Laura Vanderbloemen

Visualizing Europe’s demographic scars with coplots and contour plots

We present two enhancements to existing methods for visualizing vital statistics data. Data from the Human Mortality Database were used and vital statistics from England and Wales are used for illustration. The simpler of these methods involves coplotting mean age of death with its variance, and the more complex of these methods is to present data as a contour plot. The coplot method shows the effect of the 20th century’s epidemiological transitions. The contour plot method allows more complex and subtle age, period and cohort effects to be seen. The contour plot shows the effects of broad improvements in public health over the 20th century, including vast reductions in rates of childhood mortality, reduced baseline mortality risks during adulthood and the postponement of higher mortality risks to older ages. They also show the effects of the two world wars and the 1918 influenza pandemic on men of fighting age, women and children. The contour plots also show a cohort effect for people born around 1918, suggesting a possible epigenetic effect of parental exposure to the pandemic which shortened the cohort’s lifespan and which has so far received little attention. Although this article focuses on data from England and Wales, the associated online appendices contain equivalent visualizations for almost 50 series of data available on the Human Mortality Database. We expect that further analyses of these visualizations will reveal further insights into global public health.

Measuring expectations of benefit from treatment in acupuncture trials: A systematic review ☆

OBJECTIVES We conducted a systematic review that aimed to document and describe how (1) expectation of benefit from treatment (response expectancies) were measured and reported in acupuncture trials, and (2) examine any effect on outcomes. DESIGN We searched MEDLINE, EMBASE, AMED, CIHAHL, CENTRAL and Science and Technology Proceedings up to November 2007 for randomised (RCT) and quasi-randomised (CCT) controlled trials and prospective controlled cohorts of acupuncture as treatment for a medical or psychological condition in adults. An update citation search was conducted in April 2010. We included studies that mentioned soliciting response expectancies. RESULTS We found 58 RCTs that fulfilled our inclusion criteria. Around half referenced one of five published instruments, most of which were designed to measure sham credibility and included one question on response expectancy. A wide range of question phrasing and response scales was used. There was some evidence that response scales may influence the measurement of expectations. Eight trials analysed the association between pre-randomisation expectations for assigned treatment and outcomes, and six the effect of pre-randomisation expectations across all patients independent of treatment allocation. Some showed associations but others did not. CONCLUSIONS There is some evidence that response expectancies interact with outcomes in acupuncture trials however the variety of question phrasing and analysis methods precludes drawing a firm conclusion about for whom and under which circumstance. To further our understanding of expectations, more methodological work is needed to standardise the questions and response scales that are used.

Acanthosis nigricans predicts the clustering of metabolic syndrome components in Hispanic elementary school-aged children

Abstract Background: Acanthosis nigricans (AN) is a dermatologic condition associated with hyperinsulinemia, a marker of insulin resistance that is the principal abnormality in metabolic syndrome (MetS). We examined the association of AN with the clustering of MetS components. Methods: A cross-sectional study was conducted in an urban school-based health center in New Mexico. Students without diabetes were evaluated for AN, a family history of type 2 diabetes, body mass index (BMI), and MetS components. The clustering of MetS components by BMI category and AN status was assessed by comparing the group means of summed average z-scores of fasting insulin, triglycerides, high-density lipoprotein-cholesterol, and systolic blood pressure among the students. A multivariate model with BMI category and AN status controlling for Tanner stage was performed to identify the variables associated with the clustering of MetS components. Results: Complete data were available for 90 children (age, 9.7±1.4 years; 94% Hispanic; 60% female). In multivariate modeling of MetS cluster z-score, significant differences were found between the students with BMI <85th percentile [–0.27; 95% confidence interval (95% CI)=–0.42 to –0.11] and (a) the students with BMI 85th–94.9th percentile with AN (0.74; 95% CI=0.17–1.31) and (b) the students with BMI ≥95th percentile with AN (0.86; 95% CI=0.54–1.18). No significant differences in the MetS cluster z-score were seen between the students with BMI <85th percentile and those with BMI 85th–94.9th percentile without AN (0.24; 95% CI=–0.33 to 0.81) or those with BMI ≥95th percentile without AN (0.31; 95% CI=–0.13 to 0.75). Conclusions: Overweight/obese Hispanic elementary school-aged children with AN exhibit clustering of MetS components and could benefit from early intervention.

Visualising and quantifying ‘excess deaths’ in Scotland compared with the rest of the UK and the rest of Western Europe

Background Scotland has higher mortality rates than the rest of Western Europe (rWE), with more cardiovascular disease and cancer among older adults; and alcohol-related and drug-related deaths, suicide and violence among younger adults. Methods We obtained sex, age-specific and year-specific all-cause mortality rates for Scotland and other populations, and explored differences in mortality both visually and numerically. Results Scotlands age-specific mortality was higher than the rest of the UK (rUK) since 1950, and has increased. Between the 1950s and 2000s, ‘excess deaths’ by age 80 per 100 000 population associated with living in Scotland grew from 4341 to 7203 compared with rUK, and from 4132 to 8828 compared with rWE. UK-wide mortality risk compared with rWE also increased, from 240 ‘excess deaths’ in the 1950s to 2320 in the 2000s. Cohorts born in the 1940s and 1950s throughout the UK including Scotland had lower mortality risk than comparable rWE populations, especially for males. Mortality rates were higher in Scotland than rUK and rWE among younger adults from the 1990s onwards suggesting an age–period interaction. Conclusions Worsening mortality among young adults in the past 30 years reversed a relative advantage evident for those born between 1950 and 1960. Compared with rWE, Scotland and rUK have followed similar trends but Scotland has started from a worse position and had worse working age–period effects in the 1990s and 2000s.

Two cheers for a small giant? Why we need better ways of seeing data: A commentary on: 'Rising morbidity and mortality in midlife among White non-Hispanic Americans in the 21st century'

Two cheers for a small giant? Why we need better ways of seeing data: A commentary on: ‘Rising morbidity and mortality in midlife among White non-Hispanic Americans in the 21st century’ Jonathan Minton,* Mark Green, Gerry McCartney, Richard Shaw, Laura Vanderbloemen, Kate Pickett University of Glasgow College of Social Sciences Room 110, 25 Bute Gardens, Glasgow G12 8RS Email: jonathan.minton@glasgow.ac.uk, University of Liverpool, School of Environmental Sciences, NHS Health Scotland, Public Health Science, University of Glasgow, Medical Research Council/Chief Scientist Office, Social and Public Health Sciences Unit, 5 Imperial College, Faculty of Medicine, Department of Primary Care and Public Health, University of York, Health Sciences

Smoking and quit attempts during pregnancy and postpartum: a longitudinal UK cohort

Objectives Pregnancy motivates women to try stopping smoking, but little is known about timing of their quit attempts and how quitting intentions change during pregnancy and postpartum. Using longitudinal data, this study aimed to document women’s smoking and quitting behaviour throughout pregnancy and after delivery. Design Longitudinal cohort survey with questionnaires at baseline (8–26 weeks’ gestation), late pregnancy (34–36 weeks) and 3 months after delivery. Setting Two maternity hospitals in one National Health Service hospital trust, Nottingham, England. Participants 850 pregnant women, aged 16 years or over, who were current smokers or had smoked in the 3 months before pregnancy, were recruited between August 2011 and August 2012. Outcome measures Self-reported smoking behaviour, quit attempts and quitting intentions. Results Smoking rates, adjusting for non-response at follow-up, were 57.4% (95% CI 54.1 to 60.7) at baseline, 59.1% (95% CI 54.9 to 63.4) in late pregnancy and 67.1% (95% CI 62.7 to 71.5) 3 months postpartum. At baseline, 272 of 488 current smokers had tried to quit since becoming pregnant (55.7%, 95% CI 51.3 to 60.1); 51.3% (95% CI 44.7 to 58.0) tried quitting between baseline and late pregnancy and 27.4% (95% CI 21.7 to 33.2) after childbirth. The percentage who intended to quit within the next month fell as pregnancy progressed, from 40.4% (95% CI 36.1 to 44.8) at baseline to 29.7% (95% CI 23.8 to 35.6) in late pregnancy and 14.2% (95% CI 10.0 to 18.3) postpartum. Postpartum relapse was lower among women who quit in the 3 months before pregnancy (17.8%, 95% CI 6.1 to 29.4) than those who stopped between baseline and late pregnancy (42.9%, 95% CI 24.6 to 61.3). Conclusions Many pregnant smokers make quit attempts throughout pregnancy and postpartum, but intention to quit decreases over time; there is no evidence that smoking rates fall during gestation.

Visualising variation in mortality rates across the life course and by sex, USA and comparator states, 1933–2010

Background Previous research showed that younger adult males in the USA have, since the 1950s, died at a faster rate than females of the same age. In this paper, we quantify this difference, and explore possible explanations for the differences at different ages and in different years. Methods Using data from the Human Mortality Database (HMD), the number of additional male deaths per 10 000 female deaths was calculated for each year from 1933 to 2010, and for each year of age from 0 to 60 years, for the USA, and a number of other countries for comparison. The data were explored visually using shaded contour plots. Results Gender differences in excess mortality have increased. Coming of age (between the ages of 15 and 25 years of age) is especially perilous for men relative to women now compared with the past in the USA; the visualisations highlight this change as important. Conclusions Sex differences in mortality risks at various ages are not static. While women may today have an advantage when it comes to life expectancy, in the USA, this has greatly increased since the 1930s. Just as young adulthood for women has been made safer through safer antenatal and childbirth practices, changes in public policy can make the social environment safer for men.