Alcohol types

Multimorbidity is associated with television viewing, but not with other types of screen-time behaviors in Brazilian adults | BMC Public Health

We found significant associations between television viewing and multimorbidity in this nationally representative sample of nearly 88,000 Brazilian adults of both sexes. Importantly, we also observed an apparent dose-response relationship, since the higher the television viewing time category, the higher the associations. In contrast, in general, we found no association between other screen use and multimorbidity, except in men (highest time category) and the elderly (two highest time categories) . Sensitivity analyzes revealed that the association with television viewing time also increased with increasing number of chronic diseases, as well as in those who watched television ≥ 3 h/day there was an increased likelihood of multimorbidity in all age groups 35 years and older.

Our results confirm previous studies that have found associations between multimorbidity and television viewing in both sexes. In European adults, associations were found in all time categories for women and men, from at least 2 h/day of exposure [13], compared to 3 h/day in our study. Previous studies confirm greater associations with multimorbidity in women than in men [5, 22, 23]. These differences may be explained by the fact that women have a longer life expectancy and spend more years with multimorbidity. [23] as well as these women seek health services more frequently than men [24]. Our results showed that 35.5% of women had multimorbidity, while the proportion in men was lower (25.1%). However, the associations were similar in both sexes, with a slight predominance in women, with a significant association in all time categories assessed. Either way, our data confirms that television viewing is associated with widespread adverse health effects. [25,26,27].

Analyzes stratified by age group confirm these associations between multimorbidity and television consumption, except among the youngest. A previous study with a sample of Brazilians identified associations with ≥ 2 h of television viewing, both among adults and the elderly [15]. However, this study only used four chronic conditions to classify multimorbidity, whereas we considered 12 chronic conditions; the authors used a nationally representative sample of state capitals while we used state capitals and non-state capitals, and they studied associations with two different time categories while we studied four. Additionally, due to the threshold used in this study to categorize age groups, it was impossible to identify associations with younger adults to compare our results. There is evidence that age is correlated with multimorbidity and that the prevalence in the elderly is higher than in other age categories [7, 22, 28, 29]. In our study, the proportion of individuals with multimorbidity was only 9.5% in young adults and 60.9% in the elderly, which could explain these associations. Although it is impossible to establish a causal link, this information confirms the importance of reducing sedentary behaviors, expressed by the time spent in front of a screen, mainly in front of the television, with particular attention for the elderly. [30].

In addition to multimorbidity, we identified that watching television was also associated with a higher number of chronic diseases. A previous study with a sample of Brazilians living in state capitals has already shown associations with ≥ 2 h/day of television viewing and multimorbidity, with higher odds observed with an increasing number of chronic diseases [15]. A review study with data from six low- and middle-income countries demonstrated that the prevalence of sedentary behavior increased with the number of chronic conditions (from zero to four), however, the measure used was time spent sitting and cut-off point was from 8h/day [12]. Another study, which also used time spent sitting, found that sedentary behavior increased with the number of chronic conditions, and people with four or more conditions were significantly more likely to exhibit high sedentary behavior. [11]. From these results and the results found in our study, it appears that the number of chronic diseases influences the values ​​of the prevalence ratio of these associations, which seems to be a dose-response trend, but which can only be confirmed in a study. longitudinal to demonstrate cause and effect.

Regarding time spent on other screens, associations were only identified in stratified analyses, in the highest time category for men and the two highest time categories for the age group at from 65 years old. In our study, the proportion of individuals who spent more time on other screens was higher among those without multimorbidity, and this may have contributed to the associations being found only among these sub-groups, which are also the age group that presents proportionally the most multimorbidity. compared to other age groups. Additionally, age may have a moderating effect on this association because when the analysis was stratified by age, the association was found for the over 65 age group.

Despite studying isolated cardiometabolic markers, instead of multimorbidity, a previous study with an Asian urban population also found associations with television viewing but not with other screens (computer use and time reading). The authors believe that dietary factors and body mass index are potential moderators that explain these associations. [31]mainly due to bad habits associated with television viewing [26], which has not yet been confirmed for other screens. Another aspect that may explain the differences in results between television viewing and other screens is the time between the onset of exposure and the onset of the result, and longitudinal studies are needed to test whether the effects of time on other screens will be seen among young people in the future. Moreover, viewing television and other screens has different correlates regarding age and socio-economic variables. While more TV viewing is associated with lower education, more screens are associated with younger adults [32]. Additionally, the two behaviors may exhibit different characteristics regarding, for example, the number of hits, pauses, and overlapping behaviors, such as standing while using the cell phone.

Although the trend of decreasing time spent in front of television among Brazilian adults has already been observed [16], we did not identify this migration to other screens in our study. Besides the fact that the prevalence of individuals with multimorbidity is higher among those who spend more time in front of the television, the absence of association between multimorbidity and the other types of behaviors in front of screen (except those which appear in the stratifications by sex and age) indicate that the two types of behaviors have different relationships with the evaluated outcome. Despite the reduction in television viewing in recent years, the percentage of people who watch television 2 hours/day or more is still considerable (42.7% – see Supplementary Table 1 in Supplementary Material 1). Thus, it becomes important to observe both behavior that is more typical and behavior that is becoming more and more frequent.

Regarding strengths, our study was the first in Brazil with a nationally representative sample to use two measures of sedentary behavior (television viewing and other screen types) to investigate associations with multimorbidity, stratified by sex, age group and according to the number of chronic diseases. Our results elucidate differences in associations between types of screen-related behaviors and health outcomes. This can help direct specific actions for the control and monitoring of this condition.

Our study has certain limitations. As this is a cross-sectional study, it is not possible to infer cause and effect. Additionally, other screens (computer, tablet, or mobile phone) were assessed together, making it impossible to analyze associations for each type of screen behavior. In addition, all variables were self-reported, and the choice of chronic conditions that constituted the multimorbidity variable could render condition-specific associations adopted in this study. However, epidemiological studies conducted in low- and middle-income countries have used an average of eight to 12 chronic diseases to compose a measure of multimorbidity, and our study was based on a list of 25 most widely considered conditions to compose the measured in these studies. [33]. It is also critical to consider the possibility of reverse causation, as people with multimorbidity may exhibit increased limitations in bodily mobility, which could predispose them to higher screen-based behaviors.