Behavioral impact of national health campaigns on healthy lifestyle practices among young adults in Singapore: a cross-sectional study

A total of 665 participants were recruited via convenience sampling from 16 to 21 December 2019. The participants were interviewed using an interviewer-administered, validated survey on healthy lifestyle practices. All participants recruited in this cross-sectional, quantitative study were Singapore Citizens or Permanent Residents aged 18 to 39 years old. Young adults are defined as those between the ages of 18–39 [41‐48], as this age group captures a broad developmental period [46] prior to the onset of routine clinical interventions for all individuals at age 40 onwards [12, 49]. We excluded participants who were pre-diabetic or had either Type 1 or Type 2 DM from recruitment, as these individuals would often have regular medical follow-ups that would have affected their lifestyle practices. Participants who were concurrently schooling and working (n = 44), as well as participants who were neither schooling nor working at the time of survey (n = 27), were excluded from the data analysis step, in order to evaluate the impact of full-time working versus full-time schooling on healthy lifestyle practices. This leaves a total of 594 participants with complete information for the data analysis. This study was approved by the National University of Singapore’s Institutional Review Board (S-19-346).

Participants were recruited from a total of 23 different areas island-wide, from a mix of business districts (n = 9), tertiary institutions (n = 9), and housing estates (n = 5). Interviewers approached young adults encountered in the recruitment locations to check if they were willing to participate in our survey and if they fulfilled our inclusion criteria. Informed consent was collected from all our participants prior to participation. Verbal consent was obtained from participants aged 21 years old and above, while verbal parental consent was obtained from participants aged below 21 years old. Upon collecting informed consent and ensuring that our participants fulfilled our inclusion criteria, interviewers proceeded to read out the survey and captured the responses using their mobile devices.

All participants were asked for their demographic factors such as age, sex, ethnicity, housing type, highest educational qualification, and whether they are working or schooling. Schooling participants are attendants of full-time formal education, including but not limited to education at universities, polytechnics, or any Institute of Technical Education. We determined each participant’s exposure to the WoD campaign by asking them whether they have “seen advertisements from the WoD campaign”. Questions on lifestyle practices were taken from the Simple Lifestyle Indicator Questionnaire [50] comprising: diet, exercise, alcohol consumption, and current tobacco use. In addition, we included a question on whether participants “have used the DRA tool”, which we used as a proxy to assess participants’ participation in specific health screening services.

We dichotomized the participant’s responses into healthy or unhealthy lifestyle practices based on Singapore and international recommendations. For diet, participants who consumed at least 2 servings of fruits and vegetables per day [51, 52], and at least 2 servings of whole-grains per day [53, 54], were considered to have a healthy diet. For exercise, the Singapore HPB and WHO recommends at least 150 min of moderate intensity exercise or at least 75 min of vigorous intensity exercise per week [55, 56]. Participants who had moderate intensity exercise of at least 4 times per week or vigorous intensity exercise of at least 3 times per week with a minimum of 30 min duration in each session, were considered to have a healthy lifestyle practice. For alcohol consumption, excessive alcohol consumption was defined as more than 1 drink for females and 2 drinks for males in a day as recommended by the Centers for Disease Control and Prevention [57]. Excessive alcohol consumption was considered an unhealthy lifestyle practice. Participants who responded “yes” to current tobacco use, were also considered as having an unhealthy lifestyle practice. Lastly, participants who responded “yes” to having used the DRA tool before, were considered as having participated in screening which constitutes a healthy lifestyle practice.

Descriptive statistics consist of counts and percentages for categorical variables, and means and standard deviations for continuous variables. As our outcome measures are common activities, the odds ratio is not a good approximation for the prevalence risk ratio (PRR). Therefore, to estimate the independent effect of WoD campaign and other independent factors on each lifestyle practice, adjusted PRR for each factor was obtained using a multivariable modified Breslow-Cox proportional hazards model [58]. Exposure to WoD campaign, sex, working/schooling status, and ethnicity were included into the multivariable model for each lifestyle practice. Age and highest educational qualification were excluded from the multivariable models as these variables have high collinearity with working/schooling status. Housing type was also excluded from the multivariable models due to fundamental discrepancies between younger participants who lived with their parents and older participants who owned their own home.

The minimum sample size based on the Singapore population of 5.6 million, assuming a maximum variability of 50%, with a medium effect size of 0.5 [59, 60], with a ± 5% margin of error and 95% confidence level, was 385 complete responses.

The proportion of people who met the minimum dietary recommendations was significantly higher among those who were exposed to the WoD campaign as compared to the unexposed (adjusted PRR = 1.6, 95% CI: 1.0–2.5, p = 0.037) (Table 3). Exposure to the WoD campaign was not significantly associated with meeting minimum exercise recommendations. However, sex was a significant factor where the proportion of people who met the minimum exercise recommendations among males was two times as high as females (adjusted PRR = 2.0, 95% CI: 1.5–2.7, p < 0.001). Working status was the other significant factor where lower proportion of working individuals met the minimum exercise recommendations as compared to schooling individuals (adjusted PRR = 0.7, 95% CI: 0.5–0.9, p = 0.019). Exposure to the WoD campaign also had a significant positive association with participation in health screening (adjusted PRR = 1.2, 95% CI: 1.0–1.5, p = 0.028). The remaining factors were not significantly associated with screening participation.

All the four factors were significantly associated with current tobacco use (Table 4). The proportion of people who were currently using tobacco among those who were exposed to the WoD campaign was half as high as those who were unexposed (adjusted PRR = 0.50, 95% CI: 0.30–0.80, p = 0.003). The proportion of smokers among males was almost 4 times as high as females (adjusted PRR = 3.9, 95% CI: 2.2–6.9, p < 0.001). Working individuals were more likely to smoke as compared to schooling individuals (adjusted PRR = 1.8, 95% CI: 1.1–3.1, p = 0.024). Non-Chinese individuals were also more likely to smoke as compared to Chinese individuals (adjusted PRR = 2.8, 95% CI: 1.7–4.6, p < 0.001).

Exposure to the WoD campaign was not significantly associated with excessive alcohol consumption. Only sex was a significant factor where the proportion of people who consume alcohol excessively was higher among males as compared to females (adjusted PRR = 1.5, 95% CI: 1.0–2.3, p = 0.046) (Table 4).

In this study, we found that exposure to a national health campaign, such as the WoD campaign, had a significant positive association with the lifestyle practices among young adults. Participants who were exposed to the national health campaign were more likely to meet dietary recommendations, participate in health screening, and not currently use tobacco. However, exposure to the national health campaign was not significantly associated with alcohol consumption. This pattern of influence in the young adult age group was consistent with a review in the general population, that found mass media campaigns to have moderate to strong evidence for benefit when it comes to nutrition, tobacco usage, and screening, but only little evidence for benefit when it comes to alcohol consumption [6]. Other studies conducted in the general population have similarly reported significant reduction in tobacco usage with the use of effective health campaigns [61‐63], but only limited improvement when it comes to reducing alcohol consumption [64, 65]. Interestingly, even though alcohol and tobacco have been described to possess similar competing influences against positive behavior change (i.e. marketing, pricing, social norms and addiction) [6], reducing alcohol consumption seems to be more resistant to the effects of national health campaigns compared to tobacco use among young adults. This may be because alcohol is still considered a socially acceptable beverage in social settings such as gatherings with family and friends [66, 67]. This is in contrast to using tobacco, which is stigmatized to be a vice [68], and is fast becoming frowned upon in many professional settings [69]. As such, public health authorities can consider exploring other methods of portraying excessive alcohol consumption as an unhealthy lifestyle decision in the public consciousness. Nonetheless, these findings affirm the use of mass media campaigns to promote healthy nutrition, discourage tobacco use and encourage participation in screening programs.

Interestingly, this study found that exposure to the national health campaign was not significantly associated with meeting exercise recommendations. This was contrary to other studies that have demonstrated health campaigns to be effective in increasing physical activity among the general population [70‐72]. Health campaigns have been described to have moderate evidence for benefit [6]. This might be because the overall prevalence of participants who already met exercise recommendations, regardless of their exposure to the national health campaign, was high. This may indicate that the majority of young adults are self-motivated in keeping active, or it may be the effect of mandatory military conscription in Singapore which has boosted the prevalence.

We also found that full-time working and schooling status had a significant association with the lifestyle practices of young adults. Working participants were less likely to meet the minimum exercise recommendations and more likely to currently use tobacco compared to the schooling population. Schooling participants have more unstructured time weaved into their curriculums and more opportunities to access organized sporting activities, such as collegiate athletics or clubs [73‐75]. Whereas in the workplace, a lack of unrestricted time coupled with an unsuitable physical environment are the common barriers [76] that make working adults less likely to achieve the minimum exercise recommendations. Meanwhile, the financial independence that comes with taking on a full-time job may play a contributory role in higher rates of tobacco use among the working participants, because higher income has been found to reduce price sensitivity to cigarettes among tobacco users [77]. Financial independence also results in the autonomy to spend, as young working adults may see an elimination in parental restrictions on cigarettes bought with their own money rather than with their parents’ money [78]. Furthermore, moving from school to the workplace also meant no school-imposed sanctions on tobacco use [79, 80]. These environmental factors could explain the difference between lifestyle practices among schooling and working participants within this age group. Meanwhile, schools can consider moving beyond solely relying on disincentives to deter tobacco consumption and focus on the root cause by educating young adults on different methods to cope with stress situations while practicing financial prudence [81].

In this study, sex had a significant association with the lifestyle practices of young adults. Specifically, males were more likely to meet exercise recommendations, but also more likely to currently use tobacco and consume alcohol excessively. In Singapore, all able-bodied males above the age of 18, have to serve a mandatory full-time conscription of 2 years, and continue to be operationally fit for duty until the age of 40 years old [82]. While females are also able to join the conscription on a volunteer basis, they make up the minority of service personnel in the Singapore armed forces. This may explain why males in this age group were more likely to meet the minimum exercise recommendations, regardless of their exposure status to the WoD campaign. However, military service has been associated with the adoption of some unhealthy lifestyle practices such as tobacco use and alcohol consumption. Due to the tight laws in Singapore, participants are likely to underreport and not attribute their adoption of these practices to their time in the armed forces [83]. However, a study on the Republic of Korea Air Force soldiers found that 24.6% of survey respondents reported picking up smoking after joining military service [35]. A thorough review into the American armed forces conducted by the Department of Defense [84], also revealed that individuals were more likely to pick up tobacco use and alcohol consumption in the military. As high as 38% of service personnel had picked up tobacco use after enlisting, and 5% claim that they smoked in order to “fit in with their military unit”. In addition, 55% of tobacco users reported that they smoke when consuming alcohol, suggesting the close relationship between both unhealthy lifestyle practices. Up to 82% and 93% of participants also reported using tobacco and consuming alcohol when around their friends in order to facilitate social networking. It is believed that such activities have a long tradition in the military and that it forges stronger camaraderie among fellow soldiers [85]. Different social pressures might be one of the unique factors in this age group to consider which may explain why sex was moderately associated with some lifestyle practices. Further investigation to elucidate the source of these social pressures may be important to institute appropriate interventions to target the male young adult population.

This study has several limitations. One limitation was the cross-sectional nature of this study. Therefore, we acknowledge our inability to establish causality in the relationships. Nonetheless, we have a low proportion of respondents who were excluded from final analysis, with a final analyzed proportion of 89% of our survey respondents. Another limitation was our use of a non-probability, convenience sampling method. This was due to the presence of strict confidentiality laws in Singapore which precludes our possession of a list of personal data for the systematic recruitment of our participants. However, we have made every effort to strive at a representative sample, by recruiting a large number of participants (total n = 665) throughout an array of different locations island-wide (n = 23). In addition, the age restriction of the sample population to young adults, may limit the generalizability of the findings to individuals of other age groups. However, this also constituted a strength of this study, as to the best of our knowledge, this is one of the first studies of its kind, to have directly investigated the associations between exposure to a national health campaign and various lifestyle practices young adults aged 18 to 39 years old. In doing so, this study has raised interesting findings regarding this age group, which should serve as the springboard for further evaluation in future studies. This age group, which is neither completely appreciated when subsumed under the label of adolescence nor adults, often experience critical transitions in multiple aspects of their life at a time, which are deserving of special attention in the field of public health and preventive medicine.

We acknowledge that there are other socio-environmental factors that may affect the impact of health campaigns on the young adult age group. These factors, detailed below, were not covered in our current study but are important and could be the topic of investigation in future studies. In particular, perceived health risks amongst young adults could play a role because younger people have been found to have a more optimistic perception of their health compared to older age cohorts [86‐88]. As such, they might not prioritize adoption of healthy lifestyle practices regardless of exposure to the WoD campaign. Moreover, many young adults get married during this age group. Thus, we postulate that marriage plays a role in one’s lifestyle practices given that it has been described to be associated with body mass index and weight related behaviors, including exercise [89]. The emergence of parenthood is also a unique and interesting factor to investigate in this age group. Previous studies have revealed that new parenthood is also associated with lower physical activity levels amongst the decline of other healthy practices [90, 91]. The responsibilities and commitments of new parenthood may exhaust young adults to disregard healthy lifestyle practices [92], or it may lead to the adoption of healthy lifestyle practices so as to set an exemplary behavior for their children to follow [93]. While these factors are beyond the scope of this study, they deserve further evaluation in future studies to understand the unique circumstances young adults face, and their effect on the adoption/deterioration of certain lifestyle practices during this age.

Moreover, there are many new programs as part of the WoD campaign that are soon to be implemented in the community [94]. While this study showed encouraging early behavioral impact of the WoD campaign on individuals, change in overall outcomes takes time to manifest [95, 96]. It would be valuable for future studies to follow up on the subsequent initiatives and evaluate how they augment the campaign’s current outreach and impact.