Lisa M. Lix

Statistical Practices of Educational Researchers: An Analysis of their ANOVA, MANOVA, and ANCOVA Analyses

Articles published in several prominent educational journals were examined to investigate the use of data analysis tools by researchers in four research paradigms: between-subjects univariate designs, between-subjects multivariate designs, repeated measures designs, and covariance designs. In addition to examining specific details pertaining to the research design (e.g., sample size, group size equality/inequality) and methods employed for data analysis, the authors also catalogued whether (a) validity assumptions were examined, (b) effect size indices were reported, (c) sample sizes were selected on the basis of power considerations, and (d) appropriate textbooks and/or articles were cited to communicate the nature of the analyses that were performed. The present analyses imply that researchers rarely verify that validity assumptions are satisfied and that, accordingly, they typically use analyses that are nonrobust to assumption violations. In addition, researchers rarely report effect size statistics, nor do they routinely perform power analyses to determine sample size requirements. Recommendations are offered to rectify these shortcomings.

Use of proton pump inhibitors and risk of osteoporosis-related fractures.

Background: The use of proton pump inhibitors has been associated with an increased risk of hip fracture. We sought to further explore the relation between duration of exposure to proton pump inhibitors and osteoporosis-related fractures. Methods: We used administrative claims data to identify patients with a fracture of the hip, vertebra or wrist between April 1996 and March 2004. Cases were each matched with 3 controls based on age, sex and comorbidities. We calculated adjusted odds ratios (OR) for the risk of hip fracture and all osteoporosis-related fractures for durations of proton pump inhibitor exposure ranging from 1 or more years to more than 7 years. Results: We matched 15 792 cases of osteoporosis-related fractures with 47 289 controls. We did not detect a significant association between the overall risk of an osteoportic fracture and the use of proton pump inhibitors for durations of 6 years or less. However, exposure of 7 or more years was associated with increased risk of an osteoporosis-related fracture (adjusted OR 1.92, 95% confidence interval [CI] 1.16–3.18, p = 0.011). We also found an increased risk of hip fracture after 5 or more years of exposure (adjusted OR 1.62, 95% CI 1.02–2.58, p = 0.04), with even higher risk after 7 or more years exposure (adjusted OR 4.55, 95% CI 1.68–12.29, p = 0.002). Interpretation: Use of proton pump inhibitors for 7 or more years is associated with a significantly increased risk of an osteoporosis-related fracture. There is an increased risk of hip fracture after 5 or more years exposure. Further study is required to determine the clinical importance of this finding and to determine the value of osteoprotective medications for patients with long-term use of proton pump inhibitors.

Consequences of Assumption Violations Revisited: A Quantitative Review of Alternatives to the One-Way Analysis of Variance F Test

The presence of variance heterogeneity and nonnormality in educational and psychological data may frequently invalidate the use of the analysis of variance (ANOVA) F test in one-way independent groups designs. This article offers recommendations to applied researchers on the use of various parametric and nonparametric alternatives to the F test under assumption violation conditions. Meta-analytic techniques were used to summarize the statistical robustness literature on the Type I error properties of the Brown-Forsythe (Brown & Forsythe, 1974), James (1951) second-order, Kruskal-Wallis (Kruskal & Wallis, 1952), and Welch (1951) tests. Two variables, based on the theoretical work of Box (1954), are shown to be highly effective in deciding when a particular alternative procedure should be adopted. Based on the meta-analysis findings, it is recommended that researchers gain a clear understanding of the nature of their data before conducting statistical analyses. Of all of the procedures, the James and Welch ...

The Manitoba IBD Cohort Study: A Population-Based Study of the Prevalence of Lifetime and 12-Month Anxiety and Mood Disorders

BACKGROUNDG AND AIMS: ven the impact of anxiety and mood disorders on health, it is important to consider these disorders in persons with inflammatory bowel disease (IBD). We assessed the prevalence of anxiety and mood disorders in a population-based IBD cohort.METHODS:A structured diagnostic interview was administered to participants in the cohort (N = 351), and rates were compared to age-, gender-, and region-matched controls drawn from a national survey (N = 779).RESULTS: A comparison of lifetime prevalence suggests higher rates of panic, generalized anxiety, and obsessive-compulsive disorders and major depression and lower rates of social anxiety and bipolar disorders in the IBD sample than in national samples in the United States and New Zealand. Direct comparisons with matched controls (with data available for three anxiety disorders) found lifetime prevalence (IBD vs controls) as follows: social anxiety disorder lower in IBD (6% vs 11%, OR 0.52, 95% CI 0.32–0.85), panic disorder not significantly different (8.0% vs 4.7%, OR 1.59, 95% CI 0.96–2.63), agoraphobia without panic not significantly different (1.1% vs 0.6%, OR 1.44, 95% CI 0.37–5.55), and major depression higher (27.2% vs 12.3%, OR 2.20, 95% CI 1.64–2.95). Comparing IBD respondents with and without lifetime anxiety or mood disorder, those with a disorder reported lower quality of life and earlier onset of IBD symptoms and there was a trend toward earlier IBD diagnosis.CONCLUSIONS:Clinicians should be aware of the increased prevalence of depression and possibly other anxiety disorders in persons with IBD as these disorders may influence response to treatment and quality of life.

Predictors of medication adherence in inflammatory bowel disease.

BACKGROUND AND AIMS:This study reports cross-sectional medication adherence data from year 1 of the Manitoba Inflammatory Bowel Disease (IBD) Cohort Study, a longitudinal, population-based study of multiple determinants of health outcomes in IBD in those diagnosed within 7 yr.METHODS:A total of 326 participants completed a validated multi-item self-report measure of adherence, which assesses a range of adherence behaviors. Demographic, clinical, and psycho-social characteristics were also assessed by survey. Adherence was initially considered as a continuous variable and then categorized as high or low adherence for logistic regression analysis to determine predictors of adherence behavior.RESULTS:Using the cutoff score of 20/25 on the Medication Adherence Report Scale, high adherence was reported by 73% of men and 63% of women. For men, predictors of low adherence included diagnosis (UC: OR 4.42, 95% CI 1.66–11.75) and employment status (employed: OR 11.27, 95% CI 2.05–62.08). For women, predictors of low adherence included younger age (under 30 versus over 50 OR 3.64, 95% CI 1.41–9.43; under 30 vs. 40–49 yr: OR 2.62, 95% CI 1.07–6.42). High scores on the Obstacles to Medication Use Scale strongly related to low adherence for both men (OR 4.05, 95% CI 1.40–11.70) and women (OR 3.89, 95% CI 1.90–7.99). 5-ASA use (oral or rectal) was not related to adherence. For women, immunosuppressant use versus no use was associated with high adherence (OR 4.49, 95% CI 1.58–12.76). Low trait agreeableness was associated with low adherence (OR 2.03, 95% CI 1.12–3.66).CONCLUSIONS:Approximately one-third of IBD patients were low adherers. Predictors of adherence differed markedly between genders, although obstacles such as medication cost were relevant for both men and women.

Proton-Pump Inhibitor Use Is Not Associated With Osteoporosis or Accelerated Bone Mineral Density Loss

BACKGROUNDS & AIMS Recent studies have shown an association between proton-pump inhibitor use (PPI) and hip fracture. The mechanism by which PPI use promotes the development of hip fracture is uncharacterized. Therefore, we sought to determine whether PPI use is associated with osteoporosis or accelerated bone mineral density (BMD) loss. METHODS We used the Manitoba Bone Mineral Density Database to determine the relationship between chronic PPI use and osteoporosis on an initial assessment of BMD and on BMD loss between successive assessments of BMD. In the cross-sectional study, cases with osteoporosis at the hip or lumbar vertebrae (T-score < or =-2.5) were matched to 3 controls with normal BMD (T-score > or =-1.0). In the longitudinal analysis, the change in BMD among PPI users and nonusers between successive BMD assessments was assessed. Conditional logistic regression and multivariate linear regression were used to obtain estimates of the association between PPI use and osteoporosis and of the annualized change in BMD associated with PPI use. RESULTS PPI use was not associated with having osteoporosis at either the hip (OR, 0.84; 95% CI, 0.55-1.34) or the lumbar spine (OR, 0.79; 95% CI, 0.59-1.06) for PPI use >1500 doses over the previous 5 years. In the longitudinal study no significant decrease was observed in BMD at either site attributable to PPI use. CONCLUSIONS PPI use does not appear to be associated with either the presence of osteoporosis or accelerated BMD loss. The association between PPI use and hip fracture is probably related to factors independent of osteoporosis.

FRAX underestimates fracture risk in patients with diabetes

The study objective was to determine whether diabetes is a risk factor for incident hip or major osteoporotic fractures independent of the WHO fracture risk assessment tool (FRAX). Men and women with diabetes (n = 3518) and nondiabetics (n = 36,085) aged ≥50 years at the time of bone mineral density (BMD) testing (1990 to 2007) were identified in a large clinical database from Manitoba, Canada. FRAX probabilities were calculated, and fracture outcomes to 2008 were established via linkage with a population‐based data repository. Multivariable Cox proportional hazards models were used to determine if diabetes was associated with incident hip fractures or major osteoporotic fractures after controlling for FRAX risk factors. Mean 10‐year probabilities of fracture were similar between groups for major fractures (diabetic 11.1 ± 7.2 versus nondiabetic 10.9 ± 7.3, p = 0.116) and hip fractures (diabetic 2.9 ± 4.4 versus nondiabetic 2.8 ± 4.4, p = 0.400). Diabetes was a significant predictor of subsequent major osteoporotic fracture (hazard ratio [HR] = 1.61, 95% confidence interval [CI] 1.42–1.83) after controlling for age, sex, medication use, and FRAX risk factors including BMD. Similar results were seen after adjusting for FRAX probability directly (HR = 1.59, 95% CI 1.40–1.79). Diabetes was also associated with significantly higher risk for hip fractures (p < 0.001). Higher mortality from diabetes attenuated but did not eliminate the excess fracture risk. FRAX underestimated observed major osteoporotic and hip fracture risk in diabetics (adjusted for competing mortality) but demonstrated good concordance with observed fractures for nondiabetics. We conclude that diabetes confers an increased risk of fracture that is independent of FRAX derived with BMD. This suggests that diabetes might be considered for inclusion in future iterations of FRAX.

Longitudinal study of quality of life and psychological functioning for active, fluctuating, and inactive disease patterns in inflammatory bowel disease.

Background: The aim was to assess quality of life (QOL) and psychological functioning in inflammatory bowel disease (IBD) as related to patterns of disease activity over time. Methods: Study participants were 388 recently diagnosed individuals from the population‐based Manitoba IBD Cohort Study. They completed mail‐out surveys at 6‐month intervals and clinical interviews annually. Based on their 2‐year pattern of self‐reported disease activity, participants were assigned to 1 of 3 groups: consistently active, fluctuating, or consistently inactive disease. Disease type (Crohns disease [CD] or ulcerative colitis [UC]) was confirmed through chart review. Change over time was modeled for measures of QOL and positive and negative psychological functioning using mixed‐effects regression analyses. Results: Half of the participants had fluctuating disease activity, while almost one‐third of participants reported consistent active disease. Participants with the fluctuating activity pattern showed significant improvement in disease‐specific QOL compared to participants with consistent activity. Perceived stress, health anxiety, and pain anxiety decreased while pain catastrophizing and mastery increased over time, although the amount of change was not significantly different among disease activity patterns. However, when the data were averaged over time there were significant differences among disease activity patterns on most outcomes. Significant effects of CD versus UC were observed only for the pain measures. Conclusions: Change in IBD QOL is influenced by ones longitudinal profile of disease activity, but change in psychological functioning is not. Effects of disease activity on psychological functioning were modest, suggesting that disease has an impact even when patients are not experiencing active symptoms.

Independent clinical validation of a Canadian FRAX tool: fracture prediction and model calibration.

A FRAX model for Canada was constructed for prediction of osteoporotic and hip fracture risk using national hip fracture data with and without the use of femoral neck bone mineral density (BMD). Performance of this system was assessed independently in a large clinical cohort of 36,730 women and 2873 men from the Manitoba Bone Density Program database that tracks all clinical dual‐energy X‐ray absorptiometry (DXA) test results for the Province of Manitoba, Canada. Linkage with other provincial health databases allowed for the direct comparison of fracture risk estimates from the Canadian FRAX model with observed fracture rates to 10 years (549 individuals with incident hip fractures and 2543 with incident osteoporotic fractures). The 10‐year Kaplan‐Meier estimate for hip fractures in women was 2.7% [95% confidence interval (CI) 2.1–3.4%] with a predicted value of 2.8% for FRAX with BMD, and in men the observed risk was 3.5% (95% CI 0.8–6.2%) with predicted value of 2.9%. The 10‐year estimate of osteoporotic fracture risk for all women was 12.0% (95% CI 10.8–13.4%) with a predicted value of 11.1% for FRAX with BMD, and in men, the observed risk was 10.7% (95% CI 6.6–14.9%) with a predicted value of 8.4%. Discrepancies were observed within some subgroups but generally were small. Fracture discrimination based on receiver operating characteristic curve analysis was comparable with published meta‐analyses with area under the curve for osteoporotic fracture prediction of 0.694 (95% CI 0.684–0.705) for FRAX with BMD and for hip fractures 0.830 (95% CI 0.815–0.846), both of which were better than FRAX without BMD or BMD alone. Individual risk factors considered by FRAX made significant independent contributions to fracture prediction in one or more of the models. In conclusion, a Canadian FRAX tool calibrated on national hip fracture data generates fracture risk predictions that generally are consistent with observed fracture rates across a wide range of risk categories.

A population-based study of fatigue and sleep difficulties in inflammatory bowel disease

Background: There has been little investigation of fatigue, a common symptom in inflammatory bowel disease (IBD). The aim of this study was to evaluate fatigue more comprehensively, considering relationships with psychological and biological factors simultaneously in a population‐based IBD community sample. Methods: Manitoba IBD Cohort Study participants (n = 318; 51% Crohns disease [CD]) were assessed by survey, interview, and blood sample. Fatigue, sleep quality, daytime drowsiness, stress, psychological distress, and quality of life were measured with validated scales. Hemoglobin (Hg) and C‐reactive protein (CRP) levels were also obtained. Differences were tested across disease activity and disease subtype. Results: Elevated CRP was found for 23% of the sample and 12% were anemic; 46% had active disease. Overall, 72% of those with active and 30% with inactive disease reached clinical thresholds for fatigue (Multidimensional Fatigue Inventory; P < 0.001); 77% and 49% of those with active or inactive disease, respectively, experienced poor sleep (P < 0.001). There were few differences between those with CD and ulcerative colitis (UC) on the factors assessed, except for higher CRP levels in CD (mean 8.8 versus 5.3, P < 0.02). Multiple logistic regression analyses found that elevated fatigue was associated with active disease (odds ratio [OR] 4.2, 95% confidence interval [CI] 2.2–7.8), poor sleep quality (OR 4.0, 95% CI 1.9–8.6), and perceived stress (OR 4.2, 95% CI 2.2–8.1), but not with hours of sleep, Hg, or CRP. Conclusions: Fatigue and poor sleep are not only highly prevalent in active disease, but both are still significant concerns for many with inactive disease. Psychological factors are associated with fatigue in IBD in addition to disease and sleep considerations. (Inflamm Bowel Dis 2011;)