Quantifying Physical Resilience after Knee or Hip Surgery in Older Australian Women Based on Long-Term Physical Functioning Trajectories.

INTRODUCTION: To improve outcomes after knee or hip surgery, better insight is needed in long-term recovery patterns in the context of ageing-related decline. We examined long-term trajectories of physical functioning (PF) in older women with and without hip and knee surgery and described profiles of cases with higher and lower resilience after surgery. METHODS: This observational study used data from 10,434 women (73-79 years) who completed survey 2 of the Australian Longitudinal Study on Women's Health. Data were used from surveys 2 (1999) to 6 (2011). Covariable-adjusted linear mixed models were run to examine the surgery-by-time (-12 to +12 years) interaction in association with PF (SF-36 subscale). The differences between observed and expected PF were calculated, with positive/negative values reflecting higher/lower resilience, respectively. RESULTS: Women with hip surgery (n = 982) had lower PF than those without surgery (n = 8,117) (p < 0.001). Among hip surgery patients, the decline was more rapid pre-surgery than post-surgery (Δslope = -0.7, p < 0.001). Women with knee surgery (n = 1,144) had lower PF than those without surgery (n = 7,971), but with a slower rate of decline (p = 0.01). Among knee surgery patients, the rate of decline was similar pre- and post-surgery (Δslope = -0.3, p = 0.25). Both in hip and knee patients, women with higher resilience had fewer comorbidities and symptoms and were more often physically active and independent in daily activities than those with lower resilience (all p < 0.05). CONCLUSION: Compared with women without surgery, PF was lower and declined more rapidly around the time of hip surgery, but not for knee surgery. Women with better long-term recovery after surgery had fewer health problems and were more independent around the time of surgery.

A novel method to monitor rheumatoid arthritis prevalence using hospital and medication databases.

BACKGROUND: Most estimates of rheumatoid arthritis (RA) prevalence, including all official figures in Australia and many other countries, are based on self-report. Self-report has been shown to overestimate RA, but the 'gold standard' of reviewing individual medical records is costly, time-consuming and impractical for large-scale research and population monitoring. This study provides an algorithm to estimate RA cases using administrative data that can be adjusted for use in multiple contexts to provide the first approximate RA cohort in Australia that does not rely on self-report. METHODS: Survey data on self-reported RA and medications from 25 467 respondents of the Australian Longitudinal Study on Women's Health (ALSWH) were linked with data from the national medication reimbursement database, hospital and emergency department (ED) episodes, and Medicare Benefits codes. RA prevalence was calculated for self-reported RA, self-reported RA medications, dispensed RA medications, and hospital/ED RA presentations. Linked data were used to exclude individuals with confounding autoimmune conditions. RESULTS: Of 25 467 survey respondents, 1367 (5·4%) women self-reported disease. Of the 26 840 women with hospital or ED presentations, 292 (1·1%) received ICD-10 codes for RA. There were 1038 (2·8%) cases by the medication database definition, and 294 cases (1·5%) by the self-reported medication definition. After excluding individuals with other rheumatic conditions, prevalence was 3·9% for self-reported RA, 1·9% based on the medication database definition and 0·5% by self-reported medication definition. This confirms the overestimation of RA based on self-reporting. CONCLUSIONS: We provide an algorithm for identifying individuals with RA, which could be used for population studies and monitoring RA in Australia and, with adjustments, internationally. Its balance of accuracy and practicality will be useful for health service planning using relatively easily accessible input data.

Associations between factors in childhood and young adulthood and childlessness among women in their 40s: A national prospective cohort study.

BACKGROUND: While the risk factors for infertility are well-established, research on factors associated with voluntary childlessness is limited and mainly focused on adulthood factors. Thus, we examined the associations between factors in childhood and young adulthood and different types of childlessness. METHODS: The analysis included 4653 women from the Australian Longitudinal Study on Women's Health from 1996 to 2021. Childlessness was categorised as: voluntary, due to infertility issues, or due to other reasons. The associations between factors in childhood and young adulthood and childlessness were assessed using multinomial logistic regression models. RESULTS: In their 40s, 4.8 % of women were voluntarily childless, 6.7 % were childless due to infertility issues, and 7.8 % were childless due to other reasons. Regardless of types of childlessness, being childless was associated with poorer self-rated health during childhood and having been unpartnered and obese in young adulthood. Ex-smokers in young adulthood had lower odds of childlessness. Childhood physical abuse was associated with childlessness due to infertility issues and other reasons. Voluntary childlessness and childlessness due to infertility issues were associated with having identified as non-exclusively heterosexual in early adulthood. Lower social support in early adulthood was associated with voluntary childlessness and childlessness due to other reasons. LIMITATIONS: The direction of the associations could not be determined and using self-reported data may introduce recall bias. CONCLUSIONS: Factors in childhood and young adulthood were associated with different types of childlessness, highlighting the importance of adopting a life course perspective when studying childlessness.

Adherence to 24-h movement behaviour guidelines in families with multiple children.

BACKGROUND: In 2019, the World Health Organization (WHO) launched the first global movement guidelines for children that combined sleep, physical activity and screen time. Our previous research showed that adherence to age-specific guidelines for screen time was challenging for families with children in different age groups. We aimed to determine whether families with children in different age-based movement guideline categories have poorer adherence to the broader 24-h movement guidelines than those with all children in the same age category. METHODS: Data were from the 1973-1978 cohort of the Australian Longitudinal Study on Women's Health (seventh survey, 2015) and the women's three youngest children (aged ≤12) (Mothers and their Children's Health sub-study, 2016/2017). The sample was 1787 women (families) with 4064 children (mean age 7.2 [SD 2.9]). Whether children in the family were in the same or different age-based category was determined by matching children's ages in a family against age-based guideline categories for the 24-h movement behaviours. The association between children in the family being in the same or different age-based guideline category on adherence to 24-h movement guidelines, both collectively and individually, was analysed by adjusted logistic regression (binary and multinomial). RESULTS: Families with children in the same age guideline categories had double the odds of having all children meet 24-h movement guidelines (adjusted odds ratio [OR] 1.95 [95% confidence interval, CI: 1.32, 2.86]). Families with children in the same age categories on the screen guideline had higher odds of all children meeting (2.25 [1.73, 2.93]) and lower odds of some meeting/some failing the screen guideline (0.18 [0.14, 0.25]), than families with all children in different age categories. Families with children in the same age categories on the physical activity guideline had lower odds of all children meeting (0.57 [0.43, 0.75]) or some meeting/some failing the physical activity guideline (0.08 [0.06, 0.12]). No associations were found for sleep guidelines. CONCLUSIONS: Families with multiple children may need practical advice and strategies on how to adhere to guidelines when children span age categories. This could form part of public health strategies that raise awareness of the guidelines and may improve guideline adherence.

Is being childless associated with a woman's risk of overweight and obesity? Results from a national longitudinal study.

AIM: To examine the association between motherhood status (mothers, voluntarily childless, involuntarily childless) and overweight and obesity over 22 years. METHODS: A total of 4092 women aged 18-23 years were followed from 1996 to 2018. Motherhood status was defined by women's reports on their fertility, attempts to conceive, use of in vitro fertilisation and fertility hormones, and number of biological children. Associations between motherhood status and overweight and obesity were examined using generalised estimating equations models, adjusting for socio-demographic characteristics, lifestyle factors, depressive symptoms, early life factors, and polycystic ovary syndrome (PCOS). RESULTS: At age 40-45 years, 12% of women were voluntarily childless and 5% were involuntarily childless. The prevalence of overweight and obesity increased with age and women who were voluntarily or involuntarily childless had higher prevalence of obesity than mothers in all surveys. After adjusting for covariates, compared with mothers, women who were voluntarily childless had higher odds of being overweight (odds ratio [OR], 95% confidence interval [CI]: 1.29, 1.09-1.52) and obese (OR, 95% CI: 1.65, 1.29-2.12). Involuntary childlessness was not associated with being overweight (OR, 95% CI: 1.05, 0.82-1.33), and its association with obesity was attenuated after adjusting for PCOS in the final model (OR, 95% CI: 1.40, 0.99-1.98). CONCLUSIONS: Around one in nine Australian women remained voluntarily childless by their late reproductive years. On average, they had higher odds of being overweight and obese than mothers, suggesting that overweight and obesity prevention programs should consider tailoring their advice by motherhood status.

Managing Screen Use in the Under-Fives: Recommendations for Parenting Intervention Development.

The impact of excessive screen use on children's health and development is a public health concern and many countries have published recommendations to limit and guide the use of screen media in childhood. Despite this, international studies report that the majority of parents and children do not adhere to screen use recommendations. Existing research aiming to understand children' screen use has largely focused on older children, and on demographic and structural aspects of the child's environment. Parents play a central role in determining young children's screen use and identify numerous barriers to developing healthy screen use practices with their children. However, no clear models exist that incorporate key parenting factors in understanding children's screen use, which presents an impediment to intervention development. Likewise, while some evidence exists for interventions to improve children's screen use behaviours, most are focused on older children and parental involvement has generally been limited. In this paper, we overview key factors associated with screen use in young children (< 5 years) and summarise the existing evidence base for interventions designed to support healthy screen use. This paper proposes a conceptual model linking aspects of parenting and the socio-ecological environment to young children's screen use. Our proposed model could be used to design longitudinal studies of screen use predictors and outcomes, and inform intervention development. Finally, the paper provides key recommendations for future research, intervention development and testing.

Do mothers or females without children have better health-related quality of life across their reproductive years?

PURPOSE: To compare health-related quality of life (HRQoL) between mothers and females who were voluntarily or involuntarily childless, across their reproductive years. METHODS: 4100 females born in 1973-78 from the Australian Longitudinal Study on Women's Health were followed for 22 years. Motherhood status was defined by females' reports in Survey 8 (2018, 40-45 years) on their fertility, attempts to conceive, use of in vitro fertilization and fertility hormones, and number of children. HRQoL was assessed in each survey using the 36-Item Short Form Survey (SF-36). Linear mixed models were used to assess the associations between motherhood status and HRQoL. RESULTS: Over 22 years, compared with females who were voluntarily childless, mothers on average had better HRQoL (shown by scores 1.5 to 3.4 points higher on five of the eight SF-36 subscales), while females who were involuntarily childless scored 2.2 to 3.0 points lower on three of eight SF-36 subscales. Compared with females who were voluntarily childless, teen mothers (age at first birth < 20 years) scored lower on role limitations due to physical problems subscale (b = -5.5, 95% CI: -9.3, -1.6), while females with either two, or three or more children scored 1.6 to 4.8 points higher on seven of eight SF-36 subscales. CONCLUSIONS: Females who were childless had poorer HRQoL than mothers. Further research is needed to understand the underlying mechanisms, which could inform policymakers on how to reduce the health disparities and improve long-term health outcomes for females.

History of pre-pregnancy maternal symptoms of binge eating and childhood behavioral problems in girls and boys.

OBJECTIVE: To explore whether children of mothers with pre-pregnancy binge eating (BE) symptoms have more behavioral difficulties compared with those without and whether associations are moderated by ED symptoms and other maternal health and social factors measured during childhood. METHOD: Pre-pregnancy BE symptoms were collected by the Australian Longitudinal Study on Women's Health at Survey 1 (in 1996) and/or at Survey 2 (in 2000) using questions mapped to DSM BE criterion 1. In 2016/7, 2180 women from the 1973-78 cohort provided data on externalizing and internalizing behavior, measured by Strengths and Difficulties Questionnaire, on 4054 of their children (2-12 years) in the Mothers and their Children's Health study. Covariates were markers of other ED symptoms, sociodemographic, social support, and mental health factors collected proximally to the child outcomes. Hierarchical multivariable regression models, using generalized estimating equations accounting for clustering of children within mothers, were used. RESULTS: Pre-pregnancy BE symptoms were associated with child behavior, with associations only moderated after adjustment for proximal markers of ED (girls internalizing behavior, b (95%CI) .30 (-.02, .61); boys externalizing behavior .34 (-.04, .73)) or social support (girls externalizing behavior 0.26 (-.08, .61)). Pre-pregnancy BE symptoms were not associated with boys internalizing behavior (-.27 (-.02, 0.57)). DISCUSSION: Studies with repeated ED measures should test hypotheses that these associations vary by timing of ED measurement. Identification of young women at risk of BE symptoms pre-pregnancy, as well as when children are older, may enable health services, treatment programs, and supports to minimize longer term effects on children. PUBLIC SIGNIFICANCE STATEMENT: A history of binge eating symptoms up to 10 years pre-pregnancy in mothers is associated with behavior problems in their girls and boys at average age of 7. However, the association is moderated by behaviors of eating disorders and social support in the mothers during childhood. Identification of ED symptoms prior to pregnancy, and then after childbirth, might enable health services to intervene to maximize child and mother outcomes.

Maternal factors associated with interbirth intervals in Australia: Results from a population-based longitudinal study.

BACKGROUND AND OBJECTIVE: Short and long intervals between successive births are associated with adverse birth outcomes, especially in low-income and middle-income countries, yet the birth intervals in high-income countries remain relatively understudied. The aim was to examine maternal factors associated with birth intervals in Australia. METHODS: The sample comprised 6130 participants in the Australian Longitudinal Study on Women's Health who were born in 1973-1978, had two or more births, and responded to regular surveys between 1996 and 2018. Interbirth interval (IBI) was defined as the time between successive live births. Maternal factors were examined using accelerated failure time models. RESULTS: For women with only two births (n = 3802), the median time to the second birth was 34.0 months (IQR 23.1, 46.2) with shorter IBI associated with higher socioeconomic status (eg, university education (31.9 months), less income stress (31.1)), and longer IBI associated with age over 35 (39.7), fair/poor health (43.0), untreated fertility problems (45.5), miscarriage (39.4), or abortion (41.0). For women with three or more births (n = 2328), the median times to the second and third births were 31.2 months (19.9, 42.1) and 36.5 months (25.3, 50.1), respectively; some factors were consistent between the first IBI and second IBI (eg, university education and being married were associated with shorter IBI), whereas income stress was associated with longer first IBI but not with second IBI. CONCLUSIONS: Understanding maternal factors associated with birth intervals in a high-income country like Australia may enable more nuanced tailoring of guidelines for prepregnancy care.

Generational differences in the prevalence of postpartum depression among young Australians: a comparison of two cohorts born 17 years apart.

Whether there has been an increase in postpartum depression (PPD) over the generation remains unknown. This study aimed to compare the prevalence in two cohorts of young Australian women born 17 years apart and identified the factors associated with any generational differences. Participants were from the Australian Longitudinal Study on Women's Health, who gave birth between ages 18 and 27 (born in 1973-78 and 1989-95). PPD prevalence was calculated as the percentage of births associated with PPD. Both the prevalence of PPD diagnoses (among 1,610 births) and PPD symptoms (among 953 births) were compared. Relative risks (RRs) and 95% confidence intervals (CIs) were used to report generational differences in the prevalence for PPD diagnoses; Hazard ratios (HRs) and 95% CIs used for PPD symptoms. Factors that differed between cohorts and were associated with PPD diagnoses or PPD symptoms were adjusted. The prevalence of both PPD diagnoses (21.4% vs 10.3%; crude RR: 2.03, 95% CI: 1.59-2.60) and symptoms (20.1% vs 13.3%; crude HR: 1.60, 95% CI: 1.15-2.34) were higher in the 1989-1995 cohort than the 1973-1978 cohort. Generational differences in PPD diagnoses persisted after controlling for potential contributors (RR: 1.53, 95% CI: 1.15-2.04), while generational differences in PPD symptoms were attenuated (HR: 0.98, 95% CI: 0.64-1.49). Of all contributing factors, a history of depression explained most of the generational differences, especially in PPD symptoms (49%), to the extent that when the study sample was stratified by history of depression, no generational differences were detected (without prior depression, HR: 0.65, 95% CI: 0.20-2.08; with prior depression, HR: 1.18, 95% CI: 0.71-1.96). The higher prevalence of PPD in the recent generation was mainly due to the high prevalence of depression. Strategies that well manage pre-existing depression may benefit the prevention of PPD for the current young generation. Further research is warranted to inform detailed prevention approaches.

Rheumatoid arthritis is a preventable disease: 11 ways to reduce your patients' risk.

New evidence shows that up to 40% of rheumatoid arthritis (RA) cases are attributable to exposure to potentially modifiable factors. We can now identify people at higher risk of RA (pre-RA) through their family history, risk factors, autoantibodies and symptoms. Counselling these patients to act to modify factors known to be associated with RA risk could prevent the development of RA, and evidence shows that informing individuals of their risk and of ways to reduce it leads to positive behavioural change and is not harmful. This consumer-focussed narrative review is targeted at primary care providers and physicians to describe 11 changes that can be made, based on current evidence linking potentially modifiable factors to RA risk. These evidence-based recommendations are: (i) cease smoking; (ii) reduce exposure to inhaled silica, dusts and occupational risks; (iii) maintain a healthy weight; (iv) increase leisure time physical activity; (v) maintain good dental hygiene; (vi) maximise breastfeeding if able; (vii) maximise dietary quality and avoid high-salt diets; (viii) consume high levels of omega-3 fatty acids and fish; (ix) reduce consumption of sugar-sweetened soft drinks; (x) consume moderate levels of alcohol; and (xi) remain vitamin D replete.

The experience of delayed health care access during the COVID 19 pandemic in Australian women: A mixed methods exploration.

Delayed health care access is a potential collateral effect of pandemic conditions, health rationing strategies and social distancing responses. We investigated experiences of delayed health care access in Australian women during COVID-19. A mixed methods study used quantitative and free-text data from the Australian Longitudinal Study on Women's Health COVID-19 survey 4 (health care access or delay). Logistic regression models were used to estimate factors associated with delaying access to general practitioners (GPs), specialists and allied health services. Free-text comments were analysed thematically, employing a process of constant comparison. COVID-19 survey 4 was completed by 8,200 women and 2,727 provided free-text comments. Of the women who needed the health service, 25% (1,268/5,071) delayed seeing their GP, 23.6% (570/1,695) delayed seeing a specialist and 45% (791/1,757) delayed use of an allied health service. Younger age was most significantly associated with delaying attendance. Women born 1989-95 were significantly more likely to delay compared to women born 1946-51 (OR (95% CI): GP = 0.28 (0.22, 0.35)); Specialist = 0.65 (0.45, 0.92; Allied Health = 0.59 (0.42, 0.82)). Women born 1973-78 were also likely to delay GP visits (0.69, (0.58, 0.83)). Four qualitative themes emerged including: (1) Challenges negotiating care during a pandemic; (2) Ongoing uncertainty towards accessing health care when a specialist delays an appointment; (3) Accessing health care (or not) using Telehealth and (4) Managing complex care needs. COVID-19 has had a significant effect on access to health care. Women delayed seeking help for cancer screening, mental health, and other health conditions involving chronic and complex needs for health and social care. While there is a need to rationalise and optimise health access during a pandemic, our outcomes suggest a need for public health campaigns that clarify how to access care, engage with telehealth and respond to missed appointments.

Psychological distress in young Australian women by area of residence: findings from the Australian Longitudinal Study on Women's Health.

BACKGROUND: The association between psychological distress and area of residence has been extensively discussed and debated. However, the focus has been largely on men and area of residence is often dichotomised to urban and rural. The aim of this study is to examine the association between psychological distress and area of residence in young Australian women using a broader geographical measure. METHODS: Data were from 8961 women aged 19-26 in the 1989-95 cohort of the Australian Longitudinal Study on Women's Health in 2015. The association between area of residence (measured by the Modified Monash Model) and psychological distress (measured by the K10), adjusted for demographic factors, indicators of socioeconomic position and health behaviour characteristics, was analysed using logistic regression. RESULTS: Women in regional centres had lower odds of high to very high psychological distress compared with women in metropolitan areas (adjusted odds ratio 0.73 (95% confidence interval 0.60-0.89)). Women living in large rural towns, medium rural towns and small rural towns/remote/very remote communities had similar levels of psychological distress as women living in metropolitan areas. LIMITATIONS: ALSWH uses self-report questionnaires which may introduce potential self-report bias. The small sample size in areas outside of metropolitan areas resulted in the need to collapse small rural towns, remote communities and very remote communities into one category. CONCLUSIONS: The subtle regional differences in levels of psychological distress in young women in Australia highlights the importance of the link between health and where people live and suggests further work is required to understand the regional differences and encourage location specific mental health services.

Screen time and child behaviour and health-related quality of life: Effect of family context.

This study investigated family contextual effects on the association between screen time, behaviour and health-related quality of life (HRQOL) in child siblings. Data were from 1772 participants in the Australian Longitudinal Study of Women's Health 1973-78 cohort and their three youngest children (N = 4010 siblings) aged two to 12, collected in Mothers and their Children's Health cross-sectional sub-study (2016/17). The exposure was average daily recreational screen time (televisions, computers, tablets, mobile phones, electronic games). Outcomes were child overall behaviour and prosocial behaviour, and psychosocial and physical HRQOL. Multilevel models were used to test 1) the absolute effect of screen time, and 2) change in the child's outcomes when their own screen time changes relative to their siblings (within-family contextual effects) and change in the child's outcomes when the sibling average screen time changes (between-family contextual effects). The children were average 7.37 years (SD 2.76). There was an absolute effect of screen time on behaviour and HRQOL. Between-family effects were found for child total behaviour (0·22 (95%CI, 0.06, 0.37)) and prosocial behaviour (-0.10 (-0.17, -0.04)), but there were no within-family effects. For HRQOL, within-family effects were found for psychosocial (-0.97 (-1.91, -0.02)) and physical (-1.32 (-2.25, -0.39)) HRQOL, but no between-family effects were found. In conclusion, the higher the average screen time by all children in the family, the worse the child's behaviour, while the greater the deviation of the child's screen time from their siblings, the poorer their HRQOL. Family contextual factors should be considered in screen time research and screen time recommendations.

Association between preconception cannabis use and risk of postpartum depression: Findings from an Australian longitudinal cohort.

BACKGROUND AND AIM: Evidence suggests a positive association between cannabis use and depression however whether preconception cannabis use is associated with postpartum depression (PPD) remains unknown. This study examined the association using a longitudinal design. METHODS: Participants were from the Australian Longitudinal Study on Women's Health (cohort born in 1989-95), a sample broadly representative of similarly aged Australian women. The association of past year cannabis use at 3-15 months prior to pregnancy with PPD at ages 19-24 to 22-27 was examined in 516 women with 570 singleton live births. The associations of chronicity, recency, and initiation of cannabis use before pregnancy with PPD at ages 21-26 to 24-29 were investigated in 538 women with 590 singleton live births. PPD was ascertained from self-report of doctor diagnosis for each birth. Relative risk (RR) and 95 % confidence interval (CI) were used to calculate the associations of interest. RESULTS: Compared with no cannabis use, any past-year cannabis use at 3-15 months before pregnancy may be associated with an increased risk of PPD (1.50, 0.99-2.28). Compared with no cannabis use in the first three annual surveys before pregnancy, chronic use (a past year user in ≥ 2 surveys) was associated with an 80 % higher risk of incident PPD for births in the following 1-4 years (1.80, 1.22-2.68). Compared to never users at ages 20-25, former users who had no use within the past year had no increased risk; past year users had 73 % higher risk (1.73, 1.07-2.81); and past year users who initiated cannabis use at or after age 18 had twice the risk of incident PPD for births in the following 1-4 years (2.02, 1.17-3.49). CONCLUSION: This study provides evidence that preconception cannabis use is associated with increased risk of PPD. More studies are warranted to confirm this finding before cannabis cessation policies can be reinforced on reducing the risk of PPD.

Which Australian Women Do Not Exclusively Breastfeed to 6 Months, and why?

BACKGROUND: Rates of exclusive breastfeeding in Australia lag behind international targets. Reasons for non-exclusive breastfeeding are poorly understood. RESEARCH AIMS: To describe demographic profiles of participants reporting different feeding practices, and reasons for not exclusively breastfeeding to 6 months. METHODS: Demographics for 2888 mothers (5340 children) and reasons for 1879 mothers (3018 children) from the Mothers and Their Children's Health Study (a sub-study of the Australian Longitudinal Study on Women's Health) were examined using descriptive statistics and multivariable regression. RESULTS: Only 34.4% of children were exclusively breastfed to 6 months. Five non-exclusive feeding practices were identified: never breastfed (3.9%), breastfed < 6 months (20.8%), and breastfed to 6 months but had formula (6.8%), solids (24.5%), or both formula and solids (9.7%). Mothers of children who received < 6 months of human milk were more likely to have a lower education, be overweight/obese, smoke, and live in cities (compared to mothers of children exclusively breastfed). Reasons for never breastfeeding and for breastfeeding < 6 months were primarily insufficient milk and breastfeeding difficulties (e.g., latching issues). Reasons for introducing solids were primarily cues for solids (e.g., showing interest). Reasons for formula were insufficient milk and practical considerations (e.g., return to work). Reasons for both solids and formula were diverse, including insufficient milk, weaning cues, and practical considerations. CONCLUSIONS: Mothers who did not exclusively breastfeed to 6 months were a heterogeneous group, indicating that both targeted and universal strategies are required to increase rates of exclusive breastfeeding. Support should encompass the broad range of feeding practices.

Does premenstrual syndrome before pregnancy increase the risk of postpartum depression? Findings from the Australian Longitudinal Study on Women's Health.

BACKGROUND: Previous literature suggests a positive association between history of premenstrual syndrome (PMS) and development of postpartum depression (PPD); however, limited evidence has come from prospective population-based studies and whether history of depression affects this association is unknown. METHODS: This study included 5479 women from the 1973-78 cohort of the Australian Longitudinal Study on Women's Health and estimated the association between pre-pregnancy PMS and PPD. Participants were followed from 22-27 years in 2000 to 37-42 years in 2015. PMS was collected from a 4-category Likert-scale reporting on frequency of PMS in the last 12 months (never, rarely, sometimes, or often) at the survey preceding an index birth. PPD was ascertained from reports of doctor diagnoses for each birth. Relative risks (RRs) and 95% confidence intervals (CIs) were used to estimate the association of interest. The role of history of depression was assessed by testing its interaction with pre-pregnancy PMS. RESULTS: During 15 years' follow-up, 15.4% of participating women reported PPD; and 55.1% reported PMS (rarely: 17.2%, sometimes: 25.7%, and often: 12.2%). Compared to women who had no PMS before pregnancy, those who rarely had PMS had similar risk of PPD (1.03, 0.82-1.30); whereas those who sometimes or often had PMS had significantly higher risk of PPD (1.31, 1.09-1.57 and 1.51, 1.22-1.87, respectively). History of depression did not affect the association. LIMITATIONS: PMS was self-reported. PMS severity was not collected. CONCLUSIONS: This large population-based study provides evidence of a dose-response relationship between PMS prior to pregnancy and PPD, independent of history of depression. Evidence to date suggests PMS has the potential to help identify women at increased risk of PPD before pregnancy.