Keywords

Introduction

Late adolescence and early adulthood are times of major behavioral transition in young women as they become more independent and make choices about their health-related behaviors and lifestyle. Behaviors and lifestyle choices of young people have far-reaching consequences for future well-being, quality of life, and productivity [1], yet this is a much understudied age group [2]. Women drive health behaviors in our society [3], making it critical to understand the factors shaping health and lifestyle in young women and to implement effective interventions during the crucial ages of 16–25 years when health patterns undergo major transitions (Fig. 5.1) that can shape future health trajectories [4].

Fig. 5.1
figure 00051

The evolution of health risks in young Australian women (Kindly provided courtesy of Dr. Yeshe Fenner)

Full size image

Over 30 % of Australia’s burden of disease is due to modifiable risk factors (including smoking, alcohol abuse, physical inactivity, high blood pressure and cholesterol, low consumption of fruits and vegetables, overweight/obesity), placing preventive health care at the forefront of Australia’s national health strategy [5]. Yet changing health-related behaviors is a major challenge. Smoking causes 7 % of the total burden of disease in Australian women [6] and remains highly prevalent in young women: 18 % of females aged 18–24 were smokers in 2007 [7]. Physical inactivity is second only to smoking as the key risk factor associated with poor health outcomes [6], with poor diet a further contributor. Changes in physical activity and eating patterns during adolescence and early adulthood are associated with rising rates of obesity. The Australian government has called for action to halt this epidemic [8]: obesity rates in children and adults more than doubled over the last two decades [9]. Body mass index (BMI) in youth tracks into adult life [10] and obese or overweight young people are at higher risk of subsequent cardiovascular and metabolic disease [11].

Identifying and addressing risk factors for poor bone and joint health are also of great national importance, since musculoskeletal disorders in later life cause more disability than any other medical condition [12], imposing a large economic burden ($4.7 billion in Australia in 2000–2001 [13]). Musculoskeletal injuries and pain impact upon women’s ability to participate in physical and occupational activities and predispose to chronic conditions (e.g., low back pain and knee injuries sustained during sport, which are common and substantially increase the risk of knee osteoarthritis in later life).

In this study, we investigated the differences of key health and lifestyle factors, dietary calcium, alcohol consumption, smoking, physical activity, and BMI during different stages of adolescence and young adulthood. These factors are related to musculoskeletal development and bone health later in life (among many other health conditions). Having conducted extensive studies of lifestyle, diet, and health in female twins, more recently we have commenced population-based investigations of young women’s health determinants using recruitment through social networking sites (SNS) and online data collection methods [14] in a project called the Young Female Health Initiative (YFHI). This research complements the twin studies we report here and offers a powerful approach for ongoing health research in young people. This important line of research will help to determine how health-related behaviors and lifestyle changes influence health in young women and how we can change behaviors to improve their health outcomes.

Methods

We prospectively evaluated nutritional and lifestyle factors in 566 15–30-year-old female twins (both monozygotic and dizygotic). The participants were of various ages at first visit, attended for follow-up at variable intervals, and had variable numbers of visits, depending on opportunities for follow-up. Of the 566 twins, there were a total of 790 visits while they were in the targeted age range. These twins had participated in cross-sectional and longitudinal studies of constitutional, lifestyle, and dietary determinants of bone health in which they completed questionnaires to assess their health and lifestyle [15, 16]. This further analysis looked at the evolution of health risk factors across the crucial late adolescent and young adult years. Current daily calcium intake was measured by a short food frequency questionnaire based on one developed for Australian women [17]. The questionnaire captures approximately 60 % of dietary calcium intake on a typical Australian diet. Physical activity was determined over the previous 12 months by questionnaire [15, 16]. Hours of playing general sport per week and hours of walking per week were each recorded on a categorical scale (0–1, 2–3, 4–7, and >7 h), with the number of hours per week being entered as 0.5, 2.5, 5.5, and 8 h, respectively, for each category. Smoking habits were recorded by participants as ever smoking, current smoking, and smoking amounts. In those who had ever consumed alcohol (more than 12 standard drinks in their lifetime), alcohol consumption was calculated from average monthly consumption over the previous year and expressed as the number of standard drinks (a standard drink contains 10 g of alcohol). BMI was calculated using the standard formula [weight in kilograms/(height in meters × height in meters)]. Height was measured to the nearest 0.1 cm on a wall-mounted stadiometer, and weight was measured on a balance scale to the nearest 0.1 kg. All statistical analyses were carried out using SPSS (SPSS® 20.0, Chicago, IL, USA). Unlike previous reports on this cohort [15, 16], twinness was not included in the models since little or no effect on these descriptive data was anticipated. Nonparametric tests (chi-square, Mann-Whitney U, and Kruskal-Wallis) were used to identify significant differences in the selected lifestyle factors between age groups, to examine the age-related differences. One-way ANOVA was used for normal data. P values of <0.05 were considered statistically significant.

Results

Dietary calcium intake was relatively low [511 (321,747)] mg/day (median, IQR) and did not vary significantly with age (Fig. 5.2). The questionnaire used to calculate calcium assumes the questionnaire estimates 60 % of total dietary calcium. Taking this into account and that the daily recommended intake for most people is between 800 and 1,500 mg/day [18], almost half of this age group was below the required dietary calcium intake.

Fig. 5.2
figure 00052

Median (IQR) dietary calcium intake (mg/day) by short food frequency questionnaire representing an estimated 60 % of total calcium intake [17] in different age groups from 15 to 30 years

Full size image

The number of young women who reported ever consuming alcohol (defined as12+ standard drinks ever) increased from 50 % in those less than 18 years to 93–99 % for the 18+ age groups (p < 0.001: Fig. 5.3). Of those who consumed alcohol in the year prior to their visit, monthly intake doubled from under 18 years (5.7, 3.9, 19.0; median, IQR) to 18+ years (12.0, 4.7, 26.0; P <  0.001) with the highest consumers being 21–23 and 27–29 years (Fig. 5.4).

Fig. 5.3
figure 00053

Percentage of young women who have or have not consumed 12+ standard drinks of alcohol ever

Full size image
Fig. 5.4
figure 00054

Average monthly alcohol consumption (standard drinks) in the year prior to study visit in young women under 18 years and 18+ years

Full size image

At age 15–17 years, 14 % reported ever smoking and by age 27–29, 51 % had smoked (P  =  0.002). Under the age of 20 years, average cigarette consumption (Fig. 5.5) in smokers was six cigarettes [4, 10] per day, increasing to ten [2, 15] at 20 years and older (P  <  0.001). When looking at current smoking across the age subgroups, interestingly being a current smoker progressively increased from age 15 to 23 years of age (15 to 17 – 16.8 %, 18 to 20 – 21.4 %, 21 to 23 – 25.0 %), then declined from 24 to 30 years of age (24 to 26 – 21.9 %, 27 to 30 – 14.8 %; P  <  0.001). Therefore, the oldest age group, 27–30 years, had the least current smokers of all ages.

Fig. 5.5
figure 00055

Average daily cigarette consumption (median, IQR) in under 20 years and 20+ years

Full size image

Participation in sporting activity decreased progressively with age (P  <  0.001): 47.5 % of 15–17-year-olds undertook 4 or more hours of sport per week, compared with 23.5 % at age 27–29 years (Table 5.1). Conversely, sedentary behavior increased with age: 25.0 % of 15–17-year-olds reported 1 or less hour/week of exercise compared with 50.0 % at age 27–29 years. Changes in walking activity were complex, suggesting an increase with age (Table 5.2).

Table 5.1 Amount of sport (hours) played per week in the year prior to study visit
Full size table
Table 5.2 Number of hours spent walking per week in the year prior to study visit
Full size table

BMI increased progressively with age (P  =  0.011), the youngest group (15–17 years) having the lowest (21.3, 19.5, 23.6; median, IQR) and the oldest (27–29 years) having the highest (23.1, 21.5, 25.9) BMI (Fig. 5.6). Median BMI in all groups was within a normal range.

Fig. 5.6
figure 00056

Median (IQR) BMI in different age groups from 15 to 30 years

Full size image

There was no difference in height between age groups. Weight (kg) followed the same trend as BMI, progressively increasing with age (P  =  0.027). Again the youngest age group (15–17 years) was the lightest with mean weight 60.0 kg and the oldest age group (27–30 years) was the heaviest at 66.2 kg.

Discussion

These findings demonstrate highly significant and generally adverse changes in health-related behavior in young women as they transition into independent adult living. Many of these changes are predicted to impact adversely on bone and other health outcomes in later life. Indeed, studies of adolescent and young adult twins (age range 10–26 years) showed that the contribution of their shared environment to variance in bone mineral density (BMD) diminished dramatically as they started to live more independently in early adult life, when the contribution of environmental factors specific to each individual became evident [19]. These observations are consistent with a change in the environmental sources of variation in BMD across this age range, and most of these environmental changes are likely to relate to the individuals’ lifestyle. As we demonstrate here, physical activity, cigarette smoking, and alcohol consumption, all potentially adverse for bone health inter alia, varied significantly in this cohort of young twins across the age range 15–30 years.

There is a pressing need to improve understanding of the determinants of these changes and to develop effective interventions to improve long-term bone health and other outcomes in young women. The first step towards achieving these goals is to engage effectively with young women. However, young people in particular are underrepresented in medical and population-based studies as they are highly mobile, and recruitment and retention are difficult [2]. Traditional approaches to recruitment and retention of young women in health-related research seem unlikely to be fruitful in future, and it seems much more promising to engage with young women using the mobile and internet-based communication technologies with which they are so familiar and comfortable. Likewise, recruitment via SNS also is very appealing given their wide and almost universal reach in many countries, potential to recruit demographically representative samples where required, and the relatively modest cost of advertising via SNS. Our own group’s recent experience [14] in recruiting 16–25-year-old women for health research via Facebook was very positive. Over several months, we recruited 278 young women who completed an online health questionnaire, with approximately half choosing to do so at our study center and half remotely. Regional and remotely dwelling participants were well represented. The sample also was representative of the general population in this age range based on socioeconomic level and country of birth. Older subjects and those with a higher educational level were mildly overrepresented. Complete health questionnaire data were provided by >90 % of respondents. Among the results, self-reported weight problems were a major concern, with 24 % of 16–17-year-olds, 33 % of 18–21-year-olds, and 36 % of 22–25-year-olds classified as overweight or obese. Moreover, in a linear regression model, BMI increased by 0.29 kg/m2 per year of increase in age (P  <  0.01). In contrast, only 5–7 % reported being underweight (BMI  <  18.5).

The Facebook recruitment method also was highly cost-effective, costing only USD 20 per completing participant. Based on this experience, our group is pursuing Facebook recruitment, coupled with online and mobile data collection methods, in a range of observational health studies directed at young women. These methodologies also are readily adaptable for intervention trials, particularly targeting lifestyle and behavioral interventions in areas including nutrition, smoking cessation, physical activity, and sun exposure.

Conclusions

These findings demonstrate highly significant changes in behavior in young women as they transitioned into independent adult living. Many of these changes are predicted to impact adversely on bone and other health outcomes in later life. There is a pressing need to improve understanding of the determinants of these changes and to develop effective interventions to improve long-term bone health and other outcomes in young women. Our recent studies suggest that SNS and other information and communication technologies hold great promise for engaging young women in health research and ultimately for supporting health interventions in this demographic.