Cara L Booker

A systematic review of the effect of retention methods in population-based cohort studies

BackgroundLongitudinal studies are of aetiological and public health relevance but can be undermined by attrition. The aim of this paper was to identify effective retention strategies to increase participation in population-based cohort studies.MethodsSystematic review of the literature to identify prospective population-based cohort studies with health outcomes in which retention strategies had been evaluated.ResultsTwenty-eight studies published up to January 2011 were included. Eleven of which were randomized controlled trials of retention strategies (RCT). Fifty-seven percent of the studies were postal, 21% in-person, 14% telephone and 7% had mixed data collection methods. A total of 45 different retention strategies were used, categorised as 1) incentives, 2) reminder methods, repeat visits or repeat questionnaires, alternative modes of data collection or 3) other methods. Incentives were associated with an increase in retention rates, which improved with greater incentive value. Whether cash was the most effective incentive was not clear from studies that compared cash and gifts of similar value. The average increase in retention rate was 12% for reminder letters, 5% for reminder calls and 12% for repeat questionnaires. Ten studies used alternative data collection methods, mainly as a last resort. All postal studies offered telephone interviews to non-responders, which increased retention rates by 3%. Studies that used face-to-face interviews increased their retention rates by 24% by offering alternative locations and modes of data collection.ConclusionsIncentives boosted retention rates in prospective cohort studies. Other methods appeared to have a beneficial effect but there was a general lack of a systematic approach to their evaluation.

Stressful Life Events, Smoking Behavior, and Intentions to Smoke among a Multiethnic Sample of Sixth Graders

Objectives. Adolescent smoking has been associated with stressful life events. However, few studies have examined the associations between stress, smoking intentions, and smoking behavior among a multiethnic sample of adolescents. Methods. We compiled a checklist of stressful life events relevant to multiethnic youth and administered it to 1,074 sixth-grade students in urban Los Angeles. Results. The most frequently reported stressful events were similar across ethnic groups and generations in the USA: test taking, chores, and arguments with friends. The events reported as the most severe were disturbances in family life, such as: death, arguments between parents, and illness or injury. Whites and Latinos had reported higher levels of ever smoking and intentions to smoke than Asian/Pacific Islanders (PIs). On the positive family-related events scale Latinos scored higher than did whites or Asian/PIs. Whites scored higher than Latinos or Asian/PIs on both negative peer-related and negative personal-related events. Associations were observed between total stress, stressful life events, and smoking behavior and intention to smoke. Total stress was associated with ever smokers, smoking intentions within the next year and in high school over the entire sample. Negative peer-related events were associated with intention to smoke within the next year, among Latinos. Among Asian/PIs negative peer-related events were associated with intention to smoke within the next year and in high school. Negative school-related events were significantly associated with ever smoking and intentions to smoke in the next year and marginally associated with intentions to smoke in high school among children born in the USA whose parents were also born in the USA. Negative peer-related events and positive personal-related events were significantly associated with intentions to smoke in the next year among children born in the USA whose parents were immigrants. Conclusions. Results suggest that negative peer- and school-related events may lead to increased risk of smoking behavior and intentions to smoke among multicultural adolescents.

Psychological well-being and reactions to multiple unemployment events: adaptation or sensitisation?

Background Unemployment has been negatively associated with psychological well-being. This study examines the effect of multiple unemployment spells, specifically whether people become sensitised or adapt to unemployment if they are previously employed or economically inactive. Methods Data come from waves 1–17 of the British Household Panel Survey. Psychological well-being was measured using the 12-item General Health Questionnaire (GHQ-12), and employment status was self-reported. Multilevel modelling was used to examine the effects of unemployment, overall and by previous employment status, on well-being. Results Without consideration of prior employment status, psychological well-being was poorer at each unemployment spell. Previously employed persons had significantly higher GHQ-12 scores at the first and second unemployment spells but not at the third spell (ptrend<0.0001). Previously economically inactive persons had poorer psychological well-being at all unemployment spells, with significantly higher scores at the third spell than those at the first two spells (ptrend=0.0004). Thus, those employed prior to all unemployment spells adapted, while those previously economically inactive became more sensitised with additional unemployment spells. Pre-study unemployment and average annual household income moderated the effects of unemployment; effects varied by previous employment status and unemployment spell number. Conclusions The findings suggest that initially employed people who experience repeated unemployment cope better psychologically if they are able to regain employment in between unemployment spells. Those who make several attempts to re-enter the labour market following economic inactivity have a more difficult time, becoming more distressed with each try. This has implications for people affected by welfare to work policies.

A longitudinal analysis of stressful life events, smoking behaviors, and gender differences in a multicultural sample of adolescents.

Many studies have addressed the associations between stressful life events and adolescent smoking. Few studies, however, have examined gender differences, specifically with multicultural samples. This longitudinal study examines the relationship between 6 stress subscales and smoking behaviors 716 multicultural U.S. adolescents living in the greater Los Angeles area in 2000–2001. At baseline the ethnic break-down of the sample was 63% Latino and 26% Asian/PI and 70% were 11 years of age. Negative personal events were associated with lifetime smoking and negative school events were associated with intentions to smoke. Stratification of the sample by gender indicated that gender confounded the relationship between negative personal stress and lifetime smoking and negative school stress, positive personal stress and intentions to smoke. Two significant interactions were found. Findings indicate there are differential effects of stressful events between genders which may lead to smoking experimentation or intentions to smoke. Implications and limitations are discussed.

Media Use, Sports Participation, and Well-Being in Adolescence: Cross-Sectional Findings From the UK Household Longitudinal Study

Objectives. We investigated the relationship between selected types of screen-based media (SBM) use, total SBM use, sports participation, and markers of well-being. Methods. Data came from the youth panel (n = 4899) of Understanding Society, the UK Household Longitudinal Study, conducted in 2009. Well-being was measured by the Strengths and Difficulties Questionnaire and markers of happiness in different life domains. Results. The majority of young people used multiple types of SBM for at least 1 hour per day; only 30% participated in sports every day. Overall, young people with heavy SBM use were less happy than moderate users and more likely to have socioemotional difficulties. Chatting on social networking Web sites and game console use were associated with higher odds of socioemotional problems. Higher total SBM use was associated with lower odds of happiness and higher odds of socioemotional difficulties. Greater participation in sports was associated with higher odds of happiness and lower odds of socioemotional difficulties. Conclusions. Further longitudinal research could inform future interventions to reduce sedentary behavior and encourage healthy lifestyles among young people.

Well-Being in Adolescence—An Association With Health-Related Behaviors Findings From Understanding Society, the UK Household Longitudinal Study

The objective of this study was to investigate the demographic distribution of selected health-related behaviors and their relationship with different indicators of well-being. The data come from Wave 1 of the youth panel of Understanding Society household panel study. The Strengths and Difficulties Questionnaire (SDQ) measured socio-emotional difficulties. Markers of happiness in different life domains were combined to assess levels of happiness. Generally, younger youth participated in more health-protective behaviors, while older youth reported more health-risk behaviors. Higher consumption of fruit and vegetables and greater participation in sport were associated with higher odds of high happiness. Healthier eating was associated with lower odds of socio-emotional difficulties, while increased fast food consumption was associated with higher odds of socio-emotional difficulties. Smoking, drinking, and decreased sport participation were all associated with socio-emotional difficulties. Health-protective behaviors were associated with happiness, while health-risk behaviors were associated with socio-emotional difficulties.

Work-family life courses and markers of stress and inflammation in mid-life: evidence from the National Child Development Study

Abstract Background: This study investigated associations between work-family life courses and biomarkers of inflammation and stress in mid-life among British men and women. Gender differences in these associations were also explored. Methods: A novel statistical method—multi-channel sequence analysis—defined work-family life courses between the ages of 16 and 42 years, combining annual information on work, partnership and parenthood. Associations between work-family life courses and inflammation [C-reactive protein (CRP), fibrinogen and von Willebrand factor] and cortisol at age 44/45 years were tested using multivariate linear regression using multiply-imputed data on almost 6500 participants from the National Child Development Study 1958 British birth cohort. Results: Compared with those who combined strong ties to paid work with later transitions to stable family lives (‘Work, later family’ group), ‘Teen parents’ had higher CRP [40.6% higher, 95% confidence interval (CI): 5.6, 87.0] and fibrinogen (7.8% higher, 95% CI: 2.3, 13.5) levels, and homemakers (‘No paid work, early family’) had raised fibrinogen levels (4.7% higher, 95% CI: 0.7, 9.0), independent of childhood health and socioeconomic position, adult socioeconomic position, health behaviours and body mass index (BMI). Those who combined later transitions to stable family ties with a career break for childrearing had higher post-waking cortisol than the ‘Work, later family’ group; however, no associations were seen for other work-family types, therefore suggesting a null finding with cortisol. No statistically significant gender interactions in associations between work-family types and inflammatory or cortisol outcomes were found. Conclusions: Work-family life courses characterised by early parenthood or weak work ties were associated with a raised risk profile in relation to chronic inflammation.

Sexual orientation health inequality: Evidence from Understanding Society, the UK Longitudinal Household Study

Few studies from the United Kingdom have fully investigated inequalities between members of different sexual minority groups and heterosexuals over range of health outcomes. Using data from over 40,000 individuals, this study explores the health inequalities of sexual minority UK adults. We include respondents who identify as other and those who prefer not to say (PNS). Data come from wave three (2011-2012) of the nationally-representative Understanding Society, the UK Household Longitudinal Study. Sexual orientation was asked in the self-completion portion of the study. Markers of health include physical and mental functioning, minor psychological distress, self-rated health, substance use and disability. Multiple linear and logistic regression analyses tested for differences in markers of health between sexual orientation groups. Overall, heterosexual respondents had the best health while bisexual respondents had the worst. Gay and lesbian respondents reported poorer health than heterosexuals, specifically with regards to mental functioning, distress and illness status. The other and PNS respondents were most similar to each other and generally experienced fewer health inequalities than gay and lesbian respondents; they were less likely to use tobacco or alcohol. In sum, sexual minorities experience health inequality. The inclusion of other and PNS respondents has not been done in other studies and shows that while they may be healthier than gay/lesbian and bisexual respondents they still experiences poorer health than heterosexuals. Health promotion interventions are needed for these other and PNS individuals, who might not participate in interventions targeted toward known sexual minority groups.

OP53 Work-family life courses and markers of stress and inflammation in mid-life in the national child development study 1958 british birth cohort

Background Evidence from longitudinal studies has shown that stressful work and family situations are linked to chronic inflammation and altered cortisol profiles, both of which are risk factors for later disease. Conversely stable partnerships and work are known to be beneficial for health. However, few studies have previously investigated whether combined histories of work, partnerships and parenthood are related to later health, and in particular to markers of stress. The aim of this study was to investigate associations between work-family life courses (work, partnerships and parenthood) and biomarkers of inflammation and stress in mid-life amongst British men and women. Gender differences in these associations were also explored. Methods Multiply-imputed data on almost 6500 participants of the National Child Development Study 1958 British birth cohort were used in this study. A novel method - multi-channel sequence analysis – was used to create a typology of work-family life courses between 16–42 years, combining annual information on work, partnership and parenthood. Associations between work-family life courses and inflammation (C-reactive protein (CRP), fibrinogen and von Willebrand factor (vWF)) and cortisol at age 44/45 years were tested using multivariate linear regression. Estimates are presented as percentage difference as all outcomes were positively-skewed and therefore log-transformed. Results Compared to those who combined strong ties to paid work with later transitions to stable family lives (‘Later stable’ group), ‘Teen parents’ had higher CRP (40.6% higher, 95% CI: 5.6, 87.0) and fibrinogen (7.8% higher, 95% CI: 2.3, 13.5) levels, and ‘Homemakers’ had raised fibrinogen levels (4.7% higher, 95% CI: 0.7, 9.0), independent of childhood health and socioeconomic position, adult socioeconomic position, health behaviours and BMI. Those who combined later transitions to stable family ties with a career break for childrearing had higher post-waking cortisol (1.7% higher, 95% CI: 0.1, 3.3) than the ‘Later stable’ group. This was found to be mediated by health behaviours, particularly higher physical activity in this group. No statistically significant gender interactions in associations between work-family types and inflammatory or cortisol outcomes were found. Conclusion Work-family life courses characterised by early parenthood or weak work ties were linked to a raised risk profile in relation to stress-related biomarkers. The findings suggest that policies which enable parents to maintain strong ties to paid work, such as affordable childcare and flexible working, are likely to improve later health.

O2-2.1 Maximising the return from cohort studies

Introduction Cohort studies are important for understanding the aetiology underlying differences in disease incidence. Selective attrition is problematic as those at greater risk of ill-health are more likely to drop out, resulting in homogenous study populations with limited generalisability. Selective attrition may also bias estimates of association. For the benefits of cohort studies to be realised efforts must be made to minimise attrition and statistical methods for the analysis of studies with missing data must be developed to minimise bias. Methods We describe work examining methods used to maintain participation in cohort studies, identify best practice for reducing attrition and investigate methodologies suitable for the analysis of cohort studies with attrition under different circumstances. Results A literature search identified factors associated with minimising attrition including study design, recruitment procedures, incentives and retention methods utilised. A questionnaire including such factors was sent to 32 UK-based cohort studies; 25 (78%) returned a questionnaire. Analysis suggested that no one method is most effective, rather it is the combination of methods and study setting which may dictate the overall retention of study participants. Work with various cohort studies resulted in consideration of the appropriateness of methods for missing data. This has led to a review of inverse probability weighting (IPW) focusing on how and when to use it appropriately. The motivation for combining IPW and multiple imputation, and the theoretical justification for doing so, has been examined. A final development concerns the use of IPW when predictors of missingness are themselves missing.