Susan J. Elliott

Developing and Implementing a Triangulation Protocol for Qualitative Health Research

In this article, the authors present an empirical example of triangulation in qualitative health research. The Canadian Heart Health Dissemination Project (CHHDP) involves a national examination of capacity building and dissemination undertaken within a series of provincial dissemination projects. The Projects focus is on the context, processes, and impacts of health promotion capacity building and dissemination. The authors collected qualitative data within a parallel–case study design using key informant interviews as well as document analysis. Given the range of qualitative data sets used, it is essential to triangulate the data to address completeness, convergence, and dissonance of key themes. Although one finds no shortage of admonitions in the literature that it must be done, there is little guidance with respect to operationalizing a triangulation process. Consequently, the authors are feeling their way through the process, using this opportunity to develop, implement, and reflect on a triangulation protocol.

Overall prevalence of self-reported food allergy in Canada

Estimate 1: Including all adults Peanut 1.77 (1.21-2.33) 0.78 (0.58-0.97) 1.00 (0.80-1.20) Tree nut 1.73 (1.16-2.30) 1.07 (0.84-1.30) 1.22 (1.00-1.44) Fish 0.18 (0.00-0.36) 0.60 (0.43-0.78) 0.51 (0.37-0.65) Shellfish 0.55 (0.21-0.88) 1.91 (1.60-2.23) 1.60 (1.35-1.86) Sesame 0.23 (0.03-0.43) 0.07 (0.01-0.13) 0.10 (0.04-0.17) Milk 2.23 (1.51-2.95) 1.89 (1.56-2.21) 1.97 (1.64-2.29) Egg 1.23 (0.69-1.77) 0.67 (0.48-0.86) 0.80 (0.61-0.99) Wheat 0.45 (0.08-0.83) 0.86 (0.63-1.08) 0.77 (0.57-0.96) Soy 0.32 (0.08-0.55) 0.16 (0.07-0.25) 0.20 (0.10-0.30) Fruits 1.14 (0.68-1.60) 1.61 (1.32-1.89) 1.50 (1.25-1.75) Vegetables 0.45 (0.17-0.74) 1.29 (1.02-1.55) 1.10 (0.88-1.31) Other 1.32 (0.80-1.84) 1.67 (1.37-1.97) 1.59 (1.32-1.86) All foods 7.14 (5.92-8.36) 8.34 (7.69-8.99) 8.07 (7.47-8.67) Estimate 2: Excluding some adults All foods 7.14 (5.92-8.36) 6.56 (5.99-7.13) 6.69 (6.15-7.24) Estimate 3: Estimate 2 adjusted for nonresponse All foods 7.12 (6.07-8.28) 6.58 (6.22-6.96) 6.67 (6.19-7.17)

From Siting Principles to Siting Practices: A Case Study of Discord among Trust, Equity and Community Participation

This paper contributes to the noxious facilities siting literature by exploring some implications of adhering to some recommended principles and practices for competent siting. Through a qualitative case study of a landfill siting process in Peel (Ontario, Canada) three principles are critically assessed: trust; equity; and community participation. While laudable notions in principle, in practice they can impact each other in important ways which can (potentially) undermine the siting process. These impacts result mainly from the failure to achieve meaningful goals associated with one principle (e.g. community participation) which can exacerbate problems achieving goals associated with other principles (e.g. trust). The resulting discord can be further aggravated by the snowballing of adverse effects over time. In particular, practices for achieving trust and equity were adversely linked, as were the relationships between spatial equity and procedural equity and interregional and intraregional equity. These adverse synergisms were linked together with ineffective community participation which brought the process to a halt. Siting inertia (a process momentum difficult to redirect) and an inflexible siting context contributed to these conflicts. Implications for siting and further research are discussed.

Body mass index versus waist circumference as predictors of mortality in Canadian adults

Background:Elevated body mass index (BMI) and waist circumference (WC) are associated with increased mortality risk, but it is unclear which anthropometric measurement most highly relates to mortality. We examined single and combined associations between BMI, WC, waist–hip ratio (WHR) and all-cause, cardiovascular disease (CVD) and cancer mortality.Methods:We used Cox proportional hazard regression models to estimate relative risks of all-cause, CVD and cancer mortality in 8061 adults (aged 18–74 years) in the Canadian Heart Health Follow-Up Study (1986–2004). Models controlled for age, sex, exam year, smoking, alcohol use and education.Results:There were 887 deaths over a mean 13 (SD 3.1) years follow-up. Increased risk of death from all-causes, CVD and cancer were associated with elevated BMI, WC and WHR (P<0.05). Risk of death was consistently higher from elevated WC versus BMI or WHR. Ascending tertiles of each anthropometric measure predicted increased CVD mortality risk. In contrast, all-cause mortality risk was only predicted by ascending WC and WHR tertiles and cancer mortality risk by ascending WC tertiles. Higher risk of all-cause death was associated with WC in overweight and obese adults and with WHR in obese adults. Compared with non-obese adults with a low WC, adults with high WC had higher all-cause mortality risk regardless of BMI status.Conculsion:BMI and WC predicted higher all-cause and cause-specific mortality, and WC predicted the highest risk for death overall and among overweight and obese adults. Elevated WC has clinical significance in predicting mortality risk beyond BMI.

The COMPASS study: a longitudinal hierarchical research platform for evaluating natural experiments related to changes in school-level programs, policies and built environment resources

BackgroundFew researchers have the data required to adequately understand how the school environment impacts youth health behaviour development over time.Methods/DesignCOMPASS is a prospective cohort study designed to annually collect hierarchical longitudinal data from a sample of 90 secondary schools and the 50,000+ grade 9 to 12 students attending those schools. COMPASS uses a rigorous quasi-experimental design to evaluate how changes in school programs, policies, and/or built environment (BE) characteristics are related to changes in multiple youth health behaviours and outcomes over time. These data will allow for the quasi-experimental evaluation of natural experiments that will occur within schools over the course of COMPASS, providing a means for generating “practice based evidence” in school-based prevention programming.DiscussionCOMPASS is the first study with the infrastructure to robustly evaluate the impact that changes in multiple school-level programs, policies, and BE characteristics within or surrounding a school might have on multiple youth health behaviours or outcomes over time. COMPASS will provide valuable new insight for planning, tailoring and targeting of school-based prevention initiatives where they are most likely to have impact.

Conceptualizing Dissemination Research and Activity: The Case of the Canadian Heart Health Initiative

Cardiovascular diseases are now the worlds leading cause of death. To reduce high rates of such preventable premature deaths, evidence-based approaches to heart health promotion must be disseminated across public health systems. To succeed, we must build capacity to disseminate strategies that are practical and effective. However, we know little about such dissemination, and we lack both conceptual frameworks to guide our thinking and appropriate scientific methodologies. This article presents conceptual and analytic frameworks that integrate several approaches to understanding and studying dissemination processes within public health systems. This work is based on the Canadian Heart Health Dissemination Project, a research program examining a national heart health dissemination initiative. The primary focus is the development of a systematic protocol for measuring levels of capacity and dissemination, and determining successful conditions for, and barriers to, capacity and dissemination, as well as the nature of the relationship between these key concepts.

Prognosis after West Nile Virus Infection

Context The long-term prognosis of West Nile virus infection is not well understood. Contribution In this longitudinal study of 156 patients with West Nile virus infection, physical and cognitive function seemed to return to population norms within about 1 year. Caution Patients who died were excluded from the analysis, and the analyses depended on statistical assumptions that the data did not always meet. Implication People infected with West Nile virus seem to recover physical and mental function within about 1 year. The Editors West Nile virus, endemic to Africa, Europe, the Middle East, and Asia, has caused recurrent outbreaks in the United States and Canada since 1999 (1, 2). Approximately 20% of infected persons develop a clinical presentation that can range from a mild influenza-like illness to neuroinvasive diseases, such as meningitis, encephalitis, and acute flaccid paralysis (3). Recent studies of persons infected with West Nile virus report that symptoms and signs, such as fatigue, cognitive dysfunction, and motor abnormalities, can persist for months after symptom onset (414). Little is known, however, about how physical and mental functioning changes over time or about long-term recovery among infected persons. Understanding such change patterns is essential to provide accurate prognostic information to patients and their families, as well as to help in the planning of care and evaluation of future interventions. Existing reports provide valuable information on self-reported outcomes (413) but have limitations, including single follow-up assessments (4, 69, 1113), follow-up until 12 months after symptom onset (4, 68, 1014), and lack of validated instruments to measure physical and mental functioning (48, 1113). Moreover, factors associated with slower recovery are unknown. The primary objectives of this study were to describe patterns of physical and mental outcomes after infection with West Nile virus by using longitudinal observations and to assess long-term outcomes. We hypothesized that such long-term outcomes would be worse in patients with neuroinvasive disease than in those with nonneuroinvasive disease. We conducted a longitudinal cohort study to develop prognostic curves for patient-relevant outcomes, such as physical and mental functioning, fatigue, and depression. We also assessed factors associated with delayed recovery of physical and mental outcomes. Methods Study Participants and Protocol Patients with positive West Nile virus IgM antibody-capture enzyme-linked immunosorbent assay (15) from serum or cerebrospinal fluid samples that was subsequently confirmed by plaque reduction neutralization assay (16) were eligible. We enrolled patients with neuroinvasive disease (meningitis, encephalitis, or acute flaccid paralysis) and nonneuroinvasive disease. We aimed to enroll participants within 4 weeks of symptom onset. To allow for feasibility of follow-up, we limited enrollment to geographic regions that had 4 or more infected individuals within an approximate radius of 200 km. Because complications of West Nile virus infection in children are generally less frequent, we excluded persons younger than age 18 years (17). Because the objective of the study was to assess prognosis attributable to West Nile virus, we excluded patients receiving experimental therapy for West Nile virus, as well as those being treated for an illness unrelated to West Nile virus that could interfere with interpretation of the outcome measures. Only 1 patient, who was receiving chemotherapy for a malignant solid organ condition, met the latter criterion. We classified participants with neuroinvasive disease by using previously published criteria for meningitis, encephalitis, and acute flaccid paralysis (4) (Table 1). We classified participants who met criteria for both meningitis and encephalitis as having meningoencephalitis. Only 3 patients had meningitis alone, so we combined them with the meningoencephalitis group for analysis. Table 1. Diagnostic Criteria for Meningitis, Encephalitis, and Acute Flaccid Paralysis We classified participants who were symptomatic but did not meet any of the case definitions for neuroinvasive disease as having nonneuroinvasive disease. We reviewed laboratory, hospital, and clinic medical records to confirm the case definition. Radiologic and laboratory assessments were performed at the discretion of the attending physician. Supportive care, such as physiotherapy or psychotherapy, was under the discretion of the attending physician. We enrolled the first participant on August 2003 and the last on November 2006, with the final follow-up visit on May 2007. Provincial laboratories in Canada conducted all West Nile virus testing and forwarded the names of the physicians whose patients tested IgM-positive for West Nile virus to the study office. We asked these physicians to approach the patients (whose identity remained unknown to the research team, thereby maintaining confidentiality) or their families to see whether they were willing to be contacted about the study. A research nurse then assessed patient eligibility and obtained informed consent. We obtained ethics approval for the study protocol from the relevant review committees at McMaster University, University of Manitoba, University of Saskatchewan, and University of Alberta. All patients or their designated surrogate decision makers who agreed to participate in the study gave informed consent. Outcomes A trained research nurse assessed outcomes on enrollment into the study (baseline visit); on days 10, 20, and 30; and then every month for 12 months. In our original protocol, we planned to obtain repeated measurements at 24 and 36 months for participants enrolled in the first year of the study and to measure outcomes at 24 months (third year of the study) for those participants enrolled in the second year. Because participants were being visited more frequently to obtain blood work for an unrelated study from 15 to 36 months from their enrollment in this study, we obtained additional measurements at 3-month intervals from 15 to 36 months in these participants. A trained research nurse made most assessments during home visits, minimizing missing visits or selection bias based on participants ability to travel (although participants in Saskatchewan and Manitoba were seen in an ambulatory care setting at a tertiary care hospital for their convenience). We recorded age, sex, medical history, and premorbid chronic illnesses. Categories of comorbid conditions included cardiac disease (coronary artery disease and congestive heart failure), peripheral vascular disease, chronic obstructive pulmonary disease, diabetes, renal failure, peptic ulcer disease, cancer, and rheumatologic disease. We assessed all these conditions through interviews with participants and review of medical records. To assess physical functioning, we used the Physical Component Summary (PCS) of the Short Form-36 (18, 19). The Short Form-36 measures 8 health constructs by using 8 scales with 2 to 10 items per scale (total of 36 questions); raw scores range from 0 to 100 but are adjusted for population norms by using a linear transformation (18). For the PCS subscale, scores are standardized to the general U.S. population (mean score, 50 [SD, 10]). Very high scores indicate no physical limitations, disabilities, or decrements in well-being, as well as a high energy level. Very low scores indicate substantial limitations in self-care and physical, social, and role activities; severe bodily pain; or frequent tiredness. To assess mental functioning, we used the Mental Component Summary (MCS) of the Short Form-36. These scores are also standardized to the general U.S. population (mean score, 50 [SD, 10]). Very high scores indicate frequent positive affect and absence of psychological distress and limitations in usual social and role activities due to emotional problems. Very low scores indicate frequent psychological distress and substantial social and role disability due to emotional problems (18). We used the Depression Anxiety Stress Scale (DASS) (20, 21) and the Fatigue Severity Scale (FSS) (22) beginning in 2004 to capture depressive symptoms and persistent fatigue noted among participants after the study had begun. The DASS is a 14-item scale that assesses dysphoria and lack of interest or involvement. Scores range from 0 (no symptoms) to 42. Among a general adult population, 80% of people have a score of 9 or less and 70% have a score of 6 or less (23). The FSS measures the perceived level of fatigue by using a Likert scale, in which the score ranges from 1 (low fatigue level) to 7 (high fatigue level). Two thirds of the general population have a score between 2.7 and 5.3 (24). Statistical Analysis Prognosis We used data from 57 of 64 participants with neuroinvasive disease (meningoencephalitis [n= 32], encephalitis [n= 22], and meningitis [n= 3]) and 92 with nonneuroinvasive disease to estimate prognosis. We excluded 7 patients with acute flaccid paralysis from the analysis because this number was too small to estimate prognostic curves, and we grouped 3 participants with meningitis with the 32 participants in the meningoencephalitis group. We used nonlinear mixed-effects modeling to estimate the parameters of nonlinear models for PCS, MCS, DASS, and FSS scores (25). We compared participants with neuroinvasive disease with those with nonneuroinvasive disease. Of participants with neuroinvasive disease, we compared those with meningoencephalitis with those with encephalitis. The fixed effects in the nonlinear mixed-effects analysis describe the average pattern of change over time, and they indicate the typical course of recovery in this population. The random effects in the nonlinear mixed-effects analysis allow for orderly variations in the pattern of change among patients. We estimated predicted change curves for each patient, and we estimated the degree of i

Protective Behavior and West Nile Virus Risk

We conducted a cross-sectional, household survey in Oakville, Ontario, where an outbreak of West Nile virus (WNV) in 2002 led to an unprecedented number of cases of meningitis and encephalitis. Practicing >2 personal protective behavior traits reduced the risk for WNV infection by half.

Social capital, collective action and access to water in rural Kenya.

Globally, an estimated 748 million people remain without access to improved sources of drinking water and close to 1 billion people practice open defecation (WHO/UNICEF, 2014). The lack of access to safe water and adequate sanitation presents significant health and development challenges to individuals and communities, especially in low and middle income countries. Recent research indicates that aside from financial challenges, the lack of social capital is a barrier to collective action for community based water and sanitation initiatives (Levison et al., 2011; Bisung and Elliott, 2014). This paper reports results of a case study on the relationships between elements of social capital and participation in collective action in the context of addressing water and sanitation issues in the lakeshore village of Usoma, Western Kenya. The paper uses household data (N=485, 91% response rate) collected using a modified version of the social capital assessment tool (Krishna and Shrader, 2000). Findings suggest that investment in building social capital may have some contextual benefits for collective action to address common environmental challenges. These findings can inform policy interventions and practice in water and sanitation delivery in low and middle income countries, environmental health promotion and community development.

Associations between bacterial communities of house dust and infant gut

The human gut is host to a diverse and abundant community of bacteria that influence health and disease susceptibility. This community develops in infancy, and its composition is strongly influenced by environmental factors, notably perinatal anthropogenic exposures such as delivery mode (Cesarean vs. vaginal) and feeding method (breast vs. formula); however, the built environment as a possible source of exposure has not been considered. Here we report on a preliminary investigation of the associations between bacteria in house dust and the nascent fecal microbiota from 20 subjects from the Canadian Healthy Infant Longitudinal Development (CHILD) Study using high-throughput sequence analysis of portions of the 16S rRNA gene. Despite significant differences between the dust and fecal microbiota revealed by Nonmetric Multidimensional Scaling (NMDS) analysis, permutation analysis confirmed that 14 bacterial OTUs representing the classes Actinobacteria (3), Bacilli (3), Clostridia (6) and Gammaproteobacteria (2) co-occurred at a significantly higher frequency in matched dust-stool pairs than in randomly permuted pairs, indicating an association between these dust and stool communities. These associations could indicate a role for the indoor environment in shaping the nascent gut microbiota, but future studies will be needed to confirm that our findings do not solely reflect a reverse pathway. Although pet ownership was strongly associated with the presence of certain genera in the dust for dogs (Agrococcus, Carnobacterium, Exiguobacterium, Herbaspirillum, Leifsonia and Neisseria) and cats (Escherichia), no clear patterns were observed in the NMDS-resolved stool community profiles as a function of pet ownership.