Amalia Karahalios

A review of the reporting and handling of missing data in cohort studies with repeated assessment of exposure measures

BackgroundRetaining participants in cohort studies with multiple follow-up waves is difficult. Commonly, researchers are faced with the problem of missing data, which may introduce biased results as well as a loss of statistical power and precision. The STROBE guidelines von Elm et al. (Lancet, 370:1453-1457, 2007); Vandenbroucke et al. (PLoS Med, 4:e297, 2007) and the guidelines proposed by Sterne et al. (BMJ, 338:b2393, 2009) recommend that cohort studies report on the amount of missing data, the reasons for non-participation and non-response, and the method used to handle missing data in the analyses. We have conducted a review of publications from cohort studies in order to document the reporting of missing data for exposure measures and to describe the statistical methods used to account for the missing data.MethodsA systematic search of English language papers published from January 2000 to December 2009 was carried out in PubMed. Prospective cohort studies with a sample size greater than 1,000 that analysed data using repeated measures of exposure were included.ResultsAmong the 82 papers meeting the inclusion criteria, only 35 (43%) reported the amount of missing data according to the suggested guidelines. Sixty-eight papers (83%) described how they dealt with missing data in the analysis. Most of the papers excluded participants with missing data and performed a complete-case analysis (n = 54, 66%). Other papers used more sophisticated methods including multiple imputation (n = 5) or fully Bayesian modeling (n = 1). Methods known to produce biased results were also used, for example, Last Observation Carried Forward (n = 7), the missing indicator method (n = 1), and mean value substitution (n = 3). For the remaining 14 papers, the method used to handle missing data in the analysis was not stated.ConclusionsThis review highlights the inconsistent reporting of missing data in cohort studies and the continuing use of inappropriate methods to handle missing data in the analysis. Epidemiological journals should invoke the STROBE guidelines as a framework for authors so that the amount of missing data and how this was accounted for in the analysis is transparent in the reporting of cohort studies.

Incontinence-specific quality of life measures used in trials of treatments for female urinary incontinence: a systematic review

This systematic review examined the use of incontinence-specific quality of life (QOL) measures in clinical trials of female incontinence treatments, and systematically evaluated their quality using a standard checklist. Of 61 trials included in the review, 58 (95.1%) used an incontinence-specific QOL measure. The most commonly used were IIQ (19 papers), I-QoL (12 papers) and UDI (9 papers). Eleven papers (18.0%) used measures which were not referenced or were developed specifically for the study. The eight QOL measures identified had good clinical face validity and measurement properties. We advise researchers to evaluate carefully the needs of their specific study, and select the QOL measure that is most appropriate in terms of validity, utility and relevance, and discourage the development of new measures. Until better evidence is available on the validity and comparability of measures, we recommend that researchers consider using IIQ or I-QOL with or without UDI in trials of incontinence treatments.

Alcohol consumption and risk of glioblastoma; evidence from the Melbourne Collaborative Cohort Study

Despite the brain being highly susceptible to the action of alcohol and, therefore, potentially susceptible to its carcinogenic effects, it is not clear whether alcohol consumption is associated with risk of glioblastoma. We analyzed data from 39,766 participants of the Melbourne Collaborative Cohort Study recruited in 1990–1994 and followed to the end of 2008 for an average of 15 years. Incidence of glioblastoma of the brain was ascertained via linkage to the Victorian and other State cancer registries in Australia. During a structured face‐to‐face interview at baseline we elicited each participants history of alcoholic beverage consumption during the current decade at baseline. We used Cox regression models with age as the time metric, adjusted for country of birth, sex, total energy intake, educational attainment and coffee consumption to estimate hazard ratios (HR) and corresponding 95% confidence intervals (CI). A total of 67 glioblastomas was diagnosed in the cohort during follow‐up. The HRs associated with each additional 10 grams per day of alcohol intake was 1.16 (95% CI, 1.05 to 1.29; p for linear trend = 0.007). Compared to lifetime abstainer, the HR for glioblastoma associated with alcohol consumption were 1.07 (0.55 to 2.10) for 1 to 19 g/day, 1.79 (0.81 to 3.95) for 20 to 39 g/day, 3.07 (1.26 to 7.47) for 40 to 59 g/day and 2.54 (0.92 to 7.00) for 60 or more g/day. Alcohol consumption at baseline was associated with the risk of glioblastoma in a dose‐response relationship.

Voiding Patterns and Urodynamics After a Tension-Free Vaginal Tape Procedure

OBJECTIVE To evaluate changes in urodynamic parameters, therapeutic success, and complication rates after a tension-free vaginal tape (TVT) procedure in women with stress urinary incontinence (SUI). METHODS Sixty-three women who underwent a TVT procedure for SUI, between June 1999 and December 2001 at Mount Sinai Hospital in Toronto, Ontario, were prospectively followed. Pre- and postoperative clinical data were analyzed using the Mann-Whitney, chi-square, paired t-test, and Wilcoxon signed ranks tests. RESULTS The range of follow-up was 6 weeks to 3 years. Short-term voiding dysfunction occurred in 49% of the women but was completely resolved in all women by the 12-week follow-up. For the 45 women who completed their follow-up at 6 months or 1 year after TVT, the success rate was 85%. Provocative urodynamics at 1-year follow-up compared to preoperative data showed a decrease in maximum flow rate (P < 0.001), and increase in post-void residual urine volume (P < 0.001). CONCLUSION TVT is associated with a good short-term success rate at 6 months and 1 year. A low long-term complication rate and morbidity may be expected, but significant temporary voiding dysfunction, higher than previously reported, was observed.

The effects of resistance training on muscle strength, quality of life and aerobic capacity in patients with chronic heart failure — A meta-analysis

BACKGROUND Resistance training (RT) has been utilised to target muscle dysfunction associated with Chronic Heart Failure (CHF). However, there is limited meta-analysis evidence to support its use as a standalone therapy. This meta-analysis examined the effects of RT on muscle strength (one repetition maximum, 1RM and Peak Torque), aerobic capacity (VO2peak and 6min walk distance) and quality of life (QoL) in patients with CHF. METHODS We searched Medline, EMBASE, Cochrane and CINAHL for studies published up to July 2016, combining terms related to the population (eg, heart failure, CHF) with terms for the intervention (eg, resistance, strength training) and the outcomes (eg, QoL, VO2peak,strength, aerobic capacity). RESULTS Ten studies including 240 participants were included in our meta-analysis (aged 48-76years, Ejection Fraction 18-37%). Training duration ranged from 8 to 24weeks and intensity up to 80% of 1RM. RT increased 1RM (standardised change score=0.60; 95% Confidence Interval: 0.43, 0.77) but not strength measured via peak torque at 60°/s-1 and 180°/s-1. RT increased VO2peak (CSMD: 2.71ml/kg/min; 1.96, 3.45) and QoL (CSMD: -5.71; -9.85, -1.56). CONCLUSION RT as a single intervention can increase muscle strength, aerobic capacity and QoL in patients with CHF and may offer an alternative approach, particularly for those unable to participate in aerobic training. The effect of RT on muscle strength is mainly during slow controlled movements and not during rapid movements. Older adults and patients with advanced CHF are underrepresented in RT trials and future studies should seek to optimise their inclusion.

Weight Change and Risk of Colorectal Cancer: A Systematic Review and Meta-Analysis

Results from cohort studies of adult weight gain and risk of colorectal cancer are inconsistent. We conducted a systematic review and meta-analysis of prospective studies assessing the association of change in weight/body mass index with colorectal cancer risk. We searched Scopus and Web of Science up to June 2014 and supplemented the search with manual searches of the reference lists of the identified articles. Thirteen studies published between 1997 and 2014 were pooled by using a random-effects model, and potential heterogeneity was explored by fitting meta-regression models. The highest weight gain category, measured by weight/body mass index, compared with a reference category, was associated with increased risk of colorectal cancer (hazard ratio (HR) = 1.16, 95% confidence interval (CI): 1.08, 1.24), whereas no association was found for weight loss (HR = 0.96, 95% CI: 0.89, 1.05). There was no suggestion of heterogeneity across studies. For dose response, a 5-kg weight gain was associated with a slightly increased risk of colorectal cancer (HR = 1.03, 95% CI: 1.02, 1.05), with some heterogeneity observed (I(2) = 42%; P = 0.02), which was partially explained by sex (ratio of HRs = 1.03, 95% CI: 1.00, 1.07). In this meta-analysis, gain in weight/body mass index was positively associated with colorectal cancer risk.

Past physical activity and age-related macular degeneration: the Melbourne Collaborative Cohort Study

Background/aims To assess the association between past physical activity and early, intermediate and late age-related macular degeneration (AMD) in a community-based cohort study in Melbourne, Australia. Methods Diet and lifestyle information was recorded at baseline (1990–1994) and total recreational activity was derived from walking, vigorous and non-vigorous exercise. At follow-up (2003–2007), digital macular photographs were graded for early, intermediate and late AMD. Data were analysed using multinomial logistic regression controlling for age, sex, smoking, region of descent, diet and alcohol. Effect modification by sex was investigated. Results Out of 20 816 participants, early, intermediate and late AMD were detected at follow-up in 4244 (21%), 2661 (13%) and 122 (0.6%) participants, respectively. No association was detected between past total recreational physical activity and early, intermediate or late AMD. Frequent (≥3 times/week) and less frequent (1–2 times/week) vigorous exercise were associated with lower odds of intermediate and late AMD in univariable models. After controlling for confounders, there was evidence of effect modification by sex and frequent vigorous exercise was associated with a 22% decrease in the odds of intermediate AMD (95% CI 4% to 36%) in women, but no association was found for men. Conclusions Past frequent vigorous exercise may be inversely related to the presence of intermediate AMD in women. Further studies are needed to confirm whether physical activity and exercise have a protective effect for AMD.

Change in Body Size and Mortality: Results from the Melbourne Collaborative Cohort Study

Background The association between change in weight or body mass index, and mortality is widely reported, however, both measures fail to account for fat distribution. Change in waist circumference, a measure of central adiposity, in relation to mortality has not been studied extensively. Methods We investigated the association between mortality and changes in directly measured waist circumference, hips circumference and weight from baseline (1990–1994) to wave 2 (2003–2007) in a prospective cohort study of people aged 40–69 years at baseline. Cox regression, with age as the time metric and follow-up starting at wave 2, adjusted for confounding variables, was used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for change in body size in relation to mortality from all causes, cardiovascular disease and cancer. Results There were 1465 deaths (109 cancer, 242 cardiovascular disease) identified during an average 7.7 years of follow-up from 21 298 participants. Compared to minimal increase in body size, loss of waist circumference (HR: 1.26; 95% CI: 1.09–1.47), weight (1.80; 1.54–2.11), or hips circumference (1.35; 1.15–1.57) were associated with an increased risk of all-cause mortality, particularly for older adults. Weight loss was associated with cardiovascular disease mortality (2.40; 1.57–3.65) but change in body size was not associated with obesity-related cancer mortality. Conclusion This study confirms the association between weight loss and increased mortality from all-causes for older adults. Based on evidence from observational cohort studies, weight stability may be the recommended option for most adults, especially older adults.

The impact of missing data on analyses of a time-dependent exposure in a longitudinal cohort: a simulation study

BackgroundMissing data often cause problems in longitudinal cohort studies with repeated follow-up waves. Research in this area has focussed on analyses with missing data in repeated measures of the outcome, from which participants with missing exposure data are typically excluded. We performed a simulation study to compare complete-case analysis with Multiple imputation (MI) for dealing with missing data in an analysis of the association of waist circumference, measured at two waves, and the risk of colorectal cancer (a completely observed outcome).MethodsWe generated 1,000 datasets of 41,476 individuals with values of waist circumference at waves 1 and 2 and times to the events of colorectal cancer and death to resemble the distributions of the data from the Melbourne Collaborative Cohort Study. Three proportions of missing data (15, 30 and 50%) were imposed on waist circumference at wave 2 using three missing data mechanisms: Missing Completely at Random (MCAR), and a realistic and a more extreme covariate-dependent Missing at Random (MAR) scenarios. We assessed the impact of missing data on two epidemiological analyses: 1) the association between change in waist circumference between waves 1 and 2 and the risk of colorectal cancer, adjusted for waist circumference at wave 1; and 2) the association between waist circumference at wave 2 and the risk of colorectal cancer, not adjusted for waist circumference at wave 1.ResultsWe observed very little bias for complete-case analysis or MI under all missing data scenarios, and the resulting coverage of interval estimates was near the nominal 95% level. MI showed gains in precision when waist circumference was included as a strong auxiliary variable in the imputation model.ConclusionsThis simulation study, based on data from a longitudinal cohort study, demonstrates that there is little gain in performing MI compared to a complete-case analysis in the presence of up to 50% missing data for the exposure of interest when the data are MCAR, or missing dependent on covariates. MI will result in some gain in precision if a strong auxiliary variable that is not in the analysis model is included in the imputation model.

Incidence and survival of lymphohematopoietic neoplasms according to the World Health Organization classification: a population-based study from the Victorian Cancer Registry in Australia

We studied the incidence and relative survival of 39 837 cases of lymphohematopoietic neoplasms (LHN) reported to the Victorian Cancer Registry during 1982–2004, classified according to the World Health Organization (WHO) classification. We modeled excess mortality using Poisson regression to estimate differences in survival by age, sex, and time period. Age-standardized incidence rates varied across subtypes of lymphoid and myeloid neoplasms. All major subtypes predominantly affected the elderly except Hodgkin lymphoma (incidence peaks at 20–24 and 75–79 years) and acute lymphoblastic leukemia (0–9 years). After an initial rise, overall lymphoid and myeloid incidence stabilized in the mid-1990s. The 5-year relative survival was 58% for lymphoid and 35% for myeloid neoplasms. Survival improved during 1990–2004 for diffuse large B-cell lymphoma, follicular lymphoma, acute myeloid leukemia, chronic myeloid leukemia, and myelodysplastic syndromes (p < 0.001) and declined with advancing age for all subtypes (p < 0.001). Female sex was associated with higher survival for most myeloid subtypes. The results represent a rare epidemiological characterization of the whole range of LHN according to WHO subtypes.