Mental health-related quality of life and the timing of motherhood: a 16-year longitudinal study of a national cohort of young Australian women.

PURPOSE: We examine timing of motherhood in a longitudinal cohort of young Australian women, and its relationship with mental health-related quality of life (SF-36 MHI-5), and with sociodemographic, health behaviour and health-related variables. METHODS: We analysed longitudinal self-report data from a nationally representative cohort of 10,332 Australian women born 1973-1978, surveyed 6 times between 1996 (aged 18-23) and 2012 (aged 34-39). RESULTS: Group-based trajectory modelling identified four groups. Normative Mothers (46%, mean age at motherhood 30.5 years) made the transition to motherhood close to the Australian median age. Early Mothers (25%, 25.2 years) and Very Early Mothers (7%, 20.0 years) made this transition earlier; Not Mothers (22%) had not given birth. Generalised linear mixed models showed that all groups improved mean MHI-5 scores over time. Patterns of group differences were complex: Normative and Early Mothers scored consistently highest; Very Early Mothers scored lowest at most surveys; Not Mothers' scores increased relative to others over time. Most effects disappeared after adjustment for confounders. Early and Very Early Mothers showed multiple indicators of social disadvantage, while Not Mothers had very low rates of marriage. CONCLUSIONS: Timing of motherhood is embedded in sociodemographic and personal contexts. Women with socioeconomic advantages were characterised by higher mental health-related quality of life and later transition to motherhood, but adjustment for relative advantage attenuated differences in mental health-related quality of life. The overall findings suggest a pattern of positive adaptation to circumstances, with mental health-related quality of life improving through early adulthood regardless of timing of motherhood.

Longitudinal analysis of relationships between social support and general health in an Australian population cohort of young women.

PURPOSE: The influence of social support on health and quality of life has been well documented. There is less evidence on whether health status affects social support, and little is known about longitudinal relationships between social support and health in early adulthood. This study investigates these associations using both concurrent and time-lagged measures at 5 time-points over 12 years during early adulthood. METHODS: A population-based cohort of 9,758 young women from the Australian Longitudinal Study on Women's Health was used. Women were aged 22-27 in 2000 and 35-39 in 2012. The General Health subscale of the SF-36 and the MOS Social Support Survey 6-item Scale were used, with scores standardised to a range of 0-100. Longitudinal tobit models were used, because both social support and general health data were left skewed, with marked ceiling effects. All models were adjusted for status of the outcome of interest at the immediately previous survey. RESULTS: With both concurrent and time-lagged measures, there was a statistically significant difference in mean general health scores across social support quintiles after adjusting for demographic and behavioural covariates: lower general health was associated with lower social support. In reverse, social support mean scores were also significantly different across general health quintiles in both concurrent and time-lagged fully adjusted models. CONCLUSION: Social support is significantly associated with both current and subsequent general health in early adulthood. The significance of the reverse associations indicates that the two mutually influence each other. This study highlights the importance of social support as a health-related quality of life issue.

Longitudinal trajectory patterns of social support: correlates and associated mental health in an Australian national cohort of young women.

PURPOSE: Although social support is a significant contributor to health and well-being, little is known about patterns of perceived social support over time in young adulthood. It is also unclear which personal characteristics are associated with different patterns, and whether there is an association between social support and mental health over time. We explore these issues in a large national cohort of young women. METHODS: We used six waves of longitudinal data spanning 16 years, from 10,369 women from the Australian Longitudinal Study on Women's Health, initially aged 18-23. We used group-based trajectory modelling to identify patterns of social support across Surveys 2-6; multinomial logistic regression to identify socio-demographic and health-behaviour predictors at Survey 1 and correlates at Survey 6 for each trajectory group; and generalised linear mixed modelling to estimate mean levels of mental health over the trajectory period for each group, adjusted for confounders. RESULTS: Four distinct trajectory groups of social support were identified: 'High' (58.5 %), 'Decreasing' (20.6 %), 'Low' (9.3 %), and 'Increasing' (11.6 %). Poor health and living outside metropolitan areas at both Surveys 1 and 6 were characteristics of women in all trajectory groups other than the 'High' group, as were early motherhood and being un-partnered at age 34-39. Other characteristics were specific to one or two trajectory groups. Patterns of mental health over time were consistent with patterns of social support. CONCLUSION: Longitudinal trajectory patterns of social support are associated with mental health, health behaviours, and demographic factors even in early adulthood.

Validation of the MOS Social Support Survey 6-item (MOS-SSS-6) measure with two large population-based samples of Australian women.

PURPOSE: This study aimed to validate a 6-item 1-factor global measure of social support developed from the Medical Outcomes Study Social Support Survey (MOS-SSS) for use in large epidemiological studies. METHODS: Data were obtained from two large population-based samples of participants in the Australian Longitudinal Study on Women's Health. The two cohorts were aged 53-58 and 28-33 years at data collection (N = 10,616 and 8,977, respectively). Items selected for the 6-item 1-factor measure were derived from the factor structure obtained from unpublished work using an earlier wave of data from one of these cohorts. Descriptive statistics, including polychoric correlations, were used to describe the abbreviated scale. Cronbach's alpha was used to assess internal consistency and confirmatory factor analysis to assess scale validity. Concurrent validity was assessed using correlations between the new 6-item version and established 19-item version, and other concurrent variables. RESULTS: In both cohorts, the new 6-item 1-factor measure showed strong internal consistency and scale reliability. It had excellent goodness-of-fit indices, similar to those of the established 19-item measure. Both versions correlated similarly with concurrent measures. CONCLUSION: The 6-item 1-factor MOS-SSS measures global functional social support with fewer items than the established 19-item measure.

Disturbed eating behaviours and thoughts in Australian adolescents with type 1 diabetes.

AIM: To describe the presence and type of disturbed eating behaviours and thoughts in a combined male/female Australian sample of adolescents with type 1 diabetes, and examine the association of eating behaviours and thoughts with glycaemic control as evidenced by high glycosylated haemoglobin levels (HbA1c). METHODS: In this cross-sectional descriptive study, 124 adolescents aged 13-18 years were invited to complete three self-administered questionnaires. The Youth Eating Disorder Examination Questionnaire (YEDE-Q) and the Eating Disorder Inventory -3 Risk Composite (EDI-3RC) assessed risk for an eating disorder. The third questionnaire, the Strengths and Difficulties Questionnaire (SDQ) assessed emotional and behavioural concerns. Clinical data were collected from the medical records, routine clinic appointments and the adolescent. RESULTS: Any disturbed eating behaviour was reported by approximately one-third of participants (32.3%) and was common in females and males (37.9% vs. 25.9%). Binge eating (17.7%), driven exercise (13.0%) and dietary restraint (8.9%) were the most common disturbed eating behaviours, although restraint was not evident in males. Insulin manipulation/omission (5.6%), vomiting (3.3%), laxative (0.8%) or diuretic use (0.8%) were less common. Regression analysis showed a significant association between HbA1c and more disturbed eating behaviours and thoughts which remained significant when adjusted for confounders. CONCLUSIONS: High rates of disturbed eating behaviours and thoughts were seen in this Australian sample of adolescents with type 1 diabetes. High scores on both eating disorder measures were associated with poorer glycaemic control. These results highlight the need to screen for disordered eating in adolescents with type 1 diabetes.

Identifying comorbidity patterns of health conditions via cluster analysis of pairwise concordance statistics.

Identification of comorbidity patterns of health conditions is critical for evidence-based practice to improve the prevention, treatment and health care of relevant diseases. Existing approaches focus mainly on either using descriptive measures of comorbidity in terms of the prevalence of coexisting conditions, or addressing the prevalence of comorbidity based on a particular disease (e.g. psychosis) or a specific population (e.g. hospital patients). As coincidental comorbidity by chance increases with the prevalence rates of the conditions, which in turn depend heavily on the population under study, research findings on comorbidity patterns using those approaches may provide unreliable results. In this paper, we propose an asymmetric version of Somers' D statistic to provide a quantitative measure of comorbidity that accounts for co-occurrence of conditions by chance, and develop a unified clustering algorithm to identify comorbidity patterns with adjustment for multiple testing and control for the false discovery rate. We assess the applicability of the proposed comorbidity measure and investigate the performance of the proposed procedure for the adjustment of multiple testing by conducting a comparative study and a sensitivity analysis, respectively. The proposed method is illustrated using a national survey data set of mental health and wellbeing and a national health survey data set in Australia.

Concurrent validity of self-report measures of eating disorders in adolescents with type 1 diabetes.

AIM: Eating disorder screening tools have not been adequately validated for use with adolescents with type 1 diabetes. This study compared the Youth Eating Disorder Examination-Questionnaire (YEDE-Q) and the Eating Disorder Inventory-3 Risk Composite (EDI-3RC) against the child Eating Disorder Examination (chEDE). These screening tools were chosen because they broadly assess eating disorder psychopathology and have subscales helpful for clinical management. METHODS: In this cross-sectional study, 124 adolescents with type 1 diabetes aged 13-18 years completed two self-administered questionnaires, the YEDE-Q and the EDI-3RC. Cronbach's alpha was used to assess internal consistency of the tools. Fifty-one adolescents, randomly selected, participated in the chEDE. Intraclass correlations and Spearman's correlations were used to measure concordance of the chEDE with the YEDE-Q and EDI-3RC. RESULTS: The YEDE-Q and EDI-3RC demonstrated good subscale internal consistency; Cronbach's alpha for YEDE-Q (0.78-0.95) and EDI-3RC (0.79-0.94). High levels of concurrent validity with the chEDE were seen with both tools. CONCLUSION: Preliminary evidence is provided for the validation of the YEDE-Q and EDI-3RC for use in adolescents with type 1 diabetes. The YEDE-Q also defines individual disturbed eating behaviours with frequency ratings which can be helpful for tailoring early intervention.

A thematic analysis of the role of the organisation in building allied health research capacity: a senior managers' perspective.

BACKGROUND: Evidence-based practice aims to achieve better health outcomes in the community. It relies on high quality research to inform policy and practice; however research in primary health care continues to lag behind that of other medical professions. The literature suggests that research capacity building (RCB) functions across four levels; individual, team, organisation and external environment. Many RCB interventions are aimed at an individual or team level, yet evidence indicates that many barriers to RCB occur at an organisational or external environment level. This study asks senior managers from a large healthcare organisation to identify the barriers and enablers to RCB. The paper then describes strategies for building allied health (AH) research capacity at an organisational level from a senior managers' perspective. METHODS: This qualitative study is part of a larger collaborative RCB project. Semi-structured in-depth interviews were conducted with nine allied health senior managers. Recorded interviews were transcribed and NVivo was used to analyse findings and emergent themes were defined. RESULTS: The dominant themes indicate that the organisation plays an integral role in building AH research capacity and is the critical link in creating synergy across the four levels of RCB. The organisation can achieve this by incorporating research into its core business with a whole of organisation approach including its mission, vision and strategic planning. Critical success factors include: developing a co-ordinated and multidisciplinary approach to attain critical mass of research-active AH and enhance learning and development; support from senior managers demonstrated through structures, processes and systems designed to facilitate research; forming partnerships to increase collaboration and sharing of resources and knowledge; and establishing in internal framework to promote recognition for research and career path opportunities. CONCLUSIONS: This study identifies four key themes: whole of organisation approach; structures, processes and systems; partnerships and collaboration; and dedicated research centres, units and positions. These themes form the foundation of a model which can be applied to assist in achieving synergy across the four levels of RCB, overcome barriers and create an environment that supports and facilitates research development in AH.

Evaluating a team-based approach to research capacity building using a matched-pairs study design.

BACKGROUND: There is a continuing need for research capacity building initiatives for primary health care professionals. Historically strategies have focused on interventions aimed at individuals but more recently theoretical frameworks have proposed team-based approaches. Few studies have evaluated these new approaches. This study aims to evaluate a team-based approach to research capacity building (RCB) in primary health using a validated quantitative measure of research capacity in individual, team and organisation domains. METHODS: A non-randomised matched-pairs trial design was used to evaluate the impact of a multi-strategy research capacity building intervention. Four intervention teams recruited from one health service district were compared with four control teams from outside the district, matched on service role and approximate size. All were multi-disciplinary allied health teams with a primary health care role. Random-effects mixed models, adjusting for the potential clustering effect of teams, were used to determine the significance of changes in mean scores from pre- to post-intervention. Comparisons of intervention versus control groups were made for each of the three domains: individual, team and organisation. The Individual Domain measures the research skills of the individual, whereas Team and Organisation Domains measure the team/organisation's capacity to support and foster research, including research culture. RESULTS: In all three domains (individual, team and organisation) there were no occasions where improvements were significantly greater for the control group (comprising the four control teams, n = 32) compared to the intervention group (comprising the four intervention teams, n = 37) either in total domain score or domain item scores. However, the intervention group had a significantly greater improvement in adjusted scores for the Individual Domain total score and for six of the fifteen Individual Domain items, and to a lesser extent with Team and Organisation Domains (two items in the Team and one in the Organisation domains). CONCLUSIONS: A team-based approach to RCB resulted in considerable improvements in research skills held by individuals for the intervention group compared to controls; and some improvements in the team and organisation's capacity to support research. More strategies targeted at team and organisation research-related policies and procedures may have resulted in increased improvements in these domains.

Comprehensive health assessments for adults with intellectual disability living in the community - weighing up the costs and benefits.

BACKGROUND: Health assessments have beneficial effects on health outcomes for people with intellectual disability living in the community. However, the effect on medical costs is unknown. METHODS: We utilised Medicare Australia data on consultations, procedures and prescription drugs (including vaccinations) from all participants in a randomised control trial during 2002-03 that examined the effectiveness of a health assessment. Government health costs for adults with intellectual disability who did or did not receive an assessment were compared. Bootstrapping statistics (95% confidence interval) were employed to handle the right-skewed cost data. RESULTS: Over 12 months, patients receiving health assessments incurred total costs of $4523 (95% CI: $3521 to $5525) similar to those in usual care $4466 (95% CI: $3283 to $5649). Costs were not significantly higher compared with the 12 month pre-intervention period. DISCUSSION: Health assessments for adults with intellectual disability living in the community are encouraged as they produce enhanced patient care but do not increase overall consultation or medication costs.

Uptake of Medicare chronic disease management incentives - a study into service providers' perspectives.

BACKGROUND: Chronic disease is responsible for 80% of the burden of disease in Australia. The Australian Government Medicare Benefits Schedule (MBS) provides incentives through specific Medicare items to optimise chronic disease management (CDM), yet little is known about factors that influence their uptake. METHODS: Exploratory qualitative research was used, which incorporated focus groups and interviews with 26 staff from nine general practices in southeast Queensland, together with review of practice-specific data on CDM income. Content analysis of qualitative data was undertaken to identify barriers, enablers and service models associated with MBS CDM item uptake. Triangulation of methods and data sources facilitated confirmation of findings. RESULTS: Time pressures and unreliable MBS information were common barriers to uptake for general practitioners. Employing a nurse, team-based approaches, recall systems and using only selected MBS CDM item numbers were associated with best uptake. CONCLUSION: Improved systems within general practice and Medicare may increase the uptake of MBS CDM item numbers.

Validation of the research capacity and culture (RCC) tool: measuring RCC at individual, team and organisation levels.

Research capacity building (RCB) in Australia has recently focussed on strategies that take a whole of system approach to developing research culture at individual, team and organisation levels. Although a theoretical framework exists, no tool has been published that quantitatively measures the effectiveness of RCB interventions aimed at these three levels. A sample of 134 allied health workers was used to validate the research capacity and culture (RCC) tool. Item level analysis was undertaken using Cronbach's α and exploratory factor analysis, and test-retest reliability was examined using intra-class correlations (ICC). The tool had one factor emerge for each domain, with excellent internal consistency for organisation, team and individual domains (α=0.95, 0.96 and 0.96 respectively; and factor loadings ranges of 0.58-0.89, 0.65-0.89 and 0.59-0.93 respectively). The overall mean score (total) for each domain was: 5.4 (inter-quartile range 3.9-7.7), 4.4 (IQR 2.6-6.1) and 3.9 (IQR 2.9-6) for the organisation, team and individual domains respectively. Test-retest reliability was strong for each domain: organisation ICC=0.77, team ICC=0.83 and individual ICC=0.82. The RCC tool has three domains measuring research capacity and culture at organisation, team and individual levels. It demonstrates excellent internal consistency and strong test-retest reliability.

Distribution of volumes of individual glomeruli in kidneys at autopsy: association with physical and clinical characteristics and with ethnic group.

We have demonstrated considerable variability in the volumes of different glomeruli in given individuals (individual glomerular volume: IGV) in a stereologic study of kidneys at forensic autopsy performed to investigate sudden or unexpected death in people without manifest kidney disease. We review some important associations of IGV by subject characteristics and by ethnic groups. IGVs were measured by the Cavalieri method in 30 glomeruli in each of 111 adult males who belonged to 4 ethnic groups, i.e. US Whites, African-Americans, Africans from Senegal, and Australian Aborigines. Correlations of pooled IGV values with certain subject characteristics were evaluated in the US Whites. Pooled IGV data were compared in subjects across the 4 ethnic groups. In US Whites, mean IGV and its variance were greater with higher age, lower nephron number, lower birth weight, and with gross obesity, hypertension and cardiovascular death. In comparisons by ethnic group, mean IGV and IGV ranges were higher in African-Americans and Australian Aborigines than in US Whites and African Senegalese subjects. We conclude that glomerular enlargement with volume heterogeneity marks more advanced age, relative nephron deficiency, lower birth weight, obesity, hypertension, and advanced cardiovascular disease. The findings in African-Americans and Australian Aborigines suggest that larger IGVs and volume heterogeneity might mark populations with accentuated susceptibility to hypertension and kidney disease, but the data need to be further examined in the context of the determining characteristics defined in the US Whites.

Patterns of multimorbidity in working Australians.

BACKGROUND: Multimorbidity is becoming more prevalent. Previously-used methods of assessing multimorbidity relied on counting the number of health conditions, often in relation to an index condition (comorbidity), or grouping conditions based on body or organ systems. Recent refinements in statistical approaches have resulted in improved methods to capture patterns of multimorbidity, allowing for the identification of nonrandomly occurring clusters of multimorbid health conditions. This paper aims to identify nonrandom clusters of multimorbidity. METHODS: The Australian Work Outcomes Research Cost-benefit (WORC) study cross-sectional screening dataset (approximately 78,000 working Australians) was used to explore patterns of multimorbidity. Exploratory factor analysis was used to identify nonrandomly occurring clusters of multimorbid health conditions. RESULTS: Six clinically-meaningful groups of multimorbid health conditions were identified. These were: factor 1: arthritis, osteoporosis, other chronic pain, bladder problems, and irritable bowel; factor 2: asthma, chronic obstructive pulmonary disease, and allergies; factor 3: back/neck pain, migraine, other chronic pain, and arthritis; factor 4: high blood pressure, high cholesterol, obesity, diabetes, and fatigue; factor 5: cardiovascular disease, diabetes, fatigue, high blood pressure, high cholesterol, and arthritis; and factor 6: irritable bowel, ulcer, heartburn, and other chronic pain. These clusters do not fall neatly into organ or body systems, and some conditions appear in more than one cluster. CONCLUSIONS: Considerably more research is needed with large population-based datasets and a comprehensive set of reliable health diagnoses to better understand the complex nature and composition of multimorbid health conditions.

Health-related productivity losses increase when the health condition is co-morbid with psychological distress: findings from a large cross-sectional sample of working Australians.

BACKGROUND: The health condition of workers is known to impact on productivity outcomes. The relationship between health and productivity is of increasing interest amid the need to increase productivity to meet global financial challenges. Prevalence of psychological distress is also of growing concern in Australia with a two-fold increase in the prevalence of psychological distress in Australia from 1997-2005. METHODS: We used the cross-sectional data set from the Australian Work Outcomes Research Cost-benefit (WORC) study to explore the impacts of health conditions with and without co-morbid psychological distress, compared to those with neither condition, in a sample of approximately 78,000 working Australians. The World Health Organisation Health and Performance Questionnaire was used which provided data on demographic characteristics, health condition and working conditions. Data were analysed using negative binomial logistic regression and multinomial logistic regression models for absenteeism and presenteeism respectively. RESULTS: For both absenteeism and presenteeism productivity measures there was a greater risk of productivity loss associated when health conditions were co-morbid with psychological distress. For some conditions this risk was much greater for those with co-morbid psychological distress compared to those without. CONCLUSIONS: Co-morbid psychological distress demonstrates an increased risk of productivity loss for a range of health conditions. These findings highlight the need for further research to determine whether co-morbid psychological distress potentially exacerbates lost productivity.

Skin disease in the first two years of life in Aboriginal children in East Arnhem Land.

BACKGROUND: The most common skin infections affecting children in remote Aboriginal communities are scabies and impetigo. Group A streptococcal skin infections are linked to the high rates of heart and renal disease occurring in Aboriginal Australians. METHODS: A retrospective review of medical records was conducted in a primary health care centre in the East Arnhem region of the Northern Territory. Data was collected from all presentations to the clinic in the first 2 years of life for 99 children born between 2001 and 2005 as a component of the East Arnhem Regional Healthy Skin Project. RESULTS: The median number of presentations to the clinic in the first 2 years of life was 32. Skin disease was recorded in 22% of all presentations. By 1 year of age 82% of children had presented to the clinic with their first episode of impetigo and 68% with their first episode of scabies. Antibiotics were administered to 49% of children with impetigo. CONCLUSION: Skin infections are a major reason for presentation to primary health clinics and contribute to the high disease burden experienced by children in the first 2 years of life. This high frequency of presentation provides multiple opportunities for intervention and monitoring.

Nephron number and individual glomerular volumes in male Caucasian and African American subjects.

BACKGROUND: Glomerular hypertrophy has been described in several populations at high risk of chronic kidney disease. Total nephron (and thereby glomerular) number (N(glom)) varies widely in normal adult human kidneys and is generally inversely correlated with mean glomerular volume (V(glom)). However, little is known about the range of individual glomerular volumes (IV(glom)) within single human kidneys and the association with N(glom). The aim of the present study was to estimate IV(glom) in Caucasian and African Americans and identify any associations between heterogeneity in IV(glom) and nephron number. METHODS: Using unbiased stereological techniques, IV(glom) was determined for 30 glomeruli in each of 24 adult male kidneys from Jackson, MS, USA (12 Caucasian and 12 African American). Half of each group had 'high' N(glom) (>1.2 million nephrons per kidney) and the other half had 'low' N(glom) (<600 000). RESULTS: Caucasians with high N(glom) had a relatively homogeneous distribution of IV(glom) as well as a relatively low mean value, while those with low N(glom) had much greater heterogeneity of IV(glom), as well as a larger IV(glom) (P < 0.0001) compared with those with high N(glom). This disparity was not apparent in African Americans, however, where subjects with both high and low N(glom) showed substantial heterogeneity in IV(glom) and larger mean values (P = 0.95). CONCLUSIONS: High N(glom) appeared to protect against glomerular enlargement and volume heterogeneity in Caucasians. However, substantial variation in IV(glom) and net enlargement in glomerular size in African Americans with high nephron numbers suggest that additional forces, independent of low N(glom), are driving glomerular enlargement and heterogeneity.

Predictors of change in depressive symptoms over time: Results from the Australian Longitudinal Study on Women's Health.

BACKGROUND: Depressive symptoms fluctuate over time, and are most common amongst women in early adulthood. Understanding predictors of changes in depressive symptoms among young women may inform health promotion and early intervention. METHODS: Data were collected at three-yearly intervals from 2000 (Survey 2) to 2012 (Survey 6) from the Australian Longitudinal Study on Women's Health. The sample comprised 7663 women, aged 22-27 in 2000, who reported any indicator of poor mental health at any wave. Generalised linear mixed models identified predictors of change in depressive symptoms (CESD-10) over each three-year period. RESULTS: Depressive symptoms reduced over time. In a fully adjusted model, greater reduction in symptoms was predicted by higher initial symptoms, time, higher social support, and higher self-rated general health. Slower reduction was predicted by lower education, difficulty managing on income, high or zero alcohol consumption, stress, and history of childhood sexual assault or partner violence. Motherhood predicted an increase in depressive symptoms at Survey 2 (2000), but a decrease at Survey 5 (2009). LIMITATIONS: Although sampling was nationally representative, there is a slight bias towards Australian-born and more educated women. Further, although validated measures are used, all data are self-report. CONCLUSIONS: Fluctuations in depressive symptoms among young women are related to fixed and time-varying factors spanning multiple health and social domains. A range of factors, including education and financial resources, promotion of positive social support systems, and encouragement of health promoting lifestyles, might serve to promote young women's mental health and thus to reduce pressure on clinical services.

Characteristics of nonparticipants differed based on reason for nonparticipation: a study involving the chronically ill.

BACKGROUND: Individuals who do and do not participate in studies often have different characteristics. The profile of nonparticipants may vary by type of nonparticipation. Few cross-sectional or longitudinal studies of chronically ill individuals identify these differences at baseline and of those that do; many do not have information on type of nonparticipation. We compared characteristics of chronically ill individuals to determine if they differ by type of nonparticipation. METHODS: Data are from an evaluation of a chronic disease management program conducted in New South Wales, Australia during 2001 and 2002. Characteristics were compared using routinely collected hospital data. Reasons for nonparticipation were categorized as refusal, death, or untraceable. RESULTS: Individuals who refused to participate were older, female, discharged at risk, and have heart failure. Those untraceable were younger, not married, Indigenous, and receiving care during the intervention phase but not recruited. Individuals not participating due to death were older, male, not cohabiting, discharged at risk, and have a diagnosis of cancer or heart failure. CONCLUSION: The significant differences between untraceable, refused, and deceased individuals should be considered when designing studies and when adjusting for nonparticipation bias.

Trajectories of mental health over 16 years amongst young adult women: The Australian Longitudinal Study on Women's Health.

This article used data from 5,171 young women participating in the Australian Longitudinal Study on Women's Health, a nationally representative longitudinal cohort study, to identify longitudinal trajectory patterns of mental health across 6 surveys over 16 years of early adulthood, from age 18-23 to age 34-39. In addition, we identified both predictors and outcomes of these trajectories. Using group-based trajectory modeling, we identified 4 distinct trajectory groups of mental health. The mental health of most participants (55%) was consistently high, with 12% improving, 24% varying, and 9% frequently low. The authors considered characteristics at the beginning and end of the trajectory period, taking a life-course perspective to understand vulnerabilities to, and outcomes of, low or variable poor mental health trajectories. Financial difficulties, poor general health, and weight or shape dissatisfaction were characteristics at Survey 1 that distinguished all other trajectory groups from those with consistently high mental health. Other differences were specific to 1 or 2 groups. By the end of the trajectory period, the improving mental health group showed few differences from those with consistently high mental health. However, those with varying and low mental health showed evidence of social disadvantage, poor physical and emotional health, and unhealthy behaviors, and were less likely to be mothers. The ability to identify distinct trajectories of mental health in early adulthood, and their correlates, provides evidence to underpin population health interventions targeting the prevention of mental health problems among this population group. (PsycINFO Database Record