Adiposity and cartilage lesions following ACL reconstruction.

OBJECTIVE: To determine if global, central, or peripheral adiposity is associated with prevalent and worsening cartilage lesions following anterior cruciate ligament reconstruction (ACLR). METHODS: In 107 individuals one-year post-ACLR, adiposity was assessed globally (body mass index), centrally (waist circumference), and peripherally (knee subcutaneous adipose tissue thickness) from magnetic resonance imaging (MRI). Tibiofemoral and patellofemoral cartilage lesions were assessed from knee MRIs at 1- and 5-years post-ACLR. Poisson regression evaluated the relation of adiposity with prevalent and worsening tibiofemoral and patellofemoral cartilage lesions adjusting for age, sex, and activity level. RESULTS: The prevalence ratios of adiposity with tibiofemoral (presence in 49%) and patellofemoral (44%) cartilage lesions ranged from 0.99 to 1.03. Adiposity was more strongly associated with longitudinal changes in tibiofemoral (worsening in 21%) and patellofemoral (44%) cartilage lesions. One-unit increase in global (kg/m2), central (cm), and peripheral (mm) adiposity was associated with a higher risk of worsening tibiofemoral cartilage lesions by 17% (risk ratios [95% confidence interval (CI)]: 1.17 [1.09 to 1.23]), 5% (1.05 [1.02 to 1.08]), and 9% (1.09 [1.03 to 1.16]), and patellofemoral cartilage lesions by 5% (1.05 [1.00 to 1.12]), 2% (1.02 [1.00 to 1.04]) and 2% (1.02 [1.00 to 1.04]), respectively. CONCLUSION: Greater adiposity was a risk factor for worsening cartilage lesions up to 5 years post-ACLR. Clinical interventions aimed at mitigating excess adiposity may be beneficial in preventive approaches for early post-traumatic osteoarthritis.

RUV-III-NB: normalization of single cell RNA-seq data.

Normalization of single cell RNA-seq data remains a challenging task. The performance of different methods can vary greatly between datasets when unwanted factors and biology are associated. Most normalization methods also only remove the effects of unwanted variation for the cell embedding but not from gene-level data typically used for differential expression (DE) analysis to identify marker genes. We propose RUV-III-NB, a method that can be used to remove unwanted variation from both the cell embedding and gene-level counts. Using pseudo-replicates, RUV-III-NB explicitly takes into account potential association with biology when removing unwanted variation. The method can be used for both UMI or read counts and returns adjusted counts that can be used for downstream analyses such as clustering, DE and pseudotime analyses. Using published datasets with different technological platforms, kinds of biology and levels of association between biology and unwanted variation, we show that RUV-III-NB manages to remove library size and batch effects, strengthen biological signals, improve DE analyses, and lead to results exhibiting greater concordance with independent datasets of the same kind. The performance of RUV-III-NB is consistent and is not sensitive to the number of factors assumed to contribute to the unwanted variation.

Applying multidimensional computerized adaptive testing to the MSQOL-54: a simulation study.

BACKGROUND: The Multiple Sclerosis Quality of Life-54 (MSQOL-54) is one of the most commonly-used MS-specific health-related quality of life (HRQOL) measures. It is a multidimensional, MS-specific HRQOL inventory, which includes the generic SF-36 core items, supplemented with 18 MS-targeted items. Availability of an adaptive short version providing immediate item scoring may improve instrument usability and validity. However, multidimensional computerized adaptive testing (MCAT) has not been previously applied to MSQOL-54 items. We thus aimed to apply MCAT to the MSQOL-54 and assess its performance. METHODS: Responses from a large international sample of 3669 MS patients were assessed. We calibrated 52 (of the 54) items using bifactor graded response model (10 group factors and one general HRQOL factor). Then, eight simulations were run with different termination criteria: standard errors (SE) for the general factor and group factors set to different values, and change in factor estimates from one item to the next set at < 0.01 for both the general and the group factors. Performance of the MCAT was assessed by the number of administered items, root mean square difference (RMSD), and correlation. RESULTS: Eight items were removed due to local dependency. The simulation with SE set to 0.32 (general factor), and no SE thresholds (group factors) provided satisfactory performance: the median number of administered items was 24, RMSD was 0.32, and correlation was 0.94. CONCLUSIONS: Compared to the full-length MSQOL-54, the simulated MCAT required fewer items without losing precision for the general HRQOL factor. Further work is needed to add/integrate/revise MSQOL-54 items in order to make the calibration and MCAT performance efficient also on group factors, so that the MCAT version may be used in clinical practice and research.

Higher diet quality is associated with short and long-term benefits on SF-6D health state utilities: a 5-year cohort study in an international sample of people with multiple sclerosis.

BACKGROUND/PURPOSE: Health state utilities (HSU) are a subjective measure of an individual's health-related quality of life (HRQoL), adjusted by societal or patient relative preference weights for living in different states of health-related quality of life (HRQoL), derived from patient-reported responses to multi-attribute utility instrument (MAUI), and can be used as inputs for cost-utility analyses and in clinical assessment. This research assessed associations of diet with subsequent HSU in a large international cohort of people living with multiple sclerosis (MS), a progressive autoimmune condition of the central nervous system. METHODS: HSUs were generated from responses to Short-Form Six-Dimension (SF-6D) MAUI, and quality-of-the-diet by Diet Habits Questionnaire (DHQ). Cross-sectional, and short- and long-term prospective associations of DHQ with HSU evaluated by linear regression at 2.5- and 5-years. Pooled prospective associations between DHQ and HSU evaluated using linear and quantile regression. Analyses adjusted for relevant demographic and clinical covariates. RESULTS: Among 839 participants, baseline DHQ scores showed short- and long-term associations with subsequent HSU, each 10-unit increase in total DHQ score associated with 0.008-0.012 higher HSU (out of 1.00). These associations were dose-dependent, those in the top two quartiles of baseline DHQ scores having 0.01-0.03 higher HSU at follow-up, 0.03 being the threshold for a minimally clinically important difference. Fat, fiber, and fruit/vegetable DHQ subscores were most strongly and consistently associated with better HSU outcomes. However, baseline meat and dairy consumption were associated with 0.01-0.02 lower HSU at subsequent follow-up. CONCLUSIONS: A higher quality-of-the-diet showed robust prospective relationships with higher HSUs 2.5- and 5-years later, substantiating previous cross-sectional relationships in this cohort. Subject to replication, these results suggest interventions to improve the quality-of-the-diet may be effective to improve HRQoL in people living with MS.

A repeated cross-sectional and longitudinal study of mental health and wellbeing during COVID-19 lockdowns in Victoria, Australia.

BACKGROUND: Population surveys across the world have examined the impact of the COVID-19 pandemic on mental health. However, few have simultaneously examined independent cross-sectional data with longitudinal data, each of which have different strengths and weaknesses and facilitate the investigation of distinct research questions. This study aimed to investigate psychological distress and life satisfaction during the first and second lockdowns in the state of Victoria, Australia, and the social factors that may be affected by lockdowns and could affect mental health. METHODS: The VicHealth Victorian Coronavirus Wellbeing Impact Study included two 20-min opt-in online panel surveys conducted in May and September 2020 in Victoria, each with a sample of 2000 adults aged 18 + . A two-part study design was used: a repeated cross-sectional study of respondents who participated in Survey One and Survey Two, followed by a longitudinal nested cohort study. The primary exposures were social solidarity, social connectedness and staying connected with family and friends. Using logistic regression modelling, we explored the associations between our exposures and primary outcomes of psychological distress and life satisfaction with and without adjustment for covariates, both cross-sectionally and longitudinally. The results from the multivariable models were summarised using adjusted Odds Ratios (aOR), 95% Confidence Intervals (CI). RESULTS: Cross-sectional results indicated that the percentage of participants with low life satisfaction was significantly higher in the second survey sample (53%) compared to the first (47%). The percentage of participants with high psychological distress was higher but not significantly different between the two survey samples (14% first survey vs 16% second survey). Longitudinal study results indicated that lower social connectedness was significantly associated with higher psychological distress (aOR:3.3; 95% CI: 1.3-8.4) and lower life satisfaction (aOR:0.2; 95% CI: 0.1-0.4). Younger adults had higher psychological distress compared to older adults (aOR:6.8; 95% CI:1.5-31.1). Unemployment at the time of the first survey was significantly associated with lower life satisfaction at the second survey (aOR:0.5; 95% CI: 0.3-0.9). CONCLUSION: This study supports the findings of other international studies. It also highlights the need to promote increased social connection and maintain it at times of isolation and separation, particularly amongst younger adults.

Epididymitis rates in Australian hospitals 2009-2018: ecological analysis.

OBJECTIVES: To investigate rates of acute epididymitis diagnosed in Australian hospital settings. METHODS: Yearly hospital admission and emergency department (ED) rates of epididymitis as primary diagnoses were calculated for 15-44-year-old men for three states (Victoria, New South Wales, Queensland) from 2009 to 2014 using population denominators. Zero inflated Poisson regression models were used to analyse variation in rates by year, age, and residential area. Additionally, we investigated national epididymitis admission trends from 2009 to 2018 using generalised linear models. RESULTS: Between 2009 and 2014, there was a total of 7375 admissions and 17 281 ED presentations for which epididymitis was the main reason for care. Most epididymitis diagnoses (94.0% in admissions, 99.7% in EDs) were without abscess, and 2.5% of admissions were for chlamydial epididymitis. Almost a quarter (23.3%) of epididymitis diagnosed in EDs resulted in hospital admission. In 2014, the epididymitis rate per 100 000 men was 38.7 in admissions and 91.9 in EDs. Comparing 2014 with 2009, the overall epididymitis diagnosis rate increased in admissions by 32% (adjusted incident rate ratio (aIRR) 1.32, 95% CI 1.20 to 1.44) and in ED attendances by 40% (aIRR 1.40, 95% CI 1.31 to 1.49). By age, the highest rates were among men 35-44 years in admissions and men 15-24 years in EDs. National admission rates of epididymitis during 2009-2018 showed a similar pattern. CONCLUSION: Rates of epididymitis diagnosis in hospital admission and ED presentations increased. Different age-related rates in these settings suggest a different aetiology or differential severity by age group.

Two healthy lifestyle scores are associated with lower subsequent fatigue risk using inverse probability weighting in an international longitudinal cohort of people with multiple sclerosis.

BACKGROUND AND PURPOSE: Several modifiable lifestyle factors have been associated with the onset and health outcomes of multiple sclerosis (MS), including clinically significant fatigue. A combined lifestyle score approach represents one method of assessing their relationship with clinical outcomes. The aim was to examine the association of two lifestyle scores with clinically significant fatigue and change thereof over 2.5 years' follow-up using inverse probability treatment weighting (IPTW). METHODS: Data on sociodemographic, lifestyle, and clinical characteristics surveyed from an international cohort of people with MS at baseline and at 2.5-year follow-up were used. Fatigue was defined by the Fatigue Severity Scale (FSS >5) and healthy lifestyle by the Healthy Lifestyle Index Score (HLIS) and the Smoking, Nutrition, Alcohol Consumption and Physical Activity (SNAP) score. Analyses were by IPTW accounting for age, sex, MS type, disability, treated comorbidity number, immunomodulatory medication use, prescription antifatigue medication use, and ongoing relapse symptoms. RESULTS: In total, 1268 participants completed the FSS at both time points; approximately 62% had fatigue. Using doubly robust IPTW, high (>11/20) HLIS (odds ratio [OR] 0.90, 95% confidence interval [CI] 0.81-0.98) and high (>3/5) SNAP (OR 0.82, 95% CI 0.73-0.90) were each associated with lower risk of fatigue at follow-up. Evaluating change in fatigue, a higher SNAP score was associated with a lower risk of fatigue (OR 0.89, 95% CI 0.80-0.97) but the score for HLIS did not reach statistical significance (OR 0.93, 95% CI 0.85-1.01). CONCLUSION: These results suggest a robust role for key lifestyle factors in preventing clinically significant fatigue and may represent a place for lifestyle modification in improving clinical outcomes in MS.

Viability of a MSQOL-54 general health-related quality of life score using bifactor model.

BACKGROUND: MSQOL-54 is a multidimensional, widely-used, health-related quality of life (HRQOL) instrument specific for multiple sclerosis (MS). Findings from the validation study suggested that the two MSQOL-54 composite scores are correlated. Given this correlation, it could be assumed that a unique total score of HRQOL may be calculated, with the advantage to provide key stakeholders with a single overall HRQOL score. We aimed to assess how well the bifactor model could account for the MSQOL-54 structure, in order to verify whether a total HRQOL score can be calculated. METHODS: A large international database (3669 MS patients) was used. By means of confirmatory factor analysis, we estimated a bifactor model in which every item loads onto both a general factor and a group factor. Fit of the bifactor model was compared to that of single and two second-order factor models by means of Akaike information and Bayesian information criteria reduction. Reliability of the total and subscale scores was evaluated with Mc Donald's coefficients (omega, and omega hierarchical). RESULTS: The bifactor model outperformed the two second-order factor models in all the statistics. All items loaded satisfactorily (≥ 0.40) on the general HRQOL factor, except the sexual function items. Omega coefficients for total score were very satisfactory (0.98 and 0.87). Omega hierarchical for subscales ranged between 0.22 to 0.57, except for the sexual function (0.70). CONCLUSIONS: The bifactor model is particularly useful when it is intended to acknowledge multidimensionality and at the same time take account of a single general construct, as the HRQOL related to MS. The total raw score can be used as an estimate of the general HRQOL latent score.

Multiple imputation methods for handling missing values in longitudinal studies with sampling weights: Comparison of methods implemented in Stata.

Many analyses of longitudinal cohorts require incorporating sampling weights to account for unequal sampling probabilities of participants, as well as the use of multiple imputation (MI) for dealing with missing data. However, there is no guidance on how MI and sampling weights should be implemented together. We simulated a target population based on the Australian Bureau of Statistics Estimated Resident Population and drew 1000 random samples dependent on three design variables to mimic the Longitudinal Study of Australian Children. The target analysis was the weighted prevalence of overweight/obesity over childhood. We evaluated the performance of several MI approaches available in Stata, based on multivariate normal imputation (MVNI), fully conditional specification (FCS) and twofold FCS: a weighted imputation model, imputing missing data separately for each quintile sampling weight grouping, including the design stratum indicator in the imputation model, and using sampling weights as a covariate in the imputation model. Approaches based on available cases and inverse probability weighting (IPW), with time-varying weights, were also compared. We observed severe issues of convergence with FCS and twofold FCS. All MVNI-based approaches performed similarly, producing minimal bias and nominal coverage, except for when imputation was conducted separately for each quintile sampling weight group. IPW performed equally as well as MVNI-based approaches in terms of bias, however, was less precise. In similar longitudinal studies, we recommend using MVNI with the design stratum as a covariate in the imputation model. If this is unknown, including the sampling weight as a covariate is an appropriate alternative.

SF-6D health state utilities for lifestyle, sociodemographic and clinical characteristics of a large international cohort of people with multiple sclerosis.

BACKGROUND: While many studies have examined the impacts of multiple sclerosis (MS) on health-related quality of life (HRQoL), none have used the SF-6D multi-attribute utility instrument in a large international cohort (> 2000 subjects) of people with MS. OBJECTIVES: To derive SF-6D health state utilities (HSUs) for participants of the HOLISM (Health Outcomes and Lifestyle In a Sample of people with Multiple Sclerosis) international cohort and to describe the distribution and determinants thereof. METHODS: HSUs were generated using the SF-6D for participants with sufficient SF-36 data [n = 2185/2466 (88.6%)]. Mean HSUs for sociodemographic, clinical and modifiable lifestyle factors (including diet, physical activity, supplement use) were evaluated. Determinants of HSU were then evaluated by linear regression, adjusted for age, sex, MS type, disability, fatigue, and prescription antidepressant use. RESULTS: Mean HSU for the sample was 0.67 (SD = 0.13) and diminished with increasing MS-related disability, robust to adjustment, supporting the SF-6D's discriminatory power in people with MS. Severe disability and clinically significant fatigue were each associated with 11% lower HSU (95% CI = - 0.13, - 0.10 and - 0.12, - 0.10), and depression risk with 10%-lower HSU (95% CI = - 0.11, - 0.08). Employment, higher socioeconomic and married/partnered statuses, larger social-network size, greater physical activity, and vitamin D and omega-3 supplement use were associated with significantly higher HSU, and overweight/obese BMI and tobacco smoking with lower HSU. Age, sex, and education were not associated. CONCLUSION: Modifiable lifestyle factors including healthy diet, increased physical activity and supplement use were associated with higher HRQOL among people with MS. The SF-6D instrument revealed significant discriminatory power in this international cohort of people with MS.

Assessing measurement invariance of MSQOL-54 across Italian and English versions.

PURPOSE: The Multiple Sclerosis Quality of Life-54 (MSQOL-54) is a specific multiple sclerosis (MS) health-related quality of life inventory consisting of 52 items organized into 12 subscales plus two single items. No study was found in literature assessing its measurement invariance across language versions. We investigated whether MSQOL-54 items provide unbiased measurements of underlying constructs across Italian and English versions. METHODS: Three constrained levels of measurement invariance were evaluated: configural invariance where equivalent numbers of factors/factor patterns were required; metric invariance where equivalent factor loadings were required; and scalar invariance where equivalent item intercepts between groups were required. Comparative fit index (CFI), root mean square error of approximation (RMSEA), and standardized root mean square residual (SRMR) fit indices and their changes between nested models were used to assess tenability of invariance constraints. RESULTS: Overall, the dataset included 3669 MS patients: 1605 (44%) Italian, mean age 41 years, 62% women, 69% with mild level of disability; 2064 (56%) English-speaking (840 [41%] from North America, 797 [39%] from Australasia, 427 [20%] from UK and Ireland), mean age 46 years, 83% women, 54% with mild level of disability. The configural invariance model showed acceptable fit (RMSEA = 0.052, CFI = 0.904, SRMR = 0.046); imposing loadings and intercepts equality constraints produced negligible worsening of fit (ΔRMSEA < 0.001, ΔCFI = - 0.002, ΔSRMR = 0.002 for metric invariance; ΔRMSEA = 0.003, ΔCFI = - 0.013, ΔSRMR = 0.003 for scalar invariance). CONCLUSIONS: These findings support measurement invariance of the MSQOL-54 across the two language versions, suggesting that the questionnaire has the same meaning and the same measurement paramaters in the Italian and English versions.

Multiple imputation methods for handling missing values in a longitudinal categorical variable with restrictions on transitions over time: a simulation study.

BACKGROUND: Longitudinal categorical variables are sometimes restricted in terms of how individuals transition between categories over time. For example, with a time-dependent measure of smoking categorised as never-smoker, ex-smoker, and current-smoker, current-smokers or ex-smokers cannot transition to a never-smoker at a subsequent wave. These longitudinal variables often contain missing values, however, there is little guidance on whether these restrictions need to be accommodated when using multiple imputation methods. Multiply imputing such missing values, ignoring the restrictions, could lead to implausible transitions. METHODS: We designed a simulation study based on the Longitudinal Study of Australian Children, where the target analysis was the association between (incomplete) maternal smoking and childhood obesity. We set varying proportions of data on maternal smoking to missing completely at random or missing at random. We compared the performance of fully conditional specification with multinomial and ordinal logistic imputation, and predictive mean matching, two-fold fully conditional specification, indicator based imputation under multivariate normal imputation with projected distance-based rounding, and continuous imputation under multivariate normal imputation with calibration, where each of these multiple imputation methods were applied, accounting for the restrictions using a semi-deterministic imputation procedure. RESULTS: Overall, we observed reduced bias when applying multiple imputation methods with restrictions, and fully conditional specification with predictive mean matching performed the best. Applying fully conditional specification and two-fold fully conditional specification for imputing nominal variables based on multinomial logistic regression had severe convergence issues. Both imputation methods under multivariate normal imputation produced biased estimates when restrictions were not accommodated, however, we observed substantial reductions in bias when restrictions were applied with continuous imputation under multivariate normal imputation with calibration. CONCLUSION: In a similar longitudinal setting we recommend the use of fully conditional specification with predictive mean matching, with restrictions applied during the imputation stage.

Effectiveness of Exercise Interventions for Pain Reduction in People With Multiple Sclerosis: A Systematic Review and Meta-analysis of Randomized Controlled Trials.

OBJECTIVE: To systematically review the evidence of the effect of exercise compared with passive control on pain in people with multiple sclerosis. DATA SOURCE AND STUDY SELECTION: Five electronic databases were searched for randomized controlled trials published up to March 2017 that recruited people with multiple sclerosis where exercise was the intervention and pain was an outcome (PROSPERO registration number CRD42017060489). STATISTICAL ANALYSIS: A random-effects meta-analysis was conducted to estimate the standardized mean difference of the effect of exercise on pain between treatment and control groups. We assessed risk of bias, fitted meta-regression models to explore heterogeneity between studies, and assessed small study effects. DATA SYNTHESIS: Ten studies met the inclusion criteria (total sample size=389), and all studies were at high risk of bias. We found that exercise interventions were associated with less pain compared with passive control groups (standardized mean difference=-.46; 95% CI, -.92 to .00). There was high between-study heterogeneity (I2=77.0%), which was not explained by the prespecified study characteristics. There was also some evidence of small study effects. CONCLUSION: This is the first systematic review of the effect of exercise interventions on pain in people with multiple sclerosis, a chronic neurological disorder that affects 2.5 million people. We found some evidence that exercise compared with passive control alleviates pain in this population, but there were limitations in reporting and study quality with high risk of bias of individual studies and heterogeneity between studies.

Trends in diagnosis of pelvic inflammatory disease in an Australian sexual health clinic, 2002-16: before and after clinical audit feedback.

Background A 2006 Australian sexual health clinic audit of pelvic inflammatory disease (PID) diagnosis rates found variability between doctors. Doctors were given audit feedback towards increasing diagnosis and reducing variability. The clinic implemented other improvements to increase capacity. This study investigated PID diagnosis time trends before and after feedback. METHODS: Yearly PID diagnosis rates for women aged 16-49 years attending the clinic (2002-16) were calculated. Using multivariable generalised linear mixed models, adjusted for patient risk and lower genital infection (any of chlamydia, gonorrhoea, Mycoplasma genitalium, bacterial vaginosis) and stratified by before (2002-June 2007) and after (July 2007-2016) feedback, we assessed if PID rates changed over time, accounting for between-doctor variability. RESULTS: During 2002-16, 144 doctors undertook 84476 female consultations and diagnosed 1755 (2.1%, 95% confidence interval (CI) 2.0-2.2) with PID. Comparing 2002-03 to 2015-16, the yearly PID rate increased; 0.8% (37/4836) to 2.9% (209/7088). Comparing before and after feedback more women reported any symptoms at triage (35.1%-47.2%) or had a lower genital infection diagnosed (10.1%-14.9%). After feedback, PID rates increased by 8% yearly (incidence rate ratio (IRR) 1.08, 95% CI 1.06-1.11), but were unchanged (adjusted IRR (aIRR) 1.01, 95% CI 0.98-1.03) after adjustment for patient characteristics. Factors associated with PID were self-reported symptoms, younger age and a lower genital infection. Lower variability in doctor-specific rates was observed after feedback. CONCLUSIONS: Increasing PID diagnosis rates appeared to be driven by a greater female patient risk profile, influenced by increased capacity following service improvements.

NormalizeMets: assessing, selecting and implementing statistical methods for normalizing metabolomics data.

INTRODUCTION: In metabolomics studies, unwanted variation inevitably arises from various sources. Normalization, that is the removal of unwanted variation, is an essential step in the statistical analysis of metabolomics data. However, metabolomics normalization is often considered an imprecise science due to the diverse sources of variation and the availability of a number of alternative strategies that may be implemented. OBJECTIVES: We highlight the need for comparative evaluation of different normalization methods and present software strategies to help ease this task for both data-oriented and biological researchers. METHODS: We present NormalizeMets-a joint graphical user interface within the familiar Microsoft Excel and freely-available R software for comparative evaluation of different normalization methods. The NormalizeMets R package along with the vignette describing the workflow can be downloaded from https://cran.r-project.org/web/packages/NormalizeMets/ . The Excel Interface and the Excel user guide are available on https://metabolomicstats.github.io/ExNormalizeMets . RESULTS: NormalizeMets allows for comparative evaluation of normalization methods using criteria that depend on the given dataset and the ultimate research question. Hence it guides researchers to assess, select and implement a suitable normalization method using either the familiar Microsoft Excel and/or freely-available R software. In addition, the package can be used for visualisation of metabolomics data using interactive graphical displays and to obtain end statistical results for clustering, classification, biomarker identification adjusting for confounding variables, and correlation analysis. CONCLUSION: NormalizeMets is designed for comparative evaluation of normalization methods, and can also be used to obtain end statistical results. The use of freely-available R software offers an attractive proposition for programming-oriented researchers, and the Excel interface offers a familiar alternative to most biological researchers. The package handles the data locally in the user's own computer allowing for reproducible code to be stored locally.

Rates of pelvic inflammatory disease and ectopic pregnancy in Australia, 2009-2014: ecological analysis of hospital data.

OBJECTIVE: To analyse yearly rates of pelvic inflammatory disease (PID) and ectopic pregnancy (EP) diagnosed in hospital settings in Australia from 2009 to 2014. METHODS: We calculated yearly PID and EP diagnosis rates in three states (Victoria, New South Wales, Queensland) for women aged 15-44 years using hospital admissions and emergency department (ED) attendance data, with population and live birth denominators. We stratified PID diagnoses as chlamydial-related or gonorrhoeal-related (Chlamydia trachomatis (CT)-related or Neisseria gonorrhoeae (NG)-related), acute, unspecified and chronic, and analysed variations by year, age and residential area using Poisson regression models. RESULTS: For PID, the rate of all admissions in 2014 was 63.3 per 100 000 women (95% CI 60.8 to 65.9) and of all presentations in EDs was 97.0 per 100 000 women (95% CI 93.9 to 100.2). Comparing 2014 with 2009, the rate of all PID admissions did not change, but the rate of all presentations in EDs increased (adjusted incidence rate ratio (aIRR) 1.34, 95% CI 1.24 to 1.45), and for admissions by PID category was higher for CT-related or NG-related PID (aIRR 1.73, 95% CI 1.31 to 2.28) and unspecified PID (aIRR 1.09, 95% CI 1.00 to 1.19), and lower for chronic PID (aIRR 0.84, 95% CI 0.74 to 0.95). For EP, in 2014 the rate of all admissions was 17.4 (95% CI 16.9 to 17.9) per 1000 live births and of all ED presentations was 15.6 (95% CI 15.1 to 16.1). Comparing 2014 with 2009, the rates of all EP admissions (aIRR 1.06, 95% CI 1.04 to 1.08) and rates in EDs (aIRR 1.24, 95% CI 1.18 to 1.31) were higher. CONCLUSIONS: PID and EP remain important causes of hospital admissions for female STI-associated complications. Hospital EDs care for more PID cases than inpatient departments, particularly for young women. Updated primary care data are needed to better understand PID epidemiology and healthcare usage.

Characteristics of pelvic inflammatory disease where no sexually transmitted infection is identified: a cross-sectional analysis of routinely collected sexual health clinic data.

OBJECTIVES: Pelvic inflammatory disease (PID) occurs when pathogens, often sexually transmitted, ascend to the upper genital tract, yet a causative pathogen is not detected in a substantial proportion of diagnosed PID. We assessed the characteristics associated with PID in women in whom chlamydia, gonorrhoea, Mycoplasma genitalium (MG) and bacterial vaginosis (BV) were not detected ('pathogen-negative-PID'). METHODS: Cross-sectional analysis of routinely collected clinical data from new female patients attending a sexual health clinic between 2006 and 2013. Women were eligible if they had been diagnosed with PID and tested for genital chlamydia, gonorrhoea, MG and BV. Logistic regression was conducted to identify characteristics associated with pathogen-negative-PID. RESULTS: Among 330 women with clinically diagnosed PID, 204 (61.8%, 95% CI 56.3% to 67.1%) had pathogen-negative-PID. Compared with pathogen-positive-PID, pathogen-negative-PID cases were more likely to be aged ≥30 years (adjusted odds ratio (AOR) 1.7, 95% CI 1.0 to 3.0), had less evidence of vaginal inflammation (AOR 0.5, 95% CI 0.3 to 0.9) and reported less unprotected sex (AOR 0.6, 95% CI 0.4 to 1.0). CONCLUSIONS: These findings highlight uncertainties around PID diagnosis and aetiology. Pathogen-negative-PID could represent (i) a false positive diagnosis where the woman does not have a sexually transmitted infection (STI) or PID, (ii) PID of another microbiological aetiology or associated with a past STI or (iii) PID where the cervical infection has cleared. However, until diagnostic biomarkers are available, PID treatment should be based on clinical features and sexual risk.

A comparison of multiple imputation methods for handling missing values in longitudinal data in the presence of a time-varying covariate with a non-linear association with time: a simulation study.

BACKGROUND: Missing data is a common problem in epidemiological studies, and is particularly prominent in longitudinal data, which involve multiple waves of data collection. Traditional multiple imputation (MI) methods (fully conditional specification (FCS) and multivariate normal imputation (MVNI)) treat repeated measurements of the same time-dependent variable as just another 'distinct' variable for imputation and therefore do not make the most of the longitudinal structure of the data. Only a few studies have explored extensions to the standard approaches to account for the temporal structure of longitudinal data. One suggestion is the two-fold fully conditional specification (two-fold FCS) algorithm, which restricts the imputation of a time-dependent variable to time blocks where the imputation model includes measurements taken at the specified and adjacent times. To date, no study has investigated the performance of two-fold FCS and standard MI methods for handling missing data in a time-varying covariate with a non-linear trajectory over time - a commonly encountered scenario in epidemiological studies. METHODS: We simulated 1000 datasets of 5000 individuals based on the Longitudinal Study of Australian Children (LSAC). Three missing data mechanisms: missing completely at random (MCAR), and a weak and a strong missing at random (MAR) scenarios were used to impose missingness on body mass index (BMI) for age z-scores; a continuous time-varying exposure variable with a non-linear trajectory over time. We evaluated the performance of FCS, MVNI, and two-fold FCS for handling up to 50% of missing data when assessing the association between childhood obesity and sleep problems. RESULTS: The standard two-fold FCS produced slightly more biased and less precise estimates than FCS and MVNI. We observed slight improvements in bias and precision when using a time window width of two for the two-fold FCS algorithm compared to the standard width of one. CONCLUSION: We recommend the use of FCS or MVNI in a similar longitudinal setting, and when encountering convergence issues due to a large number of time points or variables with missing values, the two-fold FCS with exploration of a suitable time window.

Genomic prediction of coronary heart disease.

AIMS: Genetics plays an important role in coronary heart disease (CHD) but the clinical utility of genomic risk scores (GRSs) relative to clinical risk scores, such as the Framingham Risk Score (FRS), is unclear. Our aim was to construct and externally validate a CHD GRS, in terms of lifetime CHD risk and relative to traditional clinical risk scores. METHODS AND RESULTS: We generated a GRS of 49 310 SNPs based on a CARDIoGRAMplusC4D Consortium meta-analysis of CHD, then independently tested it using five prospective population cohorts (three FINRISK cohorts, combined n = 12 676, 757 incident CHD events; two Framingham Heart Study cohorts (FHS), combined n = 3406, 587 incident CHD events). The GRS was associated with incident CHD (FINRISK HR = 1.74, 95% confidence interval (CI) 1.61-1.86 per S.D. of GRS; Framingham HR = 1.28, 95% CI 1.18-1.38), and was largely unchanged by adjustment for known risk factors, including family history. Integration of the GRS with the FRS or ACC/AHA13 scores improved the 10 years risk prediction (meta-analysis C-index: +1.5-1.6%, P < 0.001), particularly for individuals ≥60 years old (meta-analysis C-index: +4.6-5.1%, P < 0.001). Importantly, the GRS captured substantially different trajectories of absolute risk, with men in the top 20% of attaining 10% cumulative CHD risk 12-18 y earlier than those in the bottom 20%. High genomic risk was partially compensated for by low systolic blood pressure, low cholesterol level, and non-smoking. CONCLUSIONS: A GRS based on a large number of SNPs improves CHD risk prediction and encodes different trajectories of lifetime risk not captured by traditional clinical risk scores.