Background: Knee osteoarthritis is a degenerative disease characterized by damaged joint tissues (e.g., cartilage) that leads to joint pain, and reduced mobility and quality of life. Various factors are involved in disease progression, including biomechanical, patient-reported outcome and mobility measures. This thesis provides important longitudinal data on the role of these factors in disease progression, and the trajectory of biomechanical factors in persons with knee osteoarthritis. Objectives: (1) Determine the extent to which changes over 2.5 years in knee cartilage thickness and volume in persons with knee osteoarthritis were predicted by the knee adduction and flexion moment peaks, and knee adduction moment impulse and loading frequency. (2) Determine the extent to which changes over 2 years in walking and stair-climbing mobility in women with knee osteoarthritis were predicted by quadriceps strength and power, pain and self-efficacy. (3) Estimate the relative and absolute test-retest reliabilities of biomechanical risk factors for knee osteoarthritis progression. Methods: Data were collected at 3-month intervals during a longitudinal (3-year), observational study of persons with clinical knee osteoarthritis (n=64). Magnetic resonance imaging of the study knee was acquired at the first and last assessments, and used to determine cartilage thickness and volume. Accelerometry and dynamometry data were acquired every 3 months, and used to determine knee loading frequency and knee muscle strength and power, respectively. Walking and stair-climbing mobility, as well as pain and self-efficacy data, were also collected every 3 months. Gait analyses were performed every 6 months, and used to calculate lower-extremity kinematics and kinetics. Results: (1) The knee adduction moment peak and impulse each interacted with body mass index to predict loss of medial tibial cartilage volume over 2.5 years. These interactions suggested that larger joint loads in those with a higher body mass index were associated with greater loss of cartilage volume. (2) In women, lower baseline self-efficacy predicted decreased walking and stair ascent performances over 2 years. Higher baseline pain intensity/frequency also predicted decreased walking performance. Quadriceps strength and power each interacted with self-efficacy to predict worsening stair ascent times. These interactions suggested that the impact of lesser quadriceps strength and power on worsening stair ascent performance was more important among women with lower self-efficacy. (3) Relative reliabilities were high for the knee adduction moment peak and impulse, quadriceps strength and power, and body mass index (i.e., intraclass correlation coefficients >0.80). Absolute reliabilities were high for quadriceps strength and body mass index (standard errors of measurement <15% of the mean). Data supported the use of interventions effective in reducing the knee adduction moment and body mass index, and increasing quadriceps strength, in persons with knee osteoarthritis. Conclusion: Findings from this thesis suggest that biomechanical factors play a modest independent role in the progression of knee osteoarthritis. However, in the presence of other circumstances (e.g., obesity, low self-efficacy, high pain intensity/frequency), biomechanical factors can vastly worsen the disease. Strategies aiming to curb structural progression and improve clinical outcomes in knee osteoarthritis should target biomechanical and clinical outcomes simultaneously. / Thesis / Doctor of Philosophy (PhD) / Knee osteoarthritis is a multifactorial disease whose progression involves worsening joint structure, symptoms, and mobility. Various factors are linked to the progression of this disease, including biomechanical, patient-reported outcome and mobility measures. This thesis provides important information on how these factors, separately and collectively, are involved in worsening disease over time, as well as benchmarks that are useful to clinicians and researchers in interpreting results from interventional or longitudinal research. First, we examined how different elements of knee loading were associated with changes in knee cartilage quantity over time in persons with knee osteoarthritis. Second, we examined how different elements of knee muscle capacity and patient-reported outcomes were related to changes in mobility over time in persons with knee osteoarthritis. Third, we examined the stability over time of various biomechanical risk factors for the progression of knee osteoarthritis. Novel results from this thesis showed that: (1) larger knee loads predicted cartilage loss over 2.5 years in obese individuals with knee osteoarthritis but not in persons of normal weight or overweight; (2) among women with knee osteoarthritis with lower self-efficacy (or confidence), lesser knee muscle capacity (strength, power) was an important predictor of declining stair-climbing performance over 2 years; and (3) clinical interventions that can positively alter knee biomechanics include weight loss, knee muscle strengthening, as well as specific knee surgery and alterations during walking to reduce knee loads. Interventions for knee osteoarthritis should target biomechanical and clinical outcomes simultaneously.

Motivation Alcohol consumption is one of the major risk factors of cardiovascular disease (CVD). Excessive alcohol drinking is the fifth leading cause of death worldwide and the prevalence of alcohol abuse continues to increase especially in low-income areas of sub-Saharan Africa. The alarming rate of urbanisation seems to be the driving force for excessive alcohol intake in the developing world. In addition to its influence on CVD, heavy drinking also results in a number of non-cardiovascular consequences that include injury, risky sexual behaviour, violent crime and family dysfunction among black South Africans, contributing to high mortality. Moreover, the highest number of individuals with human immunodeficiency virus (HIV) infection in South Africa is partly attributable to high intake of alcohol. HIV remains a major concern in South Africa with significant funding diverted to address the pandemic. The continued increases in mortality from preventable outcomes such as stroke, myocardial infarction and renal failure are largely due to urbanisation, poverty and dysfunctional health systems working with limited budgets. These are some of the factors requiring in-depth study of the scientific aspects of alcohol intake in South Africa. Although there is enough evidence that links excessive drinking with hypertension and CVD, the markers of alcohol intake – self reporting of alcohol, gamma-glutamyltransferase (GGT) and carbohydrate deficient transferrin – are still not specific enough to isolate other confounding factors in the association of alcohol intake with CVD. The markers of alcohol that independently predict CVD and mortality need to be explored. Finally, the severe lack of longitudinal investigations on alcohol-related hypertension development and total mortality in black South Africans has compromised the early identification of risk factors associated with these outcomes. This study will therefore attempt to address the limited availability of longitudinal studies and stimulate interest for continued investigation. Aim The aim of this study was to investigate whether alcohol intake of black South Africans is related to specific measures of cardiovascular function (change in blood pressure (BP), hypertension development) and mortality over a period of 5 years. Methodology This study was based on the international Prospective Urban and Rural Epidemiology (PURE) study which includes 26 countries, investigating the cause and development of cardiovascular risk factors in low, middle and high income countries. This South African leg of the PURE study started in 2005 in which the baseline data was collected from 2021 black South Africans from rural and urban areas in Ikageng, Ganyesa and Tlakgameng in the North West Province. Eleven participants presented with missing data, leaving 2010 participants with complete datasets at baseline. However, data from these 11 participants was useful, especially for Chapter 4. All participants gave informed consent and the Ethics committee of the North-West University (Potchefstroom Campus) approved the study. The follow-up data collection was done in 2010. General health questionnaires, anthropometric measurements, lipid profiles and cardiovascular measurements were taken both at baseline and follow-up using appropriate methods. We also collected blood samples and performed biochemical analyses for lipid markers, liver enzymes, inflammatory markers and percentage carbohydrate deficient transferrin (%CDT). Finally, we obtained data on cardiovascular and non-cardiovascular mortality through verbal autopsy and death certificates. We made use of analysis of variance (ANOVA) and Chi-square tests to compare means and proportions, respectively. We used dependent t-tests and the McNemar test to compare baseline and follow-up variables. Furthermore, we employed single and partial linear regression analyses to correlate alcohol markers with each other and with the cardiovascular measures. Multiple regression analyses were used to correlate dependent variables in the study with various independent variables as required. Finally, we employed multivariable-adjusted Cox regression analyses to assess the association of the selected alcohol markers with mortality while adjusting for several independent variables. Results and Conclusions of each manuscript - With the first research article (Chapter 4), we aimed to compare self-reported alcohol intake estimates with GGT and %CDT, considering their relationship with percentage change in brachial blood pressure (BP) and central systolic blood pressure (cSBP) over 5 years. The results indicated that only self-reported alcohol intake independently predicted % change in brachial BP and cSBP. This was not found for the biochemical markers GGT and %CDT. Self-reported alcohol intake seems to be an important measure to implement by health systems in low income areas of sub-Saharan Africa, where honest reporting is expected. - Given the likely presence of high GGT levels in both alcohol consumption and non-alcoholic fatty liver disease (NAFLD), the second manuscript (Chapter 5) aimed to compare the cardiovascular and metabolic characteristics of excessive alcohol users and individuals with suspected NAFLD (confirmed with self-report, GGT and %CDT). We found that different sex and cardiometabolic profiles characterised excessive alcohol users and individuals suspected with NAFLD. Lean body mass and male sex were the dominant characteristics in excessive alcohol use while the NAFLD group had a dysmetabolic profile with obese women making up the higher proportion of this group. In excessive alcohol users systolic blood pressure and pulse pressure were independently associated with high-density lipoprotein cholesterol. Diastolic blood pressure showed a significant correlation with waist circumference. These disparate profiles may guide healthcare practitioners in primary healthcare clinics to identify individuals with elevated GGT levels who may suffer from NAFLD or alcohol overuse. These results emphasise the importance of modifiable risk factors as the main contributors to CVD and that lifestyle change should be the main focus in developing countries such as South Africa. - The third manuscript (Chapter 6) aimed to determine the measure of alcohol intake (selfreported alcohol intake, GGT and %CDT) that related best with hypertension development, cardiovascular and all-cause mortality over 5 years in the same population of black South Africans. We found that GGT was the only independent predictor of hypertension development, cardiovascular as well as all-cause mortality. Moreover, self-reporting of alcohol intake predicted incident hypertension, confirming our findings from Chapter 4. The third marker, %CDT, a highly specific marker of alcohol intake, was not related with any outcome variable, perhaps due to its low sensitivity. Although self-reported alcohol intake is useful in low-resource primary healthcare settings, measurement of GGT is encouraged due to its predictive value for hypertension and mortality. GGT represents alcohol intake, non-alcoholic steatohepatitis and obesity - all known to have severe cardiovascular consequences. Discussion and Conclusions Excessive alcohol intake remains a major concern in the development of hypertension, CVD and premature death in sub-Saharan Africa. Despite their weaknesses such as bias and nonspecificity, self-reporting of alcohol consumption and GGT emerged as reliable alcohol markers that independently predicted 5-year change in BP, hypertension development and total mortality in this population. Serum %CDT did not show any association with the mentioned cardiovascular markers. Finally, we were also able to show that black South Africans with suspected NAFLD (i.e. with high GGT levels who do not consume alcohol) are typically obese women, whereas lean men were more likely to have high alcohol consumption. Further prospective investigations are encouraged regarding (a) these mentioned associations, as well as (b) other self-reporting estimates such as quantity and frequency of drinking and (c) the use of %CDT as a highly specific marker of alcohol intake. The simultaneous presence of HIV infection in alcohol abuse in this population also warrants further investigation. / PhD (Physiology), North-West University, Potchefstroom Campus, 2015

Motivation Alcohol consumption is one of the major risk factors of cardiovascular disease (CVD). Excessive alcohol drinking is the fifth leading cause of death worldwide and the prevalence of alcohol abuse continues to increase especially in low-income areas of sub-Saharan Africa. The alarming rate of urbanisation seems to be the driving force for excessive alcohol intake in the developing world. In addition to its influence on CVD, heavy drinking also results in a number of non-cardiovascular consequences that include injury, risky sexual behaviour, violent crime and family dysfunction among black South Africans, contributing to high mortality. Moreover, the highest number of individuals with human immunodeficiency virus (HIV) infection in South Africa is partly attributable to high intake of alcohol. HIV remains a major concern in South Africa with significant funding diverted to address the pandemic. The continued increases in mortality from preventable outcomes such as stroke, myocardial infarction and renal failure are largely due to urbanisation, poverty and dysfunctional health systems working with limited budgets. These are some of the factors requiring in-depth study of the scientific aspects of alcohol intake in South Africa. Although there is enough evidence that links excessive drinking with hypertension and CVD, the markers of alcohol intake – self reporting of alcohol, gamma-glutamyltransferase (GGT) and carbohydrate deficient transferrin – are still not specific enough to isolate other confounding factors in the association of alcohol intake with CVD. The markers of alcohol that independently predict CVD and mortality need to be explored. Finally, the severe lack of longitudinal investigations on alcohol-related hypertension development and total mortality in black South Africans has compromised the early identification of risk factors associated with these outcomes. This study will therefore attempt to address the limited availability of longitudinal studies and stimulate interest for continued investigation. Aim The aim of this study was to investigate whether alcohol intake of black South Africans is related to specific measures of cardiovascular function (change in blood pressure (BP), hypertension development) and mortality over a period of 5 years. Methodology This study was based on the international Prospective Urban and Rural Epidemiology (PURE) study which includes 26 countries, investigating the cause and development of cardiovascular risk factors in low, middle and high income countries. This South African leg of the PURE study started in 2005 in which the baseline data was collected from 2021 black South Africans from rural and urban areas in Ikageng, Ganyesa and Tlakgameng in the North West Province. Eleven participants presented with missing data, leaving 2010 participants with complete datasets at baseline. However, data from these 11 participants was useful, especially for Chapter 4. All participants gave informed consent and the Ethics committee of the North-West University (Potchefstroom Campus) approved the study. The follow-up data collection was done in 2010. General health questionnaires, anthropometric measurements, lipid profiles and cardiovascular measurements were taken both at baseline and follow-up using appropriate methods. We also collected blood samples and performed biochemical analyses for lipid markers, liver enzymes, inflammatory markers and percentage carbohydrate deficient transferrin (%CDT). Finally, we obtained data on cardiovascular and non-cardiovascular mortality through verbal autopsy and death certificates. We made use of analysis of variance (ANOVA) and Chi-square tests to compare means and proportions, respectively. We used dependent t-tests and the McNemar test to compare baseline and follow-up variables. Furthermore, we employed single and partial linear regression analyses to correlate alcohol markers with each other and with the cardiovascular measures. Multiple regression analyses were used to correlate dependent variables in the study with various independent variables as required. Finally, we employed multivariable-adjusted Cox regression analyses to assess the association of the selected alcohol markers with mortality while adjusting for several independent variables. Results and Conclusions of each manuscript - With the first research article (Chapter 4), we aimed to compare self-reported alcohol intake estimates with GGT and %CDT, considering their relationship with percentage change in brachial blood pressure (BP) and central systolic blood pressure (cSBP) over 5 years. The results indicated that only self-reported alcohol intake independently predicted % change in brachial BP and cSBP. This was not found for the biochemical markers GGT and %CDT. Self-reported alcohol intake seems to be an important measure to implement by health systems in low income areas of sub-Saharan Africa, where honest reporting is expected. - Given the likely presence of high GGT levels in both alcohol consumption and non-alcoholic fatty liver disease (NAFLD), the second manuscript (Chapter 5) aimed to compare the cardiovascular and metabolic characteristics of excessive alcohol users and individuals with suspected NAFLD (confirmed with self-report, GGT and %CDT). We found that different sex and cardiometabolic profiles characterised excessive alcohol users and individuals suspected with NAFLD. Lean body mass and male sex were the dominant characteristics in excessive alcohol use while the NAFLD group had a dysmetabolic profile with obese women making up the higher proportion of this group. In excessive alcohol users systolic blood pressure and pulse pressure were independently associated with high-density lipoprotein cholesterol. Diastolic blood pressure showed a significant correlation with waist circumference. These disparate profiles may guide healthcare practitioners in primary healthcare clinics to identify individuals with elevated GGT levels who may suffer from NAFLD or alcohol overuse. These results emphasise the importance of modifiable risk factors as the main contributors to CVD and that lifestyle change should be the main focus in developing countries such as South Africa. - The third manuscript (Chapter 6) aimed to determine the measure of alcohol intake (selfreported alcohol intake, GGT and %CDT) that related best with hypertension development, cardiovascular and all-cause mortality over 5 years in the same population of black South Africans. We found that GGT was the only independent predictor of hypertension development, cardiovascular as well as all-cause mortality. Moreover, self-reporting of alcohol intake predicted incident hypertension, confirming our findings from Chapter 4. The third marker, %CDT, a highly specific marker of alcohol intake, was not related with any outcome variable, perhaps due to its low sensitivity. Although self-reported alcohol intake is useful in low-resource primary healthcare settings, measurement of GGT is encouraged due to its predictive value for hypertension and mortality. GGT represents alcohol intake, non-alcoholic steatohepatitis and obesity - all known to have severe cardiovascular consequences. Discussion and Conclusions Excessive alcohol intake remains a major concern in the development of hypertension, CVD and premature death in sub-Saharan Africa. Despite their weaknesses such as bias and nonspecificity, self-reporting of alcohol consumption and GGT emerged as reliable alcohol markers that independently predicted 5-year change in BP, hypertension development and total mortality in this population. Serum %CDT did not show any association with the mentioned cardiovascular markers. Finally, we were also able to show that black South Africans with suspected NAFLD (i.e. with high GGT levels who do not consume alcohol) are typically obese women, whereas lean men were more likely to have high alcohol consumption. Further prospective investigations are encouraged regarding (a) these mentioned associations, as well as (b) other self-reporting estimates such as quantity and frequency of drinking and (c) the use of %CDT as a highly specific marker of alcohol intake. The simultaneous presence of HIV infection in alcohol abuse in this population also warrants further investigation. / PhD (Physiology), North-West University, Potchefstroom Campus, 2015

The purpose of the study was to evaluate the IMB skills model for its relevance to the Ethiopian context. According to the model, adherence-related information and motivation work through adherence-related behavioural skills to affect adherence to ART. Quantitative, analytical, observational, cross-sectional, institution-based study was conducted to evaluate the model by assessing those patients who have and do not have the right information, motivation, and behavioural skills whether they have or do not have good adherence to ART. Data was collected using structured questionnaires where a total of 400 randomly selected participants provided data on adherence-related information, motivation and behavioral skills as well as adherence behavior per se. Data was analysed using the Statistical Package for Social Sciences (SPSS) version 20.0. Both descriptive and inferential statistics used in the study. Only 90.75 % of the total sample population reported ART adherence rate of more than or equal to 95% in this study. Free and restricted model tests, through bivariate and multivariate analyses, used to assess the propositions of the IMB model of ART adherence and provided support for the interrelations between the elements proposed by the model. The study has supported the applicability of the IMB model of adherence to the Ethiopian context highlighting its application in adherence-promotion intervention efforts. The findings revealed the need for on-going educational, informational and other interventions to address the knowledge, motivation and adherence behavioural skills of patients in order to improve the current levels of ART adherence behaviour. / Health Studies / D. Litt. et. Phil. (Health Studies)

Le fonctionnement du morphème polyfonctionnel d= en syriaque des Évangiles de la Peshiṭta est décrit morphosyn-taxiquement et dans une perspective typologique, en synchronie et par comparaison, pour certains points, avec d’autres versions et avec d’autres langues sémitiques. Ancien démonstratif en proto-sémitique, d= est un relateur à plusieurs niveaux syntaxiques : support de détermination, marqueur des relations génitivales et relatives, introducteur de complétives et adverbiales. Un critère syntaxique distingue sémantiquement les syntagmes génitivaux aliénables et inaliénable même si cette distinction tend à s’estomper, à des degrés différents selon les catégories sémantiques d’inaliénables, termes de parentés et parties du corps. Si les structures syntaxiques sont semblables pour les relatives restrictives et les non-restrictives, il est impossible de relativiser ces dernières pour les relatives objet, dative et adjointe. La stratégie à trou syntaxique constitue la stratégie primaire dans la hiérarchie d’accessibilité, et toutes les fonctions de la tête dans la matrice, sauf l’objet de comparaison, sont relativisables.Les fonctions de complémenteur de d= sont plus larges qu’en sémitique ancien. d= s’emploie avec 12 des 14 types de prédicats de la classification typologique de Noonan au lieu de 3 en sémitique ancien. Le syriaque est conforme à la hiérarchie implicationnelle, Complement Deranking-Argument Hierarchy. Le critère syntaxique de saturation de la valence verbale et des critères sémantiques et contextuels permettent de distinguer les adverbiales des complétives. Lorsque d= forme des locutions conjonctives avec des prépositions ou des adverbes, la polysémie des valeurs est désambigüisée grâce au contexte, à l’exception de la conjonction monosémique temporelle mo d=. Dans le discours rapporté, le discours Reproduit est majoritairement introduit sans d=, alors qu’il l’est avec d= pour le discours Reformulé. Le grec n’a eu aucune influence sur son utilisation. / The behaviour of the polyfunctional morpheme d= in the Syriac language of the Gospels of the Peshiṭta is described in a typological perspective, in synchrony and in comparison, for some properties, with other varieties of Syriac and with other Semitic languages. d=, a former demonstrative in Proto-Semitic, is a relator at various syntactic levels: a determination place-holder, a genitival and relative marker, an introducer of completive and adverbial clauses.Chapter 1 introduces the topic of the study and the theoretical framework and chapter 2 discusses the state of art.Chapter 3 is dedicated to the study of the genitive phrase. We discovered that a syntactic criterion allows to distinguish semantically between alienable and inalienable phrases, to different degrees according to the semantic categories of the inalienable set, i.e. kinship and body part terms.In chapter 4, we showed that even though syntactic structures are similar for restrictive and non-restrictive relative clauses, it is impossible to relativize the latter for object, dative and adjoint relatives clauses. The syntactic gap strategy constitutes the primary strategy in the accessibility hierarchy and all the functions of the head in the main clause, except the object of comparison, are relativizable.Chapter 5 deals with the functions of d= as a complementizer and an introducer of adverbial clauses. The complementizer uses are wider than in Old Semitic. d= is employed with 12 of the 14 types of predicates of Noonan's typological classification as against 3 in Old Semitic. Syriac conforms to the implicational hierarchy named Complement Deranking-Argument Hierarchy. For adverbial clauses, the syntactic criterion of saturation of the verbal valency as well as semantic and contextual criteria permit to distinguish adverbial clauses from complement clauses. When d= forms conjunctive locutions with prepositions or adverbs, the polysemy of values is disambiguated thanks to the context. Only the conjunction mo d= is monosemous (with a temporal meaning).In the last chapter (chap. 6) on reported speech, we showed that direct reported speech is mostly introduced without d=, whereas it is always the case for indirect speech. Greek did not have any influence on the use of d=. / Il funzionamento del morfema polifunzionale d= del siriaco nei Vangeli della Peshiṭta è descritto in una prospettiva tipologica, in sincronia e attraverso la comparazione, in alcuni punti, con altre versioni e con altre lingue semitiche. Tale morfema deriva dal pronome dimostrativo proto-semitico *ḏV e funge da relatore in diversi livelli sintattici: supporto di determinazione, indicatore delle relazioni genitivali e relative, introduttore delle proposizioni completive e avverbiali. Il capitolo 1 introduce la problematica della ricerca e dell’ambito teorico; il capitolo 2 presenta uno stato dell’arte sull’argomento.Il capitolo 3 è consacrato allo studio del sintagma genitivale. Un criterio sintattico ha permesso di distinguere semanticamente i sintagmi genitivali alienabili e inalienabili, anche se questa distinzione tende ad attenuarsi, in diversi punti, secondo le categorie semantiche d’inalienabile, termini di parentela e parti del corpo.Nel capitolo 4 abbiamo mostrato che, se da una parte le strutture sintattiche sono simili per le relative restrittive e le non-restrittive, dall’altra è impossibile relativizzare le ultime per le relative complemento oggetto, complemento di termine e complemento circostanziale. La strategia con gap sintattico è la strategia primaria nella gerarchia d’accessibilità e tutte le funzioni della testa nella proposizione matrice, tranne l’oggetto della comparazione, sono relativizzabili.Il capitolo 5 tratta le funzioni di d= come complementatore e introduttore di proposizioni avverbiali. Gli impieghi del complementatore sono più ampi rispetto al semitico più antico. Il morfema d= si impiega infatti per 12 dei 14 tipi di predicato della classificazione tipologica di Noonan invece dei 3 del semitico più antico. Il siriaco è conforme alla gerarchia implicativa, Complement Deranking-Argument Hierarchy. Per le proposizioni avverbiali, il criterio sintattico di saturazione della valenza verbale e dei criteri semantici e contestuali permettono di distinguere le avverbiali dalle completive.Quando d= forma delle locuzioni congiuntive con delle preposizioni o degli avverbi, la polisemia dei valori è disambiguato grazie al contesto, fatta eccezione della congiunzione monosemica temporale mo d=Nell’ultimo capitolo, sul discorso indiretto, mostriamo che il discorso “riprodotto” è introdotto senza d= nella maggior parte dei casi, mentre si usa d= nel discorso “riformulato”. Il greco non ha alcuna influenza sul suo utilizzo.

Le fonctionnement du morphème polyfonctionnel d= en syriaque des Évangiles de la Peshiṭta est décrit morphosyn-taxiquement et dans une perspective typologique, en synchronie et par comparaison, pour certains points, avec d’autres versions et avec d’autres langues sémitiques. Ancien démonstratif en proto-sémitique, d= est un relateur à plusieurs niveaux syntaxiques : support de détermination, marqueur des relations génitivales et relatives, introducteur de complétives et adverbiales. Un critère syntaxique distingue sémantiquement les syntagmes génitivaux aliénables et inaliénable même si cette distinction tend à s’estomper, à des degrés différents selon les catégories sémantiques d’inaliénables, termes de parentés et parties du corps. Si les structures syntaxiques sont semblables pour les relatives restrictives et les non-restrictives, il est impossible de relativiser ces dernières pour les relatives objet, dative et adjointe. La stratégie à trou syntaxique constitue la stratégie primaire dans la hiérarchie d’accessibilité, et toutes les fonctions de la tête dans la matrice, sauf l’objet de comparaison, sont relativisables.Les fonctions de complémenteur de d= sont plus larges qu’en sémitique ancien. d= s’emploie avec 12 des 14 types de prédicats de la classification typologique de Noonan au lieu de 3 en sémitique ancien. Le syriaque est conforme à la hiérarchie implicationnelle, Complement Deranking-Argument Hierarchy. Le critère syntaxique de saturation de la valence verbale et des critères sémantiques et contextuels permettent de distinguer les adverbiales des complétives. Lorsque d= forme des locutions conjonctives avec des prépositions ou des adverbes, la polysémie des valeurs est désambigüisée grâce au contexte, à l’exception de la conjonction monosémique temporelle mo d=. Dans le discours rapporté, le discours Reproduit est majoritairement introduit sans d=, alors qu’il l’est avec d= pour le discours Reformulé. Le grec n’a eu aucune influence sur son utilisation. / The behaviour of the polyfunctional morpheme d= in the Syriac language of the Gospels of the Peshiṭta is described in a typological perspective, in synchrony and in comparison, for some properties, with other varieties of Syriac and with other Semitic languages. d=, a former demonstrative in Proto-Semitic, is a relator at various syntactic levels: a determination place-holder, a genitival and relative marker, an introducer of completive and adverbial clauses.Chapter 1 introduces the topic of the study and the theoretical framework and chapter 2 discusses the state of art.Chapter 3 is dedicated to the study of the genitive phrase. We discovered that a syntactic criterion allows to distinguish semantically between alienable and inalienable phrases, to different degrees according to the semantic categories of the inalienable set, i.e. kinship and body part terms.In chapter 4, we showed that even though syntactic structures are similar for restrictive and non-restrictive relative clauses, it is impossible to relativize the latter for object, dative and adjoint relatives clauses. The syntactic gap strategy constitutes the primary strategy in the accessibility hierarchy and all the functions of the head in the main clause, except the object of comparison, are relativizable.Chapter 5 deals with the functions of d= as a complementizer and an introducer of adverbial clauses. The complementizer uses are wider than in Old Semitic. d= is employed with 12 of the 14 types of predicates of Noonan's typological classification as against 3 in Old Semitic. Syriac conforms to the implicational hierarchy named Complement Deranking-Argument Hierarchy. For adverbial clauses, the syntactic criterion of saturation of the verbal valency as well as semantic and contextual criteria permit to distinguish adverbial clauses from complement clauses. When d= forms conjunctive locutions with prepositions or adverbs, the polysemy of values is disambiguated thanks to the context. Only the conjunction mo d= is monosemous (with a temporal meaning).In the last chapter (chap. 6) on reported speech, we showed that direct reported speech is mostly introduced without d=, whereas it is always the case for indirect speech. Greek did not have any influence on the use of d=. / Il funzionamento del morfema polifunzionale d= del siriaco nei Vangeli della Peshiṭta è descritto in una prospettiva tipologica, in sincronia e attraverso la comparazione, in alcuni punti, con altre versioni e con altre lingue semitiche. Tale morfema deriva dal pronome dimostrativo proto-semitico *ḏV e funge da relatore in diversi livelli sintattici: supporto di determinazione, indicatore delle relazioni genitivali e relative, introduttore delle proposizioni completive e avverbiali. Il capitolo 1 introduce la problematica della ricerca e dell’ambito teorico; il capitolo 2 presenta uno stato dell’arte sull’argomento.Il capitolo 3 è consacrato allo studio del sintagma genitivale. Un criterio sintattico ha permesso di distinguere semanticamente i sintagmi genitivali alienabili e inalienabili, anche se questa distinzione tende ad attenuarsi, in diversi punti, secondo le categorie semantiche d’inalienabile, termini di parentela e parti del corpo.Nel capitolo 4 abbiamo mostrato che, se da una parte le strutture sintattiche sono simili per le relative restrittive e le non-restrittive, dall’altra è impossibile relativizzare le ultime per le relative complemento oggetto, complemento di termine e complemento circostanziale. La strategia con gap sintattico è la strategia primaria nella gerarchia d’accessibilità e tutte le funzioni della testa nella proposizione matrice, tranne l’oggetto della comparazione, sono relativizzabili.Il capitolo 5 tratta le funzioni di d= come complementatore e introduttore di proposizioni avverbiali. Gli impieghi del complementatore sono più ampi rispetto al semitico più antico. Il morfema d= si impiega infatti per 12 dei 14 tipi di predicato della classificazione tipologica di Noonan invece dei 3 del semitico più antico. Il siriaco è conforme alla gerarchia implicativa, Complement Deranking-Argument Hierarchy. Per le proposizioni avverbiali, il criterio sintattico di saturazione della valenza verbale e dei criteri semantici e contestuali permettono di distinguere le avverbiali dalle completive.Quando d= forma delle locuzioni congiuntive con delle preposizioni o degli avverbi, la polisemia dei valori è disambiguato grazie al contesto, fatta eccezione della congiunzione monosemica temporale mo d=Nell’ultimo capitolo, sul discorso indiretto, mostriamo che il discorso “riprodotto” è introdotto senza d= nella maggior parte dei casi, mentre si usa d= nel discorso “riformulato”. Il greco non ha alcuna influenza sul suo utilizzo.

I performed two randomised controlled trials (RCTs) to determine the best treatments for enteric fever in Kathmandu, Nepal, an area with a high proportion of nalidixic acid resistant S. Typhi and S. Paratyphi A isolates. I recruited 844 patients with suspected enteric fever to compare chloramphenicol versus gatifloxacin. 352 patients were culture confirmed. 14/175 patients treated with chloramphenicol and 12/177 patients treated with gatifloxacin experienced treatment failure (HR=0.86 (95% CI 0.40 to 1.86), p=0.70). The median times to fever clearance were 3.95 and 3.90 days, respectively (HR=1.06 [CI 0.86 to 1.32], p=0.59). The second RCT compared ofloxacin versus gatifloxacin and recruited 627 patients. Of the 170 patients infected with nalidixic acid resistant strains, the number of patients with treatment failure was 6/83 in the ofloxacin group and 5/87 in the gatifloxacin group (Hazard Ratio, HR=0.81, 95% CI 0.25 to 2.65; p=0.73); the median times to fever clearance were 4.7 and 3.3 days respectively (HR=1.59 [CI 1.16 to 2.18], p=0.004). I compared conventional blood culture against an electricity free culture approach. 66 of 304 patients with suspected enteric fever were positive for S. Typhi or S. Paratyphi A, 55 (85%) isolates were identified by the conventional blood culture and 60 (92%) isolates were identified by the experimental method. The percentages of positive and negative agreement for diagnosis of enteric fever were 90.9% and 96.0%, respectively. This electricity free blood culture system may have utility in resource-limited settings or potentially in disaster relief and refugee camps. I performed a literature review of RCTs of enteric fever which showed that trial design varied greatly. I was interested in the perspective of patients and what they regarded as cure. 1,481 patients were interviewed at the start of treatment, 860 (58%) reported that the resolution of fever would mean cure to them. At the completion of treatment, 877/1,448 (60.6%) reported that they felt cured when fever was completely gone. We suggest that fever clearance time is the best surrogate for clinical cure in patients with enteric fever and should be used as the primary outcome in future RCTs for the treatment of enteric fever.

Les adolescents qui décrochent de l’école secondaire arrivent difficilement à s’intégrer dans une économie axée sur le savoir et éprouvent plusieurs problèmes d’ajustement à l’adolescence et à l’âge adulte. Pour prévenir le décrochage scolaire, une étape cruciale consiste à dépister efficacement les élèves les plus à risque. Deux formes de dépistage axées sur des données peuvent être utilisées en milieu scolaire: une forme utilisant des informations auto-rapportées par les élèves à partir de questionnaires, et une autre fondée sur des informations administratives consignées au dossier des élèves. Toutefois, à notre connaissance, l’efficacité de ces différentes modalités n’a jamais été comparée directement. De plus, il est possible que l’efficacité relative de ces outils de dépistage soit différente selon le sexe de l’élève. Cette étude vise à comparer différents outils de dépistage pour prédire le décrochage scolaire, en tenant compte de l’effet modérateur du sexe. Les outils utilisés seront a) un questionnaire auto-rapporté validé (Archambault et Janosz, 2009) et b) un outil conçu à l’aide de données administratives, créé par une commission scolaire du Québec. La comparaison de ces outils est effectuée en termes de qualités psychométriques et d’aspect pratique pour le milieu scolaire. Pour ce faire, un échantillon de 1557 élèves (50% de garçons), âgé entre 14 et 18 ans est utilisé. Les résultats indiquent que l’indice administratif possède une capacité discriminante adéquate, mais inférieure à celle de l’indice auto-rapportée, jugée excellente. L’effet modérateur du sexe n’a pas été confirmé. Les avantages et inconvénients respectifs de ces deux modes de dépistage sont discutés. / Adolescents who drop out of high school often struggle in a knowledge-based, technologically complex society and experience many adjustment problems in adolescence and adulthood. To prevent dropout, a crucial first step consists of correctly identifying students who are most at risk. Schools can rely on two types of screening tools to identify potential dropouts: based on self–reported questionnaire or based on administrative data. However, the predictive value of risk indices based on self-reported data has never been directly compared to that of indices calculated from administrative data. Moreover, the relative efficacy of different screening tools could be different according to the student gender. The purpose of this study is to compare the efficacy, for boys and girls, of two risk indices for high school dropout: a) a validated index based on a self-reported questionnaire (Archambault & Janosz, 2009) and b) an index based on administrative data and designed by a Quebec school board. The comparison of these two screening methods is made in term of psychometric properties and practicality for school practitioners. To do so, a sample of 1557 students (50% boys), between 14 and 18 years old is used. The results show that the administrative index has an adequate discriminant capacity, but that is lower than that of the self-reported index judged excellent. These results were not moderated by gender. The comparative advantages and disadvantages of both screening methods are discussed.

O objetivo desta tese foi identificar, analisar e sintetizar evidências científicas quanto à eficácia do Tratamento Convencional (TC) e do Tratamento Restaurador Atraumático (ART) em molares decíduos considerando os parâmetros de longevidade, custo, aceitabilidade e desfechos reportados pelo paciente. Este volume apresenta um compilado de uma revisão sistemática e dois ensaios clínicos randomizados orientados pelas recomendações PRISMA, CONSORT-PRO e CHEERS. Realizamos uma revisão sistemática, cuja pesquisa bibliográfica buscou por desfechos reportados pelo paciente (PROs) em relação a tratamentos restauradores na dentição decídua em estudos prospectivos indexados no PubMed, Scopus e OpenGrey até fevereiro de 2018. Meta-análise de Comparação de Tratamento Misto foi realizada considerando os resultados dos estudos revisados. Devido à incompatibilidade de dados, apenas dor, ansiedade e a qualidade de vida relacionada à saúde bucal (QVRSB) foram analisados estatisticamente. Adicionalmente, delineamos um ensaio clínico de não inferioridade randomizado por cluster cujo desfecho primário foi a longevidade do TC e do ART em cavidadades oclusais e oclusoproximais de molares decíduos após 24 meses de acompanhamento. Como desfechos secundários, foram avaliados a aceitabilidade, representada pelo desconforto auto relatado e cooperação dos participantes, e a custo-eficácia. Crianças entre 3 e 6 anos apresentando pelo menos uma cavidade oclusal ou oclusoproximal em molares decíduos foram randomizadas entre os grupos convencional (TC) com resina composta sob isolamento absoluto após anestesia local ou o ART. Em ambos os grupos foi realizada remoção parcial de tecido cariado. O estudo foi desenvolvido em um trailer odontológico localizado no interior de uma escola municipal de Barueri/SP. Imediatamente após cada sessão restauradora, foi mensurado o desconforto auto relatado pelo participante e o comportamento pelo operador através da Escala Facial de Wong-Baker e escala de 5 pontos, respectivamente. As restaurações oclusais e oclusoproximais foram avaliadas após 6, 12, 18 e 24 meses segundo o critério de Frencken et al. 1998 e de Roeleveld et al. 2006. Os custos profissional e do procedimento, subdividido em custos variável e do material de consumo, foram considerados para estimar o custo incremental dos tratamentos. Regressão de Poisson em análise de multinível, Regressão de Cox com fragilidade compartilhada e regressão de Bootstrap foram utilizadas para testar o desconforto e cooperação, a sobrevida e o custo entre os tratamentos e outras variáveis independentes. O nível de significância foi ajustado em 5%. Tratamentos restauradores que utilizam agentes quimicomecânicos ou apenas instrumentos manuais para remoção de tecido cariado, como o ART, resultaram em menores índices de ansiedade e menor tendência em provocar dor em crianças do que tratamentos que utilizam instrumentos rotatórios e/ou anestesia. A QVRSB não foi influenciada pelos tratamentos restauradores nem diferiu entre os grupos. Por outro lado, o TC e o ART apresentaram baixa taxa de desconforto e alto índice de cooperação sem diferença estatisticamente significante entre eles. Ambos os tratamentos apresentaram sobrevida semelhante nas cavidades oclusais após 24 meses de acompanhamento, porém o custo do ART foi menor, demonstrando ser a opção mais custo-eficaz. Em relação às restaurações oclusoproximais, apesar do desempenho do TC ser superior ao ART, ele apresentou maior custo. Pode-se concluir que ansiedade e dor estão diretamente relacionados com tratamentos restauradores mais invasivos. Por outro lado, o TC e o ART apresentaram desconforto, cooperação e custo-eficácia semelhantes no tratamento de molares decíduos. Palavras-chave: Restauração dentária permanente. Dente decíduo. Longevidade. Custos e análises de custo. Medidas de resultados relatados pelo paciente. Ensaio clínico. Revisão. Odontopediatria. / The aim of this thesis was to identify, analyze and synthesize scientific evidences regarding the efficacy of conventional treatment (CT) and Atraumatic Restorative Treatment (ART) in primary molars considering the parameters of longevity, cost, acceptability and patient reported outcomes. This volume presents a compilation of a systematic review and two randomized clinical trials guided by the PRISMA, CONSORT-PRO and CHEERS recommendations. We performed a systematic review whose literature search included outcomes reported by the pediatric patient (PROs) related to restorative treatments in the primary dentition in prospective studies indexed in PubMed, Scopus and OpenGrey until February 2018. Mixed Treatment Comparison analysis was performed considering the results of the reviewed studies. Due to data incompatibility, only pain, anxiety and oral health related to quality of life (OHRQoL) were analyzed statistically. Additionally, we designed a cluster randomized non-inferiority clinical trial whose primary outcome was the longevity of occlusal and occlusoproximal ART and TC restorations in primary molars after 24 months of follow-up. As secondary outcomes, acceptability, represented by self-reported discomfort and participant cooperation, and cost-efficacy were assessed. Children between 3 and 6 years old presenting at least one occlusal or occlusoproximal cavity in primary molars were randomized between the conventional group (TC) with composite resin under rubber dam isolation after local anesthesia or ART. Partial removal of carious tissue was performed in both groups. The study was carried out in a dental trailer located inside a municipal school in Barueri/SP. Immediately after each restorative session, self-reported discomfort and participant cooperation were measure through the Wong-Baker Facial Scale and a 5-point scale, respectively. The occlusal and occlusoproximal restorations were evaluated after 6, 12, 18 and 24 months according to the criteria of Frencken et al. 1998 and Roeleveld et al. 2006. The professional and procedural costs, subdivided into variable and material of consumption\' costs, were considered to estimate treatments\' incremental cost. Poisson regression in multilevel analysis, Cox regression with shared fragility and Bootstrap regression were used to test discomfort and cooperation, survival and cost between treatments and among others independent variables. The level of significance was set at 5%. Restorative treatments using chemomechanical agents or only manual instruments, such as ART, were related to lower anxiety rates and less tendency to provoke pain in children compared to treatments using rotary instruments and/or anesthesia. OHRQoL was not influenced by restorative treatments nor differed between groups. On the other hand, CT and ART presented low rate of discomfort and a high cooperation with no statistically significant difference between them. Both treatments presented similar survival rates in occlusal restorations after 24 months of follow-up, but the ART cost was lower, proving to be the most cost-effective option. Regarding occlusoproximal restorations, although CT performance was superior to ART, it presented a higher cost. It can be concluded that anxiety and pain are directly related to more invasive restorative treatments. Differently, CT and ART have similar discomfort, cooperation and cost-efficacy in the treatment of primary molars.

The aim of this project was to investigate what information is critical to convey to nurses when performing digital triage. In addition, the project aimed to investigate how such information could be visualized. This was done through a combined user research and machine learning approach, which enabled for a more nuanced and thorough investigation compared to only making use of one of the two fields. There is sparse research investigating how digital triaging can be improved and made more efficient. Therefore, this study has contributed with new and relevant insights. Three machine learning algorithms were implemented to predict the right level of care for a patient. Out of these three, the random forest classifier proved to have the best performance with an accuracy of 69.46%, also having the shortest execution time. Evaluating the random forest classifier, the most important features were stated to be the duration and progress of the symptoms, allergies to medicine, chronic diseases and the patient's own estimation of his/her health. These factors could all be confirmed by the user research approach, indicating that the results from the approaches were aligned. The results from the user research approach also showed that the patients' own description of their symptoms was of great importance. These findings served as a basis for a number of visualization decisions, aiming to make the triage process as accurate and efficient as possible.