 95%CI (0.91, 0.94). This pilot work demonstrated the feasibility of the proposed method and contributed to planning a full-scale validation of several syndrome definitions. STUDY 2: We focused on 5 syndromes: fever, gastrointestinal, neurological, rash, and respiratory. We selected a random sample of 3,600 physicians practicing in the province of Quebec in 2005-2007, then a stratified random sample of 10 visits per physician from their claims. We obtained chart diagnoses for all sampled visits through double-blinded chart reviews. Sensitivity, specificity, PPV, and negative predictive value (NPV) of syndrome definitions based on diagnoses in physician claims were estimated by comparison to chart review. 1,098 (30.5%) physicians completed the chart review and 10,529 visits were validated. The sensitivity of syndrome definitions ranged from 0.11, 95%CI (0.10, 0.13) for fever to 0.44, 95%CI (0.41, 0.47) for respiratory syndrome. The specificity and NPV were high for all syndromes. The PPV ranged from 0.59, 95%CI (0.55, 0.64) for fever to 0.85, 95%CI (0.83, 0.88) for respiratory syndrome. STUDY 3: We focused on the 4,330 syndrome cases identified from the claims of the 1,098 physicians who participated in study 2. We estimated the association between claim-chart agreement and physician, patient, encounter and billing characteristics using multivariate logistic regression. The likelihood of the medical chart agreeing with the physician claim about the presence of a syndrome was higher when the physician had billed many visits for the same syndrome recently (RR per 10 visits, 1.05; 95%CI, 1.01-1.08), had a lower workload (RR per 10 claims, 0.93; 95%CI, 0.90-0.97), and when the patient was younger (RR per 5 years, 0.96; 95%CI, 0.94-0.97) and less socially deprived (RR most vs least deprived, 0.76; 95%CI, 0.60-0.95). CONCLUSIONS This was the first population-based validation of syndromic surveillance case definitions based on diagnoses in physician claims. We found that the sensitivity of syndrome definitions was low, the PPV was moderate to high, and the specificity and NPV were high. We identified several physician, patient, encounter and billing characteristics associated with the PPV of syndrome definitions, many of which are readily accessible to public health departments and could be used to reduce the FP rate of syndromic surveillance systems. / CONTEXTE La surveillance des maladies infectieuses est un défi en constante évolution et un progrès continu au niveau des méthodes et des infrastructures est nécessaire pour répondre à la demande. Une nouvelle approche est la surveillance syndromique, où le personnel de santé publique, assisté de collecte automatisée de données et d'alertes statistiques, surveille des indicateurs de santé en temps quasi-réel. Plusieurs systèmes de surveillance syndromique s'appuient sur les diagnostics issus de bases de données administratives. Parce que ces codes de diagnostics ne font pas l'objet d'audits, l'effet de variations dans leur codage sur les définitions syndromiques demeure inconnu. OBJECTIFS 1) Évaluer la faisabilité d'identifier des syndromes à partir des diagnostics issus des services facturés par les médecins. 2) Évaluer l'exactitude de définitions syndromiques basées sur les diagnostics issus des services facturés par les médecins.3) Identifier les caractéristiques du médecin, du patient, de la rencontre médecin-patient et du mode de facturation associées au coefficient de prédiction positif (CPP) des définitions syndromiques. MÉTHODES & RÉSULTATS ÉTUDE 1: Cette étude a porté sur un seul syndrome (respiratoire). Nous avons comparés les cas positifs et négatifs identifiés à partir de la facturation, aux dossiers médicaux. Un échantillon de 9 médecins généralistes Montréalais a été utilisé. Les diagnostics de 3 526 visites effectuées par 729 patients ont été extraits des dossiers médicaux, et reliés à la facturation. La sensibilité et le CPP des diagnostics d'infection respiratoire issus de la facturation étaient 0.49 et 0.93. Cette étude de faisabilité a permis la planification d'une validation à grande-échelle de plusieurs définitions syndromiques. ÉTUDE 2: Cette étude a porté sur 5 syndromes: fièvre, gastro-intestinal, neurologique, cutané et respiratoire. Nous avons sélectionné aléatoirement 3600 médecins pratiquant au Québec en 2005-2007 et, parmi tous les services facturés, 10 visites par médecin. Pour chaque visite, le diagnostic du dossier médical a été obtenu grâce à une révision de dossier à double insu. La sensibilité, la spécificité, le CPP et le coefficient prédictif négatif (CPN) des définitions syndromiques basées sur les diagnostics issus de la facturation ont été estimés. 1098 (30.5%) médecins ont participé à l'étude et 10529 visites ont été validées. La sensibilité des définitions syndromiques variait de 0.11 pour la fièvre à 0.44 pour le syndrome respiratoire. La spécificité et le CPN étaient élevés pour tous les syndromes. Le CPP variait de 0.59 pour la fièvre à 0.85 pour le syndrome respiratoire. ÉTUDE 3: Nous avons restreint notre échantillon aux 4330 visites des 1098 médecins de l'étude 2 où le diagnostic de la facturation correspondait à l'un des syndromes. Nous avons utilisé une régression logistique multi-variée afin d'estimer l'association entre l'accord facturation-dossier et les caractéristiques du médecin, du patient, de la rencontre médecin-patient et du mode de facturation. La probabilité que le dossier médical confirme un syndrome présent selon la facturation était plus élevée lorsque le médecin avait facturé plusieurs visites pour le même syndrome récemment, avait une charge de travail moindre, et lorsque le patient était plus jeune et moins défavorisé socialement. CONCLUSIONS Cette étude a été la première validation à grande-échelle de définitions syndromiques basées sur les diagnostics issus des services facturés par les médecins. Nous avons découvert que la sensibilité de ces définitions est faible, le CPP varie de moyen à élevé, et la spécificité et le CPN sont élévés. Nous avons identifiés maintes caractéristiques du médecin, du patient, de la rencontre médecin-patient et du mode de facturation associées au CPP des définitions syndromiques, dont plusieurs sont accessibles aux agences de santé publique et pourraient être utilisées pour améliorer les systèmes de surveillance syndromique.

The objectives of this research are to assess the impact of prenatal courses on the birthweight of newborns, and on two intermediate variables: cigarette consumption and maternal weight gain during pregnancy. / Ninety-eight per cent (98%) of primiparous women giving birth in one of four Montreal hospitals during the nine-month study period, completed a questionnaire during their postpartum stay in hospital. Medical data on mothers and newborns were extracted from the medical records. / Women who participated in prenatal courses were of higher socioeconomic status than nonparticipants. An analysis of covariance showed that prenatal courses did not affect: the mean birthweight of newborns, the proportion of women who gained less than 20 pounds (9.1 kilos) during pregnancy, and cigarette consumption during the last five months of pregnancy. The results were consistent regardless of the > of the prenatal course which was measured by it duration, content, and trimester of enrolment in the prenatal course. / Prenatal courses did not affect the birthweight of newborns or cigarette consumption during pregnancy in women at higher risk for have low birthweight babies. Although prenatal courses did reduce by half the proportion of these higher risk women who gained less than 20 pounds, the result is of little importance since birthweight was not affected. / From a community health perspective, prenatal courses are not appropriate interventions to increase the mean birthweight of newborns. The apparent effect of prenatal courses on the birthweight of newborns is eliminated after adjusting for confounding variables. It is important to adjust for confounding variables when evaluating prenatal courses. The analysis of co-variance is an appropriate analytic technique to perform this adjustment.

The objective of this research was to study the relationship between stillbirth and chemical exposure of pregnant workers. / A case control study was conducted among women working in three sectors: Health, Personal services and Agriculture, and Manufacturing industries. Two hundred and twenty-seven (227) stillbirths aged 20 weeks of gestation or more, were matched to two hundred and twenty-seven (227) livebirths, for mother's age, gravidity and socio-economic status. Chemicals were divided into ten (10) groups. Exposure was assessed by visits to the workplaces, telephone calls or extrapolations. Conditional logistic regression analyses were done. / Women working as hairdressers-beauticians$ sp+$ (O.R. 0,01 (0,00-0,34)) and those working in the garment industry* (O.R. 0,24 (0,08-0,77)) had significantly less stillbirths while women working in metal-electrical-chemical industries* (O.R. 5,11 (0,99-26,37)) had a significant excess of stillbirth. Exposure to pesticides/germicides, irrespective of the level or frequency$ sp+$ (O.R. 2,06 (1,15-3,68)), and occasional exposure to "miscellaneous" chemicals (other chemicals potentially fetotoxic)* (O.R. 12,07 (1,22-119,9)) showed a significant increased risk of stillbirth while exposure to a low level of metals$ sp+$ (O.R. 0,28(0,10-0,83)) showed a significantly lower risk of stillbirth. / Leatherworkers* (O.R. 2,59 (0,47-14,33)), women working in the textile industry$ sp+$ (O.R. 2,55 (0,42-15,41)), agricultural workers* (O.R. 3,88 (0,47-31,88)), women exposed to a moderate (2) level of metals$ sp+$ (O.R. 2,32 (0,59-9,10)) and those exposed to a high (3) level of solvents$ sp+$ (O.R. 2,49 (0,47-13,30)) had a non significant (P $>$ 0,05) increased risk of stillbirth. / The author proposes that ergonomic factors could be a confounder of the association between stillbirth and exposure to pesticides/germicides or metals. A study looking more closely at the association between stillbirth and ergonomic factors is therefore suggested. Because exposure of pregnant workers to elevated levels of chemicals is quite rare, further epidemiological studies should include more subjects. ftn$ sp+$: stillbirths aged 20 weeks of gestation or more. ftn*: stillbirths aged 28 weeks of gestation or more.