Facilitating barriers: Contextual factors and self-management of type 2 diabetes in urban settings.

Urban environments create unique challenges for the management of type 2 diabetes (T2D). City living is associated with unhealthy occupational, nutritional, and physical activity patterns. However, it has also been linked to behaviours that promote health, such as walking and cycling for transportation. Our research is situated at the intersection of these contradictory findings. We ask: What aspects of urban living impact the ability of those living with diabetes to reach optimal health? What contextual and structural factors influence how barriers are experienced in the everyday lives of those living with T2D? We conducted semi-structured interviews with 29 individuals living in Toronto and Vancouver. Interviews were recorded, transcribed, and systematically coded for themes and sub-themes. In addition to affirming readily acknowledged barriers to diabetes management, such as accessing healthy, culturally appropriate food and the cost of management, our findings suggest that the unpredictable nature of urban living creates barriers to routinizing self-management practices. As large, cosmopolitan centres with an abundance of activities on offer, cities pulls people away from home, making adherence to self-management recommendations more difficult. Moreover, our findings challenge commonly held assumptions about the mutually exclusive and static nature of barriers and facilitators. Public transit, a readily acknowledged facilitator of healthy living, can be experienced as a barrier to diabetes management. Participants report intentional non-adherence to their medication regimens for fear of hypoglycemia in subway or traffic delays. While the stimulating nature of cities promotes walkability, it produces barriers as well: participants partake in more restaurant eating than they would if they lived in a rural area and were home to cook their own meals. Understanding how barriers are experienced by people living with diabetes will help mitigate some of the unintended consequences associated with various contextual factors. We recommend that healthcare professionals acknowledge and support people with T2D in routinizing self-management and developing contingency plans for the unpredictability and complexity that urban living entails. We suggest further research be carried out to develop contextually-tailored municipal policies and interventions that will support self-management and improve outcomes for individuals living with T2D in urban settings.

The influence of measurement error on calibration, discrimination, and overall estimation of a risk prediction model.

BACKGROUND: Self-reported height and weight are commonly collected at the population level; however, they can be subject to measurement error. The impact of this error on predicted risk, discrimination, and calibration of a model that uses body mass index (BMI) to predict risk of diabetes incidence is not known. The objective of this study is to use simulation to quantify and describe the effect of random and systematic error in self-reported height and weight on the performance of a model for predicting diabetes. METHODS: Two general categories of error were examined: random (nondirectional) error and systematic (directional) error on an algorithm relating BMI in kg/m2 to probability of developing diabetes. The cohort used to develop the risk algorithm was derived from 23,403 Ontario residents that responded to the 1996/1997 National Population Health Survey linked to a population-based diabetes registry. The data and algorithm were then simulated to allow for estimation of the impact of these errors on predicted risk using the Hosmer-Lemeshow goodness-of-fit χ2 and C-statistic. Simulations were done 500 times with sample sizes of 9,177 for males and 10,618 for females. RESULTS: Simulation data successfully reproduced discrimination and calibration generated from population data. Increasing levels of random error in height and weight reduced the calibration and discrimination of the model. Random error biased the predicted risk upwards whereas systematic error biased predicted risk in the direction of the bias and reduced calibration; however, it did not affect discrimination. CONCLUSION: This study demonstrates that random and systematic errors in self-reported health data have the potential to influence the performance of risk algorithms. Further research that quantifies the amount and direction of error can improve model performance by allowing for adjustments in exposure measurements.

Identifying factors likely to influence compliance with diagnostic imaging guideline recommendations for spine disorders among chiropractors in North America: a focus group study using the Theoretical Domains Framework.

BACKGROUND: The Theoretical Domains Framework (TDF) was developed to investigate determinants of specific clinical behaviors and inform the design of interventions to change professional behavior. This framework was used to explore the beliefs of chiropractors in an American Provider Network and two Canadian provinces about their adherence to evidence-based recommendations for spine radiography for uncomplicated back pain. The primary objective of the study was to identify chiropractors' beliefs about managing uncomplicated back pain without x-rays and to explore barriers and facilitators to implementing evidence-based recommendations on lumbar spine x-rays. A secondary objective was to compare chiropractors in the United States and Canada on their beliefs regarding the use of spine x-rays. METHODS: Six focus groups exploring beliefs about managing back pain without x-rays were conducted with a purposive sample. The interview guide was based upon the TDF. Focus groups were digitally recorded, transcribed verbatim, and analyzed by two independent assessors using thematic content analysis based on the TDF. RESULTS: Five domains were identified as likely relevant. Key beliefs within these domains included the following: conflicting comments about the potential consequences of not ordering x-rays (risk of missing a pathology, avoiding adverse treatment effects, risks of litigation, determining the treatment plan, and using x-ray-driven techniques contrasted with perceived benefits of minimizing patient radiation exposure and reducing costs; beliefs about consequences); beliefs regarding professional autonomy, professional credibility, lack of standardization, and agreement with guidelines widely varied ( social/professional role & identity); the influence of formal training, colleagues, and patients also appeared to be important factors ( social influences); conflicting comments regarding levels of confidence and comfort in managing patients without x-rays ( belief about capabilities); and guideline awareness and agreements ( knowledge). CONCLUSIONS: Chiropractors' use of diagnostic imaging appears to be influenced by a number of factors. Five key domains may be important considering the presence of conflicting beliefs, evidence of strong beliefs likely to impact the behavior of interest, and high frequency of beliefs. The results will inform the development of a theory-based survey to help identify potential targets for behavioral-change strategies.

The role of ethnicity in predicting diabetes risk at the population level.

BACKGROUND: The current form of the diabetes population risk tool (DPoRT) includes a non-specific category of ethnicity in concordance with publicly data available. Given the importance of ethnicity in influencing diabetes risk and its significance in a multi-ethnic population, it is prudent to determine its influence on a population-based risk prediction tool. OBJECTIVE: To apply and compare the DPoRT with a modified version that includes detailed ethnic information in Canada's largest and most ethnically diverse province. METHODS: Two additional diabetes prediction models were created: a model that contained predictors specific to the following ethnic groups--White, Black, Asian, south Asian, and First Nation; and a reference model which did not include a term for ethnicity. In addition to discrimination and calibration, 10-year diabetes incidence was compared. The algorithms were developed in Ontario using the 1996-1997 National Population Health Survey (N=19,861) and validated in the 2000/2001 Canadian community health survey (N=26,465). RESULTS: All non-white ethnicities were associated with higher risk for developing diabetes with south Asians having the highest risk. Discrimination was similar (0.75-0.77) and sufficient calibration was maintained for all models except the detailed ethnicity models for males. DPoRT produced the lowest overall ratio between observed and predicted diabetes risk. DPoRT identified more high risk cases than the other algorithms in males, whereas in females both DPoRT and the full ethnicity model identified more high risk cases. Overall DPoRT and full ethnicity algorithms were very similar in terms of predictive accuracy and population risk. CONCLUSION: Although from the individual risk perspective, incorporating information on ethnicity is important, when predicting new cases of diabetes at the population level and accounting for other risk factors, detailed ethnic information did not improve the discrimination and accuracy of the model or identify significantly more diabetes cases in the population.

Validation of a health administrative data algorithm for assessing the epidemiology of diabetes in Canadian children.

OBJECTIVE: To validate a case definition of pediatric diabetes using administrative health data and describe trends in incidence and prevalence over time in Ontario, Canada. METHODS: We sampled hospital records of 700 children from 1994 to 2003 with a prior history of at least one outpatient or hospital record for diabetes mellitus and 300 randomly selected children with no diabetes records. We defined patients as having diabetes based on diagnoses and drug utilization from chart abstraction and compared sensitivity and specificity of a number of combinations of overall health care use using administrative data to develop a highly specific definition. We used Poisson regression to test changes in incidence over time (1994-2003). RESULTS: Use of four physician claims and no hospital records over a 2-yr period yielded the most specific definition (83% sensitivity, 99% specificity). Using this definition overall age/sex standardized incidence per 100,000 was 32.3 [95% confidence intervals (CI) 30.4, 34.4] and prevalence 241.5 per 100 000 (95% CI 236.2-249.9) in 2003/2004. Overall incidence differs by age, (peaking in 10-14 yr olds) but not significantly by sex. The overall incidence has increased on average by 3.1% per year since 1994 (95% CI 1.02-1.04), with no difference in the rate of increase by age. CONCLUSIONS: Population-based surveillance of diabetes in children is possible using administrative data. This will facilitate further study of trends in incidence but also in use of health services and outcomes. Further work to differentiate type 1 and 2 diabetes will be important.

Looking inside the black box: a theory-based process evaluation alongside a randomised controlled trial of printed educational materials (the Ontario printed educational message, OPEM) to improve referral and prescribing practices in primary care in Ontario, Canada.

BACKGROUND: Randomised controlled trials of implementation strategies tell us whether (or not) an intervention results in changes in professional behaviour but little about the causal mechanisms that produce any change. Theory-based process evaluations collect data on theoretical constructs alongside randomised trials to explore possible causal mechanisms and effect modifiers. This is similar to measuring intermediate endpoints in clinical trials to further understand the biological basis of any observed effects (for example, measuring lipid profiles alongside trials of lipid lowering drugs where the primary endpoint could be reduction in vascular related deaths).This study protocol describes a theory-based process evaluation alongside the Ontario Printed Educational Message (OPEM) trial. We hypothesize that the OPEM interventions are most likely to operate through changes in physicians' behavioural intentions due to improved attitudes or subjective norms with little or no change in perceived behavioural control. We will test this hypothesis using a well-validated social cognition model, the theory of planned behaviour (TPB) that incorporates these constructs. METHODS/DESIGN: We will develop theory-based surveys using standard methods based upon the TPB for the second and third replications, and survey a subsample of Ontario family physicians from each arm of the trial two months before and six months after the dissemination of the index edition of informed, the evidence based newsletter used for the interventions. In the third replication, our study will converge with the "TRY-ME" protocol (a second study conducted alongside the OPEM trial), in which the content of educational messages was constructed using both standard methods and methods informed by psychological theory. We will modify Dillman's total design method to maximise response rates. Preliminary analyses will initially assess the internal reliability of the measures and use regression to explore the relationships between predictor and dependent variable (intention to advise diabetic patients to have annual retinopathy screening and to prescribe thiazide diuretics for first line treatment of uncomplicated hypertension). We will then compare groups using methods appropriate for comparing independent samples to determine whether there have been changes in the predicted constructs (attitudes, subjective norms, or intentions) across the study groups as hypothesised, and will assess the convergence between the process evaluation results and the main trial results.

Trends in deliveries, prenatal care, and obstetrical complications in women with pregestational diabetes: a population-based study in Ontario, Canada, 1996-2001.

OBJECTIVE: To describe recent trends in the proportion of deliveries in women with pregestational diabetes (PGD), their use of services, and diabetes-related obstetrical complications. RESEARCH DESIGN AND METHODS: In this population-based retrospective cohort study, comprehensive administrative data were used to identify all women (with and without PGD) who gave birth in an Ontario, Canada, hospital from 1996 to 2001. Data on maternal complications and interventions were obtained from hospital discharge records; data on use of prenatal services were obtained from fee-for-service claims. RESULTS: The proportion of deliveries in women with PGD increased steadily from 0.8% in 1996 to 1.2% in 2001 (P < 0.001). In 2001, women with PGD were more likely to be diagnosed with shoulder dystocia (adjusted odds ratio 2.00 [95% CI 1.55-2.58]), hypertension (4.13 [3.44-4.96]), and preeclampsia/eclampsia (4.44 [3.43-5.73]) and have higher rates of inductions (1.69 [1.52-1.88]) and caesarean sections (1.78 [1.60-1.98]) than women without PGD. In 2001, 50% of the women with PGD had a visit to a diabetes specialist during pregnancy and only 30% of women had claims for a prenatal retinal examination. Both of these rates have decreased over the study period. CONCLUSIONS: Women with PGD now account for a larger proportion of deliveries. These women continue to have higher obstetrical complication and intervention rates than women without PGD and many do not receive recommended specialty care during pregnancy.

Serum lipids and outcome of early-stage breast cancer: results of a prospective cohort study.

PURPOSE: The prognosis of women with early-stage breast cancer is influenced by insulin and body mass index (BMI). High levels of serum insulin and obesity often coexist with dyslipidemia in the insulin resistance syndrome (IRS), but the contribution of lipids to breast cancer outcome is unclear. Here, we examine whether serum levels of total cholesterol (TC) and triglycerides (TG) influence breast cancer outcome. PATIENTS AND METHODS: A cohort of 520 women without known hyperlipidemia or diabetes, with stage T1-T3, N0-N1, M0 breast cancer, was assembled from July 1989 to June 1996. Fasting blood was collected at baseline. Subjects were followed prospectively, for recurrence (local, regional, distant) and death. Cox models were used to calculate the prognostic effect of TC and TG levels. Two-sided significance levels were set at 0.025. RESULTS: TC was correlated with age (Spearman's r = 0.44) and low tumor grade (p = 0.01), while TG was correlated with insulin (r = 0.43) and BMI (r = 0.45). At a median follow-up of 8.7 years, TC and TG were not associated with breast cancer recurrence or death before of after adjustment for age, tumor-related variables, BMI or fasting insulin levels. In multivariate analysis adjusting for age, tumor-related variables and BMI, a trend towards an adverse effect of TC on disease recurrence was seen (HR recurrence = 1.62 for the 4th versus. 1st quartile, 2-sided p = 0.03). CONCLUSIONS: Fasting TG was not associated with outcome. A trend towards risk of recurrence was seen with higher TC in multivariate analysis. This potential association should be explored in future studies.

Antihypertensive therapy and incidence of type 2 diabetes in an elderly cohort.

OBJECTIVE: The aim of this study was to determine whether the incidence of type 2 diabetes differed among elderly users of four major antihypertensive drug classes. RESEARCH DESIGN AND METHODS: This was a retrospective, observational cohort study of previously untreated elderly patients (aged > or = 66 years) identified as new users of an antihypertensive drug class between April 1995 and March 2000. Using a Cox proportional hazards model, the primary analysis compared diabetes incidence in users of ACE inhibitors, beta-blockers, and calcium channel blockers (CCBs), with thiazide diuretics allowed as second-line therapy. In the secondary analysis, thiazide diuretics were added as a fourth study group. RESULTS: In the multivariable-adjusted primary analysis (n = 76,176), neither ACE inhibitor use (hazard ratio 0.96 [95% CI 0.84-1.1]) nor beta-blocker use (0.86 [0.74-1.0]) was associated with a statistically significant difference in type 2 diabetes incidence compared with the CCB control group. In the secondary analysis (n = 100,653), compared with CCB users, type 2 diabetes incidence was not significantly different between users of ACE inhibitors (0.97 [0.83-1.1]), beta-blockers (0.84 [0.7-1.0]), or thiazide diuretics (1.0 [0.89-1.2]). CONCLUSIONS: Type 2 diabetes incidence did not significantly differ among users of the major antihypertensive drug classes in this elderly, population-based administrative cohort. These results do not support the theory that different antihypertensive drug classes are relatively more or less likely to cause diabetes.