Personalised medicine describes the practice of tailoring medical care to the individual characteristics of each patient. Fundamental to this practice is the identification of markers associated with differential treatment response. Such markers can be identified through the assessment of treatment effect modification using statistical methods. Randomised controlled trials provide the optimal setting for evaluating differential response to treatment. Due to restrictions regarding sample size, study length and ethics, observational studies are more appropriate in many circumstances, particularly for the identification of markers associated with adverse side-effects and long term response to treatments. However, the analysis of observational data raises some additional challenges. The overall aim of this thesis was to develop statistical methodology for the analysis of observational data, specifically primary care databases, to identify and evaluate markers associated with differential treatment response. Three aspects of the assessment of treatment effect modification in an observational setting were addressed. The first aspect related to the assessment of treatment effect modification on the additive measurement scale which corresponds to a comparison of absolute treatment effects across patient subgroups. Various ways in which this can be assessed in an observational setting were reviewed and a novel measure, the ratio of absolute effects, which can be calculated from certain multiplicative regression models, was proposed. The second aspect regarded the confounding adjustment and it was investigated how the presence of interactions between the moderator and confounders on both treatment receipt and outcome can bias estimates of treatment effect modification if unaccounted for using Monte Carlo simulations. It was determined that the presence of bias differed across different confounding adjustment methods and, in the majority of settings, the bias was reduced when the interactions between the moderator and confounders were accounted for in the confounding adjustment model. Thirdly, it has been proposed that patient data in observational studies be organised into and analysed as series of nested nonrandomised trials. This thesis extended this study design to evaluate predictive markers of differential treatment response and explored the benefits of this methodology for this purpose. It was suggested how absolute treatment effect estimates can be estimated and compared across patient subgroups in this setting. A dataset comprising primary care medical records of adults with rheumatoid arthritis was used throughout this thesis. Interest lay in the identification of characteristics predictive of the onset of type II diabetes associated with steroid (glucocorticoid) therapy. The analysis in this thesis suggested older age may be associated with a higher risk of steroid-associated type II diabetes, but this warrants further investigation. Overall, this thesis demonstrates how observational studies can be analysed such that accurate and meaningful conclusions are made within personalised medicine research.

京都大学 / 新制・課程博士 / 博士(医学) / 甲第23097号 / 医博第4724号 / 新制||医||1050(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 近藤 尚己, 教授 稲垣 暢也, 教授 石見 拓 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

Background: Many randomized controlled trials (RCTs) and meta-analyses include analyses of effect modification (also known as subgroup, interaction, or moderation analyses). Methodologists have widely acknowledged the challenges in deciding whether an apparent effect modification is credible or likely the result of chance or bias. Various sets of credibility criteria are available (Chapter 2 provides an example) but are inconsistent, vague in wording, lack guidance for deciding on overall credibility, and have not been systematically tested. Objective: To systematically develop a formal instrument to assess the credibility of effect modification analyses (ICEMAN) in RCTs and meta-analyses of RCTs. Methods: Key steps in the development process included 1) a systematic survey of the literature to identify available criteria, rationales, and previous instruments, 2) a formal consensus study among 10 leading experts, and 3) a formal user-testing study to refine the instrument based on interviews with trial investigators, systematic reviewer authors, and journal editors who applied drafts of the instrument to published claims of effect modification. Results: The systematic survey identified 150 relevant publications, 36 candidate credibility criteria with associated rationales, and 30 existing checklists (Chapter 3). The consensus study consisted of two main video conferences and multiple rounds of written discussion. The user-testing involved 17 users (including systematic review authors, trial investigators, and journal editors) who suggested substantial improvements based on detailed interviews. The final instrument provides separate versions for RCTs (five core questions) and meta-analyses (eight core questions) with explicit response options, and an overall credibility rating ranging from very low to high credibility. A detailed manual provides rationales, supporting references, examples from the literature, and suggestions for use in combination with other quality appraisal tools and reporting (Chapter 4). Discussion: ICEMAN is a rigorously developed instrument to evaluate claims of effect modification and addresses the main limitations of previous approaches. / Thesis / Doctor of Philosophy (PhD) / Randomized controlled trials and meta-analyses provide the best available evidence to evaluate whether effects of a therapy vary among individual patients. Efforts to decide whether treatment effects differ across patients are important and frequently done but difficult to interpret. The fundamental challenge is to decide whether apparent differences in effect are real or due to chance. To aid this decision, experts have suggested various sets of credibility criteria, all with important limitations. This thesis documents how we systematically addressed the limitations of previous approaches. Key steps were a systematic survey of the available credibility criteria, a consensus study among leading methodologists, and a formal user-testing study. The result is a new instrument for assessing the credibility of effect modification analyses (ICEMAN).

Background: Black South African women are known to have a high usage rate of injectable contraceptives. Breast cancer is the second leading cancer after malignant cervical neoplasms in black South African women. There is evidence that sex hormones are associated with an increased risk of developing breast cancer. In the Western Cape, investigators suggested that injectable contraceptives, more specifically DMPA, may increase breast cancer risk. In another study conducted in the same province, a weak association between breast cancer and women taking combined oestrogen/progesterone oral contraceptives was found, though no risk associated with injectable progestogen contraceptives (DMPA) was confirmed. Study Objective: This study aimed to determine whether there is an association between hormonal contraceptive use and an increased risk of cancer of the breast. Methods: Data was obtained from an ongoing case control study set up by MRC/Wits/NHLS Cancer Epidemiology Research Group (CERG) in 1995 to investigate risk factors associated with cancer among the black population in Johannesburg. Data was processed using STATA, version8 and analysed using univariate, bivariate and multivariate unmatched logistic regression models. Results: There was evidence that an overall use of oral contraceptives increases the risk of breast cancer; cases (n= 221), controls :( n= 153), OR=2.01 (95% CI:1.45, 2.80), p<0.0001. There was evidence of an association between use of injectable contraception and the risk of breast cancer; cases (n=244), controls (n=202), OR=1.51(CI: 1.14, 2.01),p=0.004 Surprisingly, no other use characteristic of either hormonal contraceptive method was statistically significantly associated with the risk of breast cancer in our dataset. The combined use of both oral and injectable contraception was associated with an increased risk of breast cancer, OR=1.68(1.21, 2.33), p =0.002. There was a strong effect modification (interaction) between oral contraceptive use and injectable progesterone associated with the risk of breast cancer, (p=0.008). Conclusion: After adjusting for all potential risk and confounding factors, as collected in the dataset, there was evidence of an association between combined oral contraceptive use and breast cancer. An association between cancer of the breast and overall use of injectable progesterone use was also established. There was evidence of association between the use of both hormonal contraceptive methods and an increased risk of breast cancer. However, whether these findings reflect the reality in terms of causal relationship or are the result of bias must be ascertained.

<p> Asthma is a multifactorial chronic disease that has shown a marked increase in prevalence over the past few decades, both in Canada and worldwide. Basic knowledge gaps remain about the pathways through which risk factors influence adult asthma. More adult women than men have asthma, and a growing body of research suggests that associations between certain risk factors and asthma may differ by sex. The aim of this thesis was to investigate the socioeconomic, environmental and personal correlates of asthma in men and women.</p> <p> Data for this thesis were obtained from a cross-sectional study conducted in 2003 in the rural Canadian town of Humboldt, Saskatchewan. The survey response rate was 71% of the resident target population, with 1177 females and 913 males aged 18 to 79 participating in the study. Researchers collected objective data on atopy (skin prick test), and body mass index. Exposures and history of physician-diagnosed asthma in the past year (current asthma) and during the participants lifetime (ever asthma) were self-reported. Multivariable logistic regression models adjusted for age, atopy, and parental asthma history were used to evaluate associations of correlates with asthma. The model building process was based on a conceptual framework of three categories: socioeconomic variables, home and work environment, and personal factors. </p> <p> The prevalence of asthma was higher in women than men (ever asthma: 10.2% of women versus 5.8% of men; current asthma: 6.2% of women versus 2.8% of men). The logistic regression models for ever asthma and current asthma showed several sex differences. The sequential addition of each category of socioeconomic, environmental, and personal variables contributed significantly to model fit in women, but not in men. Living in a mobile, attached or multiple-family home, household dampness, and overweight/obesity were strong risk factors for female asthma, while farm living, occupational grain dust exposure, and regular alcohol use emerged as protective factors. Male models revealed a strong significant association between household dampness and current asthma. A significant interaction between home type and age was found only in females. Women living in homes other than single-family detached dwellings were more likely to have asthma, an association that decreased in strength with increasing age. </p> <p> These results suggest that several risk factors for adult asthma may be sex-specific, therefore emphasizing the importance of considering sex as a potential effect modifier in future adult asthma epidemiology studies.</p>

<p> Asthma is a multifactorial chronic disease that has shown a marked increase in prevalence over the past few decades, both in Canada and worldwide. Basic knowledge gaps remain about the pathways through which risk factors influence adult asthma. More adult women than men have asthma, and a growing body of research suggests that associations between certain risk factors and asthma may differ by sex. The aim of this thesis was to investigate the socioeconomic, environmental and personal correlates of asthma in men and women.</p> <p> Data for this thesis were obtained from a cross-sectional study conducted in 2003 in the rural Canadian town of Humboldt, Saskatchewan. The survey response rate was 71% of the resident target population, with 1177 females and 913 males aged 18 to 79 participating in the study. Researchers collected objective data on atopy (skin prick test), and body mass index. Exposures and history of physician-diagnosed asthma in the past year (current asthma) and during the participants lifetime (ever asthma) were self-reported. Multivariable logistic regression models adjusted for age, atopy, and parental asthma history were used to evaluate associations of correlates with asthma. The model building process was based on a conceptual framework of three categories: socioeconomic variables, home and work environment, and personal factors. </p> <p> The prevalence of asthma was higher in women than men (ever asthma: 10.2% of women versus 5.8% of men; current asthma: 6.2% of women versus 2.8% of men). The logistic regression models for ever asthma and current asthma showed several sex differences. The sequential addition of each category of socioeconomic, environmental, and personal variables contributed significantly to model fit in women, but not in men. Living in a mobile, attached or multiple-family home, household dampness, and overweight/obesity were strong risk factors for female asthma, while farm living, occupational grain dust exposure, and regular alcohol use emerged as protective factors. Male models revealed a strong significant association between household dampness and current asthma. A significant interaction between home type and age was found only in females. Women living in homes other than single-family detached dwellings were more likely to have asthma, an association that decreased in strength with increasing age. </p> <p> These results suggest that several risk factors for adult asthma may be sex-specific, therefore emphasizing the importance of considering sex as a potential effect modifier in future adult asthma epidemiology studies.</p>

To help mitigate the escalating prevalence of Type 2 Diabetes (T2D) and alleviate society of its associated morbidity and economic burden on health care, it is crucial to understand its aetiology. Both genetic and the environmental risk factors are known to be involved. Healthy diets have been proven to reduce the risk of T2D in primary prevention trials, however, which components and exact mechanisms are involved is not fully understood, in particular, the role of macronutrient intake. Body weight, glycaemic markers and T2D are all to some extent genetically regulated. There may also be genetic influences on how people digest, absorb or metabolise macronutrients. This poses the possibility that the interplay between genes and our diet may help us unravel T2D’s aetiology. The aim of this PhD was to investigate gene-diet interactions on the risk of incident T2D, focusing primarily on macronutrient intake as the dietary factor. First, I systematically evaluated the current evidence before taking a step-wise approach (hypothesis driven to hypothesis-free) to interrogate gene-macronutrient interactions. This identified 13 publications, with 8 unique interactions reported between macronutrients (carbohydrate, fat, saturated fat, dietary fibre, and glycaemic load derived from self-report of dietary intake and circulating n-3 polyunsaturated fatty acids) and genetic variants in or near TCF7L2, GIPR, CAV2 and PEPD (p < 0.05) on T2D. All studies were observational with moderate to serious risk of bias and limitations that included lack of adequate adjustment for confounders, lack of reported replication and insufficient correction for multiple testing. Second, these reported interactions did not replicate in a large European multi-centre prospective T2D case-cohort study called EPIC-InterAct. We concluded that the heterogeneity between our results and those published could be explained by methodological differences in dietary measurement, population under study, study design and analysis but also by the possibility of spurious interactions. Third, given the paucity of gene-macronutrient interaction research using genetic risk scores (GRS), we examined the interaction between three GRS (for BMI (97 SNPs), insulin resistance (53 SNPs) and T2D (48 SNPs)) and macronutrient intake (quantity and quality indicators) in EPIC-InterAct. We did not identify any statistically significant interactions that passed multiple testing corrections (p≥0.20, with a p value threshold for rejecting the null hypothesis of 0.0015 (based on 0.05/33 tests)). We also examined 15 foods and beverages identified as being associated with T2D, and no significant interactions were detected. Lastly, we applied a hypothesis-free method to examine gene-macronutrient interactions and T2D risk by using a genome-environment-wide-interaction-study. Preliminary findings showed no significant interactions for total carbohydrate, protein, saturated fat, polyunsaturated fat and cereal fibre intake on T2D. In conclusion, the consistently null findings in this thesis using a range of statistical approaches to examine interactions between genetic variants and macronutrient intake on the risk of developing T2D have two key implications. One, based on the specific interactions examined, this research does not confirm evidence for gene-diet interactions in the aetiology of T2D and two, this research suggests that the association between macronutrient intake and the risk of developing T2D does not differ by genotype.

Introduction Although studies have examined the association between childhood asthma and parental smoking and secondhand smoke, little research has explored the moderating role of birth weight and prematurity (BWP) status on this association. We examined the association between secondhand smoke exposure, asthma, and asthma severity in children aged six to 17 as well as the modifying effect of BWP on parental smoking and asthma. Methods We used data from 36,954 children from the National Survey of Children's Health 2017-2018. In addition to univariate analysis, adjusted and unadjusted logistic regression models were used to estimate the effect of secondhand smoke on asthma. The interaction term between parental smoking and BWP was tested. Multinomial regression was used to evaluate the association between secondhand smoke on asthma severity. Results About 15.1 % of children had asthma and 15.4% of parents reported smoking. Odds of asthma were higher in children living with an outdoor (AOR, 1.27; 95% CI, 1.06-1.52) and indoor (AOR, 1.46; 95% CI, 1.01-2.11) smoker in the adjusted model. The association of parental smoking with asthma differed by birth weight and premature status. Normal weight children who are premature had the highest odds ratio (AOR, 2.15; 95% CI, 1.2-3.86). In the multinomial model, low birth weight and premature children had higher odds of mild (AOR, 1.90; 95% CI, 1.40-2.56) and moderate/severe (AOR, 1.81; 95% CI, 1.16-2.84) asthma compared to the no asthma group. Conclusion The Association of parental smoking on asthma was modified by BWP. Focused asthma interventions in children should inquire about BWP status as well as parental smoking and household smoke exposure to reduce asthma morbidity and mortality.

BACKGROUND: Previous studies show a positive association between household wealth and overweight in sub-Saharan African (SSA) countries; however, the manner in which this relationship differs in the presence of educational attainment has not been well-established. This study examined the multiplicative effect modification of educational attainment on the association between middle-income and rich household wealth and overweight status among adult females in 22 SSA countries. We hypothesized that household wealth was associated with a greater likelihood of being overweight among middle income and rich women with lower levels of educational attainment compared to those with higher levels of educational attainment. METHODS: Demographic and Health Survey (DHS) data from 2006 to 2016 for women aged 18-49 years in SSA countries were used for the study. Overweight was defined as a body mass index (BMI) ≥ 25 kg/m2. Household wealth index tertile was the exposure and educational attainment, the effect modifier. Potential confounders included age, ethnicity, place of residence, and parity. Descriptive analysis was conducted, and separate logistic regression models were fitted for each of the 22 SSA countries to compute measures of effect modification and 95% confidence intervals. Analysis of credibility (AnCred) methods were applied to assess the intrinsic credibility of the study findings and guide statistical inference. RESULTS: The prevalence of overweight ranged from 12.6% in Chad to 56.6% in Swaziland. Eighteen of the 22 SSA countries had measures of effect modification below one in at least one wealth tertile. This included eight of the 12 low-income countries and all 10 middle income countries. This implied that the odds of overweight were greater among middle-income and rich women with lower levels of educational attainment than those with higher educational attainment. On the basis of the AnCred analysis, it was found that the majority of the study findings across the region provided some support for the study hypothesis. CONCLUSIONS: Women in higher wealth strata and with lower levels of educational attainment appear to be more vulnerable to overweight compared to those in the same wealth strata but with higher levels of educational attainment in most low- and middle- income SSA countries.

Exposé de la situation : Des études menées sur les animaux démontrent que le système endocannabinoide est important dans le maintien de l’homéostasie de l’énergie et que les effets de sa modulation sont différents selon le sexe et l’exposition à la nicotine. Deux études longitudinales ont étudié l’association entre l’usage du cannabis (UC) et le changement de poids et ont obtenus des résultats contradictoires. L’objectif de ce mémoire est de décrire la modification de l’association entre l’UC et le changement de poids par la cigarette chez les jeunes hommes et femmes. Méthodes : Des donnés de 271 hommes et 319 femmes ont été obtenues dans le cadre de l’étude NICO, une cohorte prospective (1999-2013). L’indice de masse corporelle (IMC) et la circonférence de taille (CT) ont été mesurés à l’âge de 17 et 25 ans. L’UC dans la dernière année et de cigarette dans les derniers trois mois ont été auto-rapportées à 21 ans. Les associations entre l’UC et le changement d’IMC et de CT ont été modélisées dans une régression polynomiale stratifiée par sexe avec ajustement pour l’activité physique, la sédentarité et la consommation d’alcool. Résultats : Uniquement, chez les hommes, l’interaction de l’UC et cigarettes était statistiquement significative dans le model de changement IMC (p=0.004) et celui de changement de CT (p=0.043). L’UC était associé au changement d’adiposité dans une association en forme de U chez les homes non-fumeurs et chez les femmes, et dans une association en forme de U-inversé chez les hommes fumeurs. Conclusion : La cigarette semble modifier l’effet du cannabis sur le changement d’IMC et CT chez les hommes, mais pas chez les femmes. / Background: Animal studies suggest that the endocannabinoid system is a regulator of energy homeostasis, whose effects are modified by sex and nicotine. Two studies in humans have examined the association between cannabis use and change in adiposity, and obtained conflicting results. This thesis aimed to determine if the association between cannabis use and change in adiposity is modified by cigarette smoking in young adults. Methods: Data were available for 271 males and 319 females participating in the Nicotine Dependence In Teens study, a prospective cohort investigation (1999-2013). Body mass index (BMI) and waist circumference (WC) were measured at ages 17 and 25 years. Self-report data on past-year cannabis use and past three-month cigarette smoking were collected at age 21 years. Modification of the association between cannabis use and change in adiposity by cigarette smoking was tested separately in each sex, in polynomial linear regression models controlling for physical activity and sedentary behavior in both sexes, and alcohol use in males only. Results: In males only, the interaction between cannabis use and cigarette smoking was statistically significant in both the model for change in BMI (p=0.004) and the model for change in WC (p=0.043). Cannabis use was associated with change in adiposity in a U-shaped form in females and in non smoking males, and in an inverted U-shaped association in males who smoked more than 10 cigarettes per day. Conclusion: Smoking cigarettes appears to attenuate the association between cannabis use and change in adiposity in young men, but not in young women.