En chirurgie, les critères d’évaluation des soins sont surtout relatifs à la technique et à la biologie, et les facteurs psychologiques et sociaux sont peu présents. En particulier, la subjectivité du chirurgien et son rôle dans les soins prodigués sont des aspects peu étudiés, malgré (i) un taux de burnout des chirurgiens proche de 40%, (ii) la singularité du chirurgien (son expérience, sa personnalité, ses choix de prise en charge, etc.), variable clé dans les soins prodigués (comme le psychothérapeute en psychothérapie), (iii) l’environnement social et culturel particulier dans lequel travaillent les chirurgiens. L’objectif de cette thèse est d’explorer les facteurs psychosociaux qui peuvent influencer la pratique de la chirurgie et le bien-être des chirurgiens. Trois études ont été menées. La première est une revue systématique de la littérature qualitative (méta-synthèse) portant sur comment les chirurgiens vivent leur pratique. A l’issue de la première étude, nous avons approfondi les résultats au travers d’une étude qualitative sur l’expérience émotionnelle des chirurgiens hépatobiliaires (chirurgie à haut risque ; surtout oncologique). Une troisième étude prospective de cohorte s’est intéressée à la détermination de l’influence des facteurs psychologiques du patient (dépression et anxiété) sur l’outcome de la chirurgie hépatique, et a permis d'explorer comment les chirurgiens prennent en compte ces facteurs dans leur pratique. Nous avons discuté les implications de nos résultats par rapport à la santé mentale et aux burnouts des chirurgiens, ainsi que par rapport à la mise en œuvre du modèle bio-psycho- social dans les soins des patients chirurgicaux. / In surgery, psychological and social factors are not widely studied, and the only variables considered important in surgical care are those related to operating technique and biology. Additionally, surgeons’ experience of their practice is neglected, despite the fact that (i) epidemiological studies showed a 40% burnout rate among surgeons, (ii) the difference between individual surgeons is a key variable in the performed care (like psychotherapists in psychotherapy), and (iii) surgeons work in a high salient social and cultural environment. The aim of this PhD thesis is to explore the psychosocial factors influencing surgical practice and surgeons’ wellbeing. Our findings are based on three studies. First, we performed a systematic review of international qualitative studies (metasynthesis) in order to describe how surgeons experience their practice from their perspective. Then, a second qualitative study extended the findings of the previous one exploring in-depth the everyday emotional experience of liver surgeons (high risk surgery, mainly cancer-related). Finally, a prospective cohort study investigated the influence of preoperative patient-related psychological factors (depression and anxiety) on the outcome of liver surgery, and how surgeons take these data into account in their practice. Several implications of our findings have been discussed, concerning surgeons’ mental wellbeing and burnout, as well as the implementation of the bio-psycho-social model in the care of surgical patients.

Coronary heart disease (CHD) is a leading cause of morbidity and mortality in the industrialised world, and places a heavy burden on society in terms of personal disability and health care costs. The first signs of CHD often present acutely as a myocardial infarction (AMI), commonly known as a heart attack. Survivors of a heart attack remain vulnerable to poor health-related quality of life (HRQOL), further cardiac events, and increased morbidity due to a progression of CHD. Thus, the implementation of interventions to reduce these risks is an important public health strategy. To date, secondary prevention and rehabilitation efforts post-AMI focus primarily on the patient. However, it is argued that recovery from AMI occurs within a social context, and that risk reduction strategies are likely to be enhanced if interventions take into account the impact of the event on the quality of life of patients and their partners. Evidence from a review of couple relationship literature indicates that a significant proportion of couples experiences poor HRQOL (i.e., physical and emotional wellbeing) when coping with stressful life events, and that interactive aspects of a couple relationship (i.e., dyadic functioning and behaviour) are associated with individual well-being at such a time. Information from studies of couples dealing with recovery from heart attack is sparse, but tends to reflect the findings from the broader literature. Further research is required with post-AMI couples, though, as there are a number of shortcomings associated with the existing evidence. For instance, it is derived from studies conducted with, mostly small, samples of convenience; many different instruments are used to collect the data; and no studies specifically measure HRQOL. Analytically, most evidence is obtained with univariate and bivariate statistics, and data are analysed as groups of patients or partners, as opposed to dyads. Where multivariable analyses are undertaken a number of bivariate relationships are no longer significant after accounting for covariates such as age and gender; and few researchers investigate predictive associations between dyadic functioning/behaviour and HRQOL outcomes. Finally, there is a paucity of information from comparative analyses. Thus, it is not known whether the well-being of post-AMI couples over time is better than, similar to, or worse than, for example, that in the general population. The research program underpinning this thesis, the QUT-AMI Project, comprised two studies designed to address these methodological issues. The first was an observational, cross-sectional, pilot study conducted in 1998 with 26 post-AMI couples. The main investigation was a prospective cohort study of 93 post-AMI couples undertaken in 1999-2000. In both studies the samples comprised a consecutive series of adult males younger than 75 years who had experienced a first AMI, and their female partners. The average couple in both studies was middle-aged, had been married for many years, and both members of the dyad were working at the time of the heart attack. Prospective participants were identified in major clinical centres that admit cardiac patients, and couples were recruited to the project soon after the patient's heart attack. Clinical data were collected in hospital. Further data were collected with self-administered questionnaires during a home visit at 1 month (pilot and main study), and by mailed questionnaire or during a home visit at 6 months(main study) after the heart attack. The pilot study was undertaken to test recruitment and data collection procedures in preparation for the second (main) study, measure couples' HRQOL at 1 month after the event using the SF-36, and qualitatively investigate life issues for couples coping with recovery from AMI. In the main study couples' HRQOL outcomes were measured at 1 and 6 months post-AMI using the SF-36, and examined for changes over that time. The outcomes were also compared with those from matched population norms to estimate the impact of a heart attack on couples' HRQOL during the early and later recovery period. Additionally, the following relationships were investigated to determine the extent to which:* patients' dyadic functioning (e.g. happiness/satisfaction with relationship, measured with the Marital Adjustment Scale) and use of dyadic behaviour (e.g., hiding concerns and negative feelings from the other member of the dyad, measured with the Protective Buffering Scale) at 1 month predicted patients' emotional well-being at 6 months post-AMI;* partners' dyadic functioning and behaviour at 1 month predicted partners' emotional well-being at 6 months post-AMI;* patients' and partners' dyadic functioning at 1 month predicted patients' or partners' emotional well-being at 6 months post-AMI; and* patients' and partners' dyadic behaviour at 1 month predicted patients' or partners' emotional well-being at 6 months post-AMI. Exploratory analyses were also undertaken to determine the effect of dyadic discrepancies in functioning and behaviour, at 1 month after the heart attack, on patients' and partners' emotional well-being at 6 months after the event. Important findings were as follows:* At 1 month after an AMI the HRQOL of couples is impaired. The major impact is on physical well-being for patients, and emotional well-being for their partners.* In general, couples' HRQOL improves between 1 and 6 months after an AMI.* At 6 months after an AMI, the HRQOL of average couples is similar to that of their peers in the normal population.* There are subgroup variations in the quality of life of post-AMI couples, and these are associated with age, clinically poor physical health, and depression.* The combination of patients' and partners' use of dyadic behaviour at 1 month after an AMI explains 7% of the variation in patients' emotional well-being at 6months after the event, after adjustment for patients' concurrent physical wellbeing and prior levels of emotional well-being, as well as duration of couple relationships.* The combination of partners' perceptions of dyadic functioning and use of dyadic behaviour at 1 month after an AMI explains 5% of the variation in partners' emotional well-being at 6 months after the event, after adjustment for partners' concurrent physical well-being and prior levels of emotional well-being, as well as duration of couples' relationships.* Patients have poorer emotional well-being at 6 months after an AMI if partners use dyadic behaviour infrequently at 1 month after the event.* Partners have poorer emotional well-being at 6 months after an AMI if they are not satisfied/unhappy with the functioning of their relationships at 1 month after the event. Adjusted exploratory analyses, undertaken to determine the extent to which dyadic discrepancies in perceptions of functioning or use of protective buffering behaviour, predict emotional well-being, show that patients who are less satisfied/unhappier with functioning than their partners at 1 month after an AMI have poor emotional wellbeing at 6 months after the event; patients who use the behaviour more frequently than their partners at 1 month after an AMI have poor emotional well-being at 6months after the event; and partners who are less satisfied/unhappier with functioning than their patients at 1 month after an AMI have poor emotional well-being at 6months after the event. The evidence from the QUT-AMI Project supports the proposition that the interaction that occurs within a couple relationship, combined with individual characteristics of members of a dyad, influences the extent to which a heart attack impacts on couples' HRQOL. It is argued that it is not enough to merely focus on implementing secondary prevention strategies with post-AMI patients. Given that poor emotional well-being is known to predict adverse cardiac events, and premature mortality due to cardiac disease, it is recommended that a couple-focused intervention designed to meet specific needs should be implemented with at-risk couples as a public health strategy to improve not only the patients' quality of life but also that of his partner. Further research is recommended to determine the extent to which such an intervention improves post-AMI couples' quality of life.

Contexte. La pandémie de la COVID-19 a amené son lot de détresse psychologique chez les travailleurs de la santé et des services sociaux (TSSS) du Québec. Ceux-ci se trouvaient déjà à risque d’éprouver des difficultés de santé mentale avant la pandémie, étant exposés à plusieurs événements critiques dans leur travail quotidien. Parallèlement à cette réalité, la pandémie a été très médiatisée et des points de presse quotidiens animés par la santé publique et le gouvernement étaient diffusés sur plusieurs plates-formes d’informations. Les études montrent que l’exposition médiatique à des traumas collectifs, tel que la COVID-19, est associée à des niveaux de détresse psychologique plus élevés chez la population générale. Les femmes sont aussi susceptibles de vivre des niveaux plus élevés de détresse psychologique que les hommes. Or, aucune étude ne fait état de l’impact de l’exposition aux médias sur la détresse psychologique vécue chez les TSSS en contexte de pandémie, ni en général. Objectifs. (1) Mesurer l’impact du degré d’exposition aux médias sur l’évolution de la détresse psychologique des TSSS pendant 37 semaines. (2) Mesurer l’effet modérateur du sexe biologique sur la relation entre le degré d’exposition aux médias et la détresse psychologique des TSSS. Méthode. Le devis est longitudinal, étant une étude de cohorte prospective. Ce sont 830 TSSS québécois qui ont participé à l’étude pendant les 1ière et 2ième vagues de la COVID-19, de mai 2020 à janvier 2021. La détresse psychologique, définie par des symptômes dépressifs et anxieux, a été mesurée à l’aide du PHQ-9 et du GAD-7. Des analyses statistiques à l’aide du modèle linéaire mixte ont été effectuées. Résultats. (1) Les TSSS qui se sont beaucoup ou énormément exposés aux médias rapportent un nombre significativement plus élevé de symptômes dépressifs que ceux étant très peu exposés. Les TSSS qui se sont beaucoup exposés aux médias rapportent un nombre significativement plus élevé de symptômes anxieux que ceux étant très peu exposés. (2) Il n’y a pas de différence significative entre les hommes et les femmes quant à la détresse psychologique vécue en fonction du degré d’exposition aux médias. Conclusions. Des campagnes de sensibilisation et d’éducation devraient être effectuées dans le réseau de la santé et des services sociaux quant à l’exposition aux médias des TSSS afin de ne pas créer de détresse psychologique inutile à ces derniers. L’exploration d’un temps limite d’exposition et du type de médias à privilégier constituent des pistes de recherches intéressantes pour la prévention de la détresse psychologique. / Context. The COVID-19 pandemic has brought its share of psychological distress to Quebec’s health and social services workers (HSSWs). These workers were already at risk of experiencing mental health issues prior to the pandemic, as they were exposed to several critical events in their daily work. Along with this reality, the pandemic was highly publicized, with daily press briefings by public health and other government officials being broadcast on several news platforms. Studies show that media exposure to mass trauma, such as COVID-19, is associated with higher levels of psychological distress in the general population. Women are also likely to experience higher levels of psychological distress than men. Yet, there are no studies reporting on the impact of media exposure on psychological distress experienced among HSSWs during the pandemic, or in general. Objectives. (1) To measure the impact of media exposure level on changes in psychological distress among HSSWs over 37 weeks. (2) To measure the moderating effect of biological sex on the relationship between media exposure level and psychological distress among HSSWs. Method. The design was longitudinal, being a prospective cohort study. A total of 830 Quebec’s HSSWs participated in the study during waves 1 and 2 of COVID-19, from May 2020 to January 2021. Psychological distress, conceptualized as depression and anxiety symptoms, was measured using the PHQ-9 and GAD-7. Statistical analyses using linear mixed models were performed. Results. (1) HSSWs who considered themselves often or greatly exposed to media reported significantly more depression symptoms than those who considered themselves rarely exposed. HSSWs who considered themselves often exposed to media reported significantly more anxiety symptoms than those who considered themselves rarely exposed. (2) There was no significant difference between men and women in experienced psychological distress as a function of media exposure level. Conclusions. Awareness and education campaigns should be carried out in the health and social services network regarding the media exposure of HSSWs in order to avoid creating unnecessary psychological distress for them. The exploration of a limit exposure time and the type of media to be favored are interesting avenues of research to prevent psychological distress.

Untersuchungsziel: Untersucht wurden Auswirkungen einer ambulanten orthopädischen Rehabilitation nach der Implantation einer Hüft- bzw. Knieendoprothese auf die individuelle Lebensqualität. Um zu erkennen, ob Zusammenhänge von individuellen Parametern wie Geschlecht, Alter, Bildungsstand und dem Rehabilitationsverlauf bzw. –ergebnis bestehen, wurden medizinisch-funktionelle und psychosoziale Daten erhoben, sowie ergänzende Variablen durch Schichtenindices und mittels eines strukturierten Interviews erfasst. Methode: Systematische Datenerhebung von medizinisch funktionellen (Gelenkbeweglichkeit, Muskelkraft) und psychosozialen (Aktivitäten des täglichen Lebens) Daten von n=60 Pat. mit Knie-TEP und n=61 Pat. mit Hüft-TEP zu vier Messzeitpunkten (Anfang Reha=t0, Ende Reha=t1, 6 Monate nach Reha= t2 und 24 Monate nach Reha= t3) und strukturiertes Interview zur Erfassung von soziodemographischen Merkmalen, Motivation und Behandlungszuversicht. Analysemethoden: Signifikanzanalyse und Berechnung der Effektstärke (Cohens d) der Veränderungen zwischen t0 – t1 für abhängige Stichproben (getrennt nach Indikation), Längsschnittdarstellung der funktionellen und psychosozialen Parameter mit deskriptivstatistischen Analysen getrennt nach Indikation und Geschlecht. Ergebnisse: Für beide Indikationen ergaben sich hohe Effekte für Verbesserungen der funktionellen Parameter und mittlere (teilhabebezogen) bzw. hohe (aktivitätsbezogen) Effekte bei den psychosozialen Parametern am Ende der Rehabilitation. Die teilstandardisierten Rehabilitationsmaßnahmen wirken sich indikations- und geschlechtsbezogen unterschiedlich aus. Unter Beachtung der psychophysischen Leistungsvoraussetzungen und sozialen Rahmenbedingungen ist die ambulante Rehabilitation gut geeignet für Patienten im höheren Lebensalter. / Study objective: Effects of an outpatient orthopedic rehabilitation after the implantation of a hip or knee replacement on the individual quality of life were studied. In order to analyze the existence of significant relationships between individual parameters such as gender, age, education and the rehabilitation outcome medical-functional and psychosocial data were collected, and additional layers of indices and variables by using a structured interview were applied. Method: Systematic data collection from medical functional (joint mobility, muscle strength) and psychosocial (activities of daily living) data of n = 60 patient with total knee replacement and n = 61 patient with total hip replacement to four time points (beginning of rehabilitation = t0, end rehab = t1, t2 = 6 months after rehabilitation and 24 months after discharge = t3) and structured interview for socio-demographic characteristics, motivation and confidence in treatment Analysis: Significance analysis and calculation of effect size (Cohen''s d) of the changes between t0 and t1 for dependent samples (separated by indication), longitudinal section view of the functional and psychosocial parameters with descriptive-statistical analyses apart from indication and sex Results: For both indications high effects of improvement in functional parameters, mean effects (participatory based) and high (activity-related) effects on the psycho-social parameters at the end of rehabilitation could be identified. The semi-standardized rehabilitation measures had shown differences by indication just like gender. In accordance with the psychophysical performance conditions and social environment outpatient rehabilitation is well suited for patients in advanced age.

La majorité des personnes utilisatrices de drogues par injection (PUDI) contracteront le virus de l’hépatite C (VHC), les mettant à risque accru de complications hépatiques graves et parfois mortelles. Les comportements les plus risqués pour l’acquisition du VHC incluent le partage de matériel d’injection et l’injection à haute fréquence. Un facteur jusqu’ici négligé dans l’évaluation du risque de VHC est l’aspect dynamique de l’injection, c.-à-d. la manière dont elle varie dans le temps, incluant l’effet des périodes sans injection et celui des changements dans la fréquence d’injection. On reconnaît également l’effet délétère que l’instabilité résidentielle peut avoir sur le risque de VHC, bien que les mécanismes sous-jacents soient mal compris. Cette thèse s’intéresse à l’effet des variations temporelles de l’injection sur le risque de VHC, et à la manière dont la fréquence d’injection évolue en concomitance avec les conditions résidentielles dans le temps, afin d’aider au développement de nouvelles stratégies de prévention du VHC. Les données ont été recueillies entre mars 2011 et juin 2016 dans la Hepatitis Cohort, une cohorte de PUDI suivies trimestriellement à Montréal, au Québec. Une première analyse a évalué l’effet des périodes sans injection de trois mois ou moins sur le risque de VHC sur 916 personnes-années de suivi, par régression de Cox (N=372). Celle-ci suggère que les PUDI présentant des périodes sans injection courtes (3/3 mois sans injection) et sporadiques (1/3 ou 2/3 mois sans injection) sont respectivement 76% et 44% moins à risque de VHC que celles s’injectant de manière persistante (0/3 mois sans injection). Une deuxième analyse a utilisé la modélisation de trajectoires fondée sur le groupement pour identifier cinq types distincts de trajectoires de fréquence d’injection suivies sur une année, lesquels ont ensuite été comparés en termes d’incidence du VHC sur des périodes de suivi allant de 71 à 355 personnes-années (N=386). Les résultats suggèrent que les PUDI dont la fréquence reste élevée (injection fréquente) ou change dans le temps (croissante, décroissante) sont à plus haut risque de VHC que celles s’injectant à basse fréquence (sporadique, peu fréquente). Une dernière analyse a identifié trois types de trajectoires de stabilité résidentielle suivies sur un an (persistance, déclin, amélioration; N=386), lesquels ont été évalués en association avec les trajectoires de fréquence d’injection suivies simultanément. Les résultats suggèrent qu’il existe un lien entre l’amélioration des conditions résidentielles et la diminution de la fréquence d’injection, mais aussi que la probabilité d’injection à fréquence croissante est plus élevée chez les PUDI maintenant des conditions résidentielles stables que celles chez qui elles s’améliorent. Collectivement, les résultats ont de nombreuses implications en termes de prévention du VHC. Cliniquement, l’instabilité de la fréquence d’injection semble être un facteur de risque à monitorer régulièrement. En termes de santé publique, les interventions favorisant l’engagement dans des périodes sans injection ou le maintien d’une basse fréquence d’injection pourraient être prometteuses. Enfin, les stratégies visant l’amélioration des conditions résidentielles pourraient éventuellement aider les PUDI à réduire leur fréquence d’injection, mais être insuffisantes pour maintenir celle-ci à basse fréquence une fois la stabilité atteinte. / The majority of people who inject drugs (PWID) will become infected with hepatitis C virus (HCV), placing them at risk of serious and sometimes fatal liver complications. Injecting behaviours with higher risk of HCV transmission include injecting equipment sharing and high frequency injecting. One factor that has been overlooked when assessing HCV acquisition risk is the dynamic aspect of drug injecting, i.e., how drug injecting varies over time, including the role of injecting cessation episodes and that of changes in injecting frequency. Moreover, there is growing recognition of the deleterious effect unstable housing can have on HCV acquisition risk, although the underlying mechanisms are not yet fully understood. This thesis examines how temporal variations in drug injecting relate to HCV acquisition risk and further explores how housing conditions and injecting frequency evolve together over time, for the purposes of contributing to the development of novel HCV prevention strategies. Data were collected between March 2011 and June 2016 in the Hepatitis Cohort, a prospective cohort study of PWID interviewed and tested for HCV infection at three-monthly intervals in Montréal, Québec. A first analysis examined the effect of injecting cessation episodes of three months or less on the risk of contracting HCV during 916 person-years of follow-up, using Cox regression (N=372). Results suggest that PWID with short injecting cessation episodes (3/3 months without injecting) or sporadic injecting cessation episodes (1/3 or 2/3 months without injecting) are 76% and 44% less at risk of contracting HCV than those with persistent injecting (0/3 months without injecting), respectively. A second analysis used group-based trajectory modeling to identify five distinct types of one-year injecting frequency trajectories and compared these in terms of HCV incidence over follow-up periods ranging from 71 to 355 person-years (N=386). Findings suggest that PWID injecting with consistently high frequencies (frequent) or time-varying frequencies (increasing, decreasing) are at greater HCV acquisition risk compared with those maintaining low injecting frequencies (sporadic, infrequent). Finally, a third analysis identified three types of one-year housing stability trajectories (sustained, declining, improving) and examined their associations with concomitant injecting frequency trajectories (N=386). Findings suggest an association between improving housing stability and decreasing injecting frequency, but also a higher probability of increasing injecting frequency among PWID who maintain housing stability compared to those that improve it. Collectively, these findings have numerous implications for HCV prevention. Clinically, instability in injecting frequency appears to be a risk factor that should be monitored regularly. From a public health perspective, interventions that promote engagement in injecting cessation episodes or maintenance of low injecting frequency may be promising. Finally, strategies aimed to improve housing stability may help PWID to decrease their injecting frequency but may not be sufficient to help them maintain low injecting frequencies once housing stability is achieved.