Abstract
Background:
Population-based studies on the prevalence of screen-time based activities and their role in health-related quality of life (HRQoL) remain lacking.
Objective:
This study aimed to examine the associations between screen time and HRQoL as well as the mediation roles of physical activity and loneliness in these relationships in community-dwelling adults.
Methods:
A population-based survey was conducted through telephone interviews in Hong Kong. Participants (N = 1070) were asked about their status as internet gamers and social media users, time spent on internet games and social media, moderate-to-vigorous physical activities (MVPA), loneliness, and HRQoL (EQ-5D and EQ-VAS). Descriptive statistics, regression, and path analyses were conducted.
Results:
The mediation models of status of social media users and internet gamers by age groups (younger adults: 18-65 years vs older adults: >65 years) showed that MVPA significantly fully mediated the effect of social media users on EQ VAS in younger adults, and partially mediated its effect on EQ-5D index in older adults. However, internet gamers were only associated with more loneliness and better EQ VAS in younger adults but not older adults. The mediation model of social media time among social media users by age groups showed that loneliness significantly and completely mediated the effect of social media time on EQ-5D index only in younger adults.
Conclusions:
Internet games and social media may have different roles in health-related behaviors and quality of life. The findings highlight the importance of recognizing the multifaceted influences of screen time on adults’ well-being across ages.
Keywords: screen time, physical activity, loneliness, quality of life, community-dwelling adults
Introduction
With the popularity of mobile phones and the Internet, internet gaming and social media use have become major activities for social connection, entertainment, and information. Studies on young to middle-aged adults reported an average social media usage of 2 to 3 hours per day.1-3 A Canadian study of 1238 adults aged 18 to 88 years found that 48.2% of the participants engaged in video gaming for at least 8 hours per week, 4 whereas another study with 844 participants aged 18 to 94 years reported a lower average of 22.87 minutes of gaming per day. 5 Such emerging screen time and technology-related lifestyles may significantly affect individuals’ health-related quality of life (HRQoL).
HRQoL is a subset of quality of life that focuses on health aspects, including individuals’ functioning and well-being in the physical, mental, and social domains of life.6-8 The negative relationship between screen time and HRQoL has been consistently reported in adolescent populations,9,10 while few studies were conducted in adults. 11 We only identified one study in community-dwelling adults, but screen-time was measured by the time watching TV and using a computer during leisure time and/or at work. 11 No study investigated whether and how internet gaming versus social media use may affect HRQoL in community-dwelling adults.
Time spent on internet games and social media may impact HRQoL via several mechanisms (eg, lifestyle and social mechanisms). From a lifestyle perspective, escalating screen time on internet gaming and social media may bring about lifestyle changes. The most direct correlates may be changes in sedentary behavior and physical activity. Prolonged screen time may imply a sedentary lifestyle and insufficient physical activity. Studies demonstrated a negative association between screen time and physical activity among young adults or college students,12,13 while studies on older adults were lacking. In turn, individuals who engage in sedentary behavior and spend minimal time on physical activities have been proved to be at risk of developing chronic diseases and poor HRQoL.14,15 A nationwide cross-sectional study conducted across all adult age groups in China reported that individuals with vigorous physical activity had a better emotional state and less screen time than those with light physical activity. 16
Socially, screen time may affect feelings of loneliness, which in turn could reduce HRQoL. 17 Loneliness has been a well-documented social determinant of HRQoL. 18 However, the relationship between screen time and loneliness remains unclear. Previous studies have found positive correlations between screen time and loneliness, 19 while other studies reported negative correlations. 20 Most related studies were conducted in young adults and college students. More empirical studies in population-based populations are warranted.
Theoretically, the potential mediation role of physical activity and loneliness could be supported by displacement theory,21,22 which asserts that when individuals devote more time and attention to one activity, such as online gaming and social media, it can detract from the time and attention they allocate to other activities, like sufficient exercise and face-to-face communication or interactions. This theory implies that there is a finite amount of time and attention available to individuals. As a result, engaging heavily in one activity may lead to a reduction in the time and resources allocated to other activities. However, the theory of information and communication technologies (ICTs) may suggest a different role of screen time. 23 According to this theory, ICTs can increase access to health-related information, coping resources, and entertainment, enable communication transfer, improve self-management, and help people cope with loneliness. 23 People using ICTs may experience enhanced social interactions, facilitated access to support networks, improved self-efficacy in managing health concerns, and greater empowerment in making informed health-related decisions. Therefore, screen time may increase physical activity and reduce loneliness, which in turn may benefit HRQoL. These 2 theories thus offer competing predictions regarding the direction of the mediation pathways.
Purpose
This study aims to examine the relationship among screen time (on internet games and social media), physical activity, loneliness, and HRQoL in a population-based sample of community-dwelling adults in Hong Kong. It is hypothesized that screen time would be negatively associated with HRQoL and reduce HRQoL indirectly via decreasing physical activity and increasing loneliness. In addition, subgroup analyses by age are conducted given that the importance of specific HRQoL domains varies across the adult lifespan. In younger adults (18-65 years), HRQoL may be typically more influenced by social functioning and emotional problems. In contrast, for older adults (>65 years), HRQoL is more strongly determined by physical functioning, mobility, and the presence of chronic diseases. Younger adults also have been found to be likely to have more screen time, 24 more physical activities, and better HRQoL 25 than older adults. These age-related differences in both screen time patterns and the determinants of HRQoL suggest that the relationships among screen time, physical activity, loneliness, and HRQoL may not be uniform across age groups. It is hypothesized that negative effects (displacement) may be more pronounced in younger adults who spend more time on screens, while positive effects (ICT benefits) may be more evident in older adults who may benefit from digital social connection. Understanding the role of these modifiable factors and mediators will guide the formulation of targeted lifestyle interventions and policies with the goal of promoting HRQoL across the adult lifespan.
Methods
Study Design and Study Population
This study was a population-based cross-sectional telephone survey conducted in Hong Kong between May 14 and June 4, 2020. Inclusion criteria of the participants included: (1) Chinese speaking, (2) aged 18 years or older, and (3) residents of Hong Kong holding valid Hong Kong identity cards. Of the 1882 individuals contacted, 1070 consented to participate in and finished the survey. The study procedures were carried out in accordance with the Declaration of Helsinki. Ethics approval was obtained from the Survey and Behavioral Research Ethics Committee of the corresponding author’s affiliating university (Ref.# SBRE-19-645). All participants provided verbal informed consent.
Data Collection
To ensure a representative sample that included working individuals, telephone interviews were conducted between 6 PM and 10 PM. The interviewers, supervised by a senior project coordinator, were extensively trained and possessed a minimum of 6 months of interviewing experience. Random phone numbers were drawn from the most recent residential telephone directory using a random phone number generator program. Landline telephone numbers were selected as seed numbers, and 3 additional sets of numbers were generated by randomizing the last 2 digits to include unlisted numbers. Eligible household members, chosen based on their birth dates closest to the survey date, were invited to participate. For unanswered calls, 2 follow-up attempts were made before considering the telephone number invalid. Verbal informed consent was obtained from the participants, and the anonymous interviews lasted 10 to 15 minutes. No incentives were provided to the participants.
Measures
Status of Screen Time
In this study, ‘screen time’ is conceptualized along 2 dimensions. First, user status (yes/no for internet gaming such as Candy Crush, Minecraft, League of Legends, or Fortnite and social media use such as WhatsApp, Twitter, or WeChat) reflects whether an individual engages with these digital platforms at all in the past year, capturing the potential benefits of access to information, social support, and entertainment. Second, time spent (daily hours) on internet gaming and social media, respectively, reflects the intensity or dosage of use, which may capture risks associated with excessive engagement. The concrete hours provided by the participants were recorded.
Physical Activity
Participants were asked how much time they spent engaging in moderate-to-vigorous physical activities (MVPA) in the last 7 days according to the definition of the World Health Organization. 26 Participants were educated about various forms of MVPA, such as brisk walking, dancing, gardening, household chores, social gatherings, walking pets, lifting objects weighing over 20 kilograms, running, fast-paced uphill walking/hiking, cycling at a fast pace, aerobic exercises, fast swimming, competitive sports, and games such as traditional sports, soccer, volleyball, hockey, basketball. 27 Their responses included (1) None at all; (2) Less than 50 minutes; (3) 50 to 100 minutes; (4) 100 to 150 minutes; or (5) More than 150 minutes over the 7 days.
Loneliness
Loneliness was measured using the 6-item De Jong Gierveld Loneliness Scale (DJGLS). 28 It measures social and emotional loneliness (range: 0-3). Higher scores indicate higher loneliness. The DJGLS is a reliable and valid loneliness scale for Chinese older adults. 28 The scale reliability was acceptable in the current sample (Cronbach’s alpha = 0.70).
Health-Related Quality of Life
The EuroQol Group 5-Dimension (EQ-5D) was used to measure HRQoL. 29 It consists of 2 components. First, a descriptive system (EQ-5D index) comprising 5 dimensions (mobility, self-care, usual activities, pain/discomfort, and anxiety/depression), each measured by a single item (total 5 items). Each item is rated on a 5-point severity scale: 1 = no problems, 2 = slight problems, 3 = moderate problems, 4 = severe problems, 5 = unable to/extreme problems. After reversing the scores, higher scores indicate better HRQoL on each dimension. These 5 items can be combined into a summary index (EQ-5D index) using a population-specific value set, with higher scores indicating better HRQoL. Second, the EQ VAS (visual analogue scale) is a separate, single-item measure on which respondents rate their overall current health from 0 (“the worst health you can imagine”) to 100 (“the best health you can imagine”). The EQ VAS thus captures a global self-perceived health status distinct from the index score. Scale reliability for the EQ-5D index in the current sample was good (Cronbach’s alpha = 0.77).
Statistical Analyses
Descriptive statistics were calculated for variables related to sociodemographic characteristics. To examine the relationships between socio-demographic characteristics, screen time variables (internet gamers, internet gaming time, social media users, and social media time), MVPA, loneliness, and HRQoL, univariable and multivariable linear regression analyses were conducted. Path analysis was used to examine the relationship between screen time variables and HRQoL with MVPA and loneliness as mediators and significant socio-demographic variables of outcomes as controlled factors. The goodness-of-model fit was evaluated using χ2/df value, comparative fit index (CFI), nonnormed fit index (NNFI), and root mean square error of approximation (RMSEA). The following cut points were applied to indicate a good fit between the hypothesized model and the observed data: (1) χ2/df ratio ≤3; (2) CFI and NNFI ≥0.90; (3) RMSEA ≤0.08.30,31 Bootstrapping analyses were conducted to identify the significant mediators. For subgroup analyses, age was dichotomized into younger adults (18-65 years) and older adults (>65 years). SPSS version 29 was used for these analyses, and IBM SPSS Amos Version 26.0 (IBM Corp., Armonk, NY, USA) was used for mediation analysis. The level of statistical significance was .05.
Results
Participant Characteristics and Descriptive Statistics
Of the 1070 participants, 658 (61.5%) were social media users, while 177 (16.5%) were internet gamers in the past year (all of them were also social media users); 67.7% of the participants were female, 51.6% were middle aged (36-65 years old), 70% were cohabiting or married, and 69.1% had secondary school education or below. Most participants had monthly household income ≤20 000 HKD and had no chronic disease. The distribution of socio-demographic characteristics differed significantly between social media users and nonusers, as well as between internet gamers and nongamers (see Table 1 for full comparisons).
Table 1.
Socio-Demographic Characteristics of the Participants.
| Socio-demographic characteristics | All participants (N = 1070) |
Social media user (n = 658) |
Social media non-user (n = 412) |
P a | Internet gamer (n = 177) |
Internet non-gamer (n = 893) |
P a |
|---|---|---|---|---|---|---|---|
| Sex, n (%) | <.001 | <.001 | |||||
| Male | 346 (32.3) | 242 (36.8) | 104 (25.2) | 84 (47.5) | 262 (29.3) | ||
| Female | 724 (67.7) | 416 (63.2) | 308 (74.8) | 93 (52.5) | 631 (70.7) | ||
| Age, n (%) | <.001 | <.001 | |||||
| 18-25 years | 74 (6.9) | 72 (10.9) | 2 (0.5) | 46 (26.0) | 28 (3.1) | ||
| 26-35 years | 41 (3.8) | 41 (6.2) | 0 (0.0) | 25 (14.0) | 16 (1.8) | ||
| 36-45 years | 106 (9.9) | 98 (14.9) | 8 (1.9) | 28 (15.8) | 78 (8.7) | ||
| 46-55 years | 146 (13.6) | 131 (19.9) | 15 (3.6) | 18 (10.2) | 128 (14.3) | ||
| 56-65 years | 301 (28.1) | 195 (29.6) | 106 (25.7) | 34 (19.2) | 267 (29.9) | ||
| >65 years | 383 (35.8) | 110 (16.7) | 273 (66.3) | 21 (11.9) | 362 (40.5) | ||
| Refuse to answer/missing value | 19 (1.8) | 11 (1.7) | 8 (1.9) | 5 (2.8) | 14 (1.6) | ||
| Current marital status, n (%) | <.001 | <.001 | |||||
| Single | 201 (18.8) | 176 (26.7) | 25 (6.1) | 84 (47.5) | 117 (13.1) | ||
| Cohabiting/married | 749 (70.0) | 448 (68.1) | 301 (73.1) | 83 (46.9) | 666 (74.6) | ||
| Separated/divorced | 15 (1.4) | 8 (1.2) | 7 (1.7) | 2 (1.1) | 13 (1.5) | ||
| Widowed | 92 (8.6) | 16 (2.4) | 76 (18.4) | 5 (2.8) | 87 (9.7) | ||
| Refused to answer/missing value | 13 (1.2) | 10 (1.5) | 3 (0.7) | 3 (1.7) | 10 (1.1) | ||
| Education level, n (%) | <.001 | <.001 | |||||
| Primary school or below | 355 (33.2) | 74 (11.2) | 281 (68.2) | 16 (9.0) | 339 (38.0) | ||
| Secondary school | 384 (35.9) | 281 (42.7) | 103 (25.0) | 64 (36.2) | 320 (35.8) | ||
| College or above | 294 (27.5) | 281 (42.7) | 13 (3.2) | 92 (52.0) | 202 (22.6) | ||
| Refused to answer/missing value | 37 (3.5) | 22 (3.3) | 15 (3.6) | 5 (2.8) | 32 (3.6) | ||
| Monthly household income, n (%) | <.001 | <.001 | |||||
| ≤20 000 HKD | 627 (58.7) | 265 (40.3) | 362 (87.9) | 74 (41.8) | 553 (61.9) | ||
| 20 001-30 000 HKD | 124 (11.6) | 112 (17.0) | 12 (2.9) | 34 (19.2) | 90 (10.1) | ||
| 30 001-50 000 HKD | 104 (9.7) | 101 (15.3) | 3 (0.7) | 28 (15.8) | 76 (8.5) | ||
| >50 000 HKD | 87 (8.1) | 81 (12.3) | 6 (1.5) | 16 (9.0) | 71 (8.0) | ||
| Refused to answer/missing value | 128 (12.0) | 99 (15.0) | 29 (7.0) | 25 (14.1) | 103 (11.5) | ||
| Chronic diseases, n (%) | <.001 | <.001 | |||||
| No | 716 (66.9) | 513 (78.0) | 203 (49.3) | 144 (81.4) | 572 (64.1) | ||
| Yes | 354 (33.1) | 145 (22.0) | 209 (50.7) | 33 (18.6) | 321 (35.9) |
Abbreviation: HKD: Hong Kong Dollar.
Chi-square tests.
Associations Between Socio-Demographics/Study Variables and HRQoL
According to univariable regression analyses (Table 2), women exhibited lower HRQoL compared to men (EQ-5D: β = −0.09, 95% CI for B [−0.54 to −0.11]; EQ VAS: β = −0.09, 95%CI for B [−3.72 to −0.67]). Older age was associated with reduced HRQoL (EQ-5D: β = −0.34, 95%CI for B [−0.13 to −0.07]; EQ VAS: β = −0.27, 95%CI for B [−9.59 to −3.88]). Compared to single individuals, those who were cohabiting/married and widowed had significantly lower HRQoL (EQ-5D: β = −0.20, 95%CI for B [−3.70 to −0.11]; EQ VAS: β = −0.24, 95%CI for B [−3.79 to −0.10]). Higher education levels (EQ-5D: β = 0.31, 95%CI for B [5.12 to 8.71]; EQ VAS: β = 0.20, 95%CI for B [5.11 to 8.69]) and higher household income (EQ-5D: β = 0.39, 95%CI for B [3.73 to 9.12]; EQ VAS: β = 0.38, 95%CI for B [3.74 to 9.02]) were linked to better HRQoL. The presence of chronic diseases was associated with lower HRQoL (EQ-5D: β = −0.25, 95%CI for B [−8.03 to −4.95]; EQ VAS: β = −0.26, 95%CI for B [−8.00 to −5.05]). Internet gamers (EQ-5D: β = 0.11, 95% CI for B [0.22 to 0.75]; EQ VAS: β = 0.13, 95% CI for B [2.40 to 6.22]) and social media users (EQ-5D: β = 0.30, 95% CI for B [0.79 to 1.18]; EQ VAS: β = 0.22, 95% CI for B [3.91 to 6.78]) had significantly higher HRQoL compared to nonusers. Higher levels of MVPA were associated with better HRQoL (EQ-5D: β = 0.11, 95% CI for B [0.02 to 0.07]; EQ VAS: β = 0.12, 95% CI for B [1.57 to 6.34]). Increased loneliness was linked to lower HRQoL (EQ-5D: β = −0.07, 95% CI for B [−0.01 to −0.00]; EQ VAS: β = −0.13, 95% CI for B [−1.15 to −0.43]). According to multivariable regression analyses, widowed (β = −0.12, 95% CI for B [−0.10 to −0.02]), secondary school education levels (β = 0.13, 95% CI for B [0.02 to 0.06]), the presence of chronic diseases (β = −0.16, 95% CI for B [−0.07 to −0.03]), social media users (β = 0.09, 95% CI for B [0.01 to 0.05]), and the frequency of MVPA (less than 50 minutes: β = 0.08, 95% CI for B [0.00 to 0.04]; 50-100 minutes: β = 0.08, 95% CI for B [0.01 to 0.05]; more than 150 minutes: β = 0.08, 95% CI for B [0.01 to 0.06]) were significantly associated with EQ-5D. Regarding EQ VAS, significant independent variables included the presence of female participants (β = −0.06, 95% CI for B [−3.02 to −0.01]), the presence of chronic diseases (β = −0.19, 95% CI for B [−6.32 to −3.13]), the level of loneliness (β = −0.13, 95% CI for B [−1.11 to −0.42]), and the frequency of MVPA (less than 50 minutes: β = 0.11, 95% CI for B [0.77 to 4.39]; 50-100 minutes: β = 0.13, 95% CI for B [1.76 to 5.93]; 100-150 minutes: β = 0.10, 95% CI for B [2.03 to 8.78]; more than 150 minutes: β = 0.11, 95% CI for B [1.31 to 5.94]).
Table 2.
Associations Between Socio-Demographics/Study Variables and HRQoL (N = 1070).
| Variables | EQ-5D index | EQ-5D index | EQ VAS | EQ VAS |
|---|---|---|---|---|
| βu (95% CI for B) | βm (95% CI for B) | βu (95% CI for B) | βm (95% CI for B) | |
| Sex | ||||
| Male | Ref. | Ref. | Ref. | Ref. |
| Female | −0.09 (−0.54, −0.11) | −0.04 (−0.03, 0.01) | −0.09 (−3.72, −0.67) | −0.06 (−3.02, −0.01) |
| Age | ||||
| 18-25 years | Ref. | Ref. | Ref. | Ref. |
| 26-35 years | −0.02 (−0.06, 0.04) | −0.01 (−0.05, 0.04) | 0.01 (−3.83, 4.73) | 0.03 (−2.56, 5.98) |
| 36-45 years | −0.01 (−0.04, 0.04) | −0.01 (−0.05, 0.04) | 0.02 (−2.55, 4.22) | 0.06 (−1.67, 6.02) |
| 46-55 years | −0.02 (−0.05, 0.03) | −0.03 (−0.06, 0.03) | −0.02 (−3.89, 2.55) | 0.02 (−3.08, 4.56) |
| 56-65 years | −0.10 (−0.07, 0.00) | −0.03 (−0.05, 0.03) | −0.14 (−6.63, −0.79) | −0.03 (−4.61, 2.83) |
| >65 years | −0.34 (−0.13, −0.07) | −0.09 (−0.07, 0.02) | −0.27 (−9.59, −3.88) | −0.03 (−4.83, 3.22) |
| Current marital status | ||||
| Single | Ref. | Ref. | Ref. | Ref. |
| Cohabiting/married | −0.20 (−3.70, −0.11) | 0.04 (−0.02, 0.04) | −0.24 (−3.79, −0.10) | 0.02 (−1.90, 2.82) |
| Separated/divorced | −0.03 (−7.01, 5.23) | −0.03 (−0.10, 0.03) | −0.06 (−7.07, 5.35) | 0.01 (−5.16, 6.37) |
| Widowed | −0.22 (−9.98, −4.23) | −0.12 (−0.10, −0.02) | −0.23 (−9.95, −4.11) | −0.02 (−4.09, 2.69) |
| Education level | ||||
| Primary school/below | Ref. | Ref. | Ref. | Ref. |
| Secondary school | 0.30 (3.77, 7.16) | 0.13 (0.02, 0.06) | 0.36 (3.79, 7.14) | 0.05 (−0.87, 3.23) |
| College/above | 0.31 (5.12, 8.71) | 0.06 (−0.01, 0.05) | 0.20 (5.11, 8.69) | 0.02 (−2.24, 3.17) |
| Monthly household income | ||||
| ≤20 000 HKD | Ref. | Ref. | Ref. | Ref. |
| 20 001-30 000 HKD | 0.36 (2.13, 6.66) | 0.04 (−0.01, 0.04) | 0.38 (2.11, 6.65) | 0.02 (−1.49, 3.01) |
| 30 001-50 000 HKD | 0.31 (2.15, 6.95) | 0.03 (−0.01, 0.04) | 0.30 (2.05, 6.94) | 0.01 (−2.12, 2.79) |
| >50 000 HKD | 0.39 (3.73, 9.12) | 0.03 (−0.02, 0.05) | 0.38 (3.74, 9.02) | 0.05 (−0.70, 4.74) |
| Chronic diseases | ||||
| No | Ref. | Ref. | Ref. | Ref. |
| Yes | −0.25 (−8.03, −4.95) | −0.16 (−0.07, −0.03) | −0.26 (−8.00, −5.05) | −0.19 (−6.32, −3.13) |
| Social media user | ||||
| No | Ref. | Ref. | Ref. | Ref. |
| Yes | 0.30 (0.79, 1.18) | 0.09 (0.01, 0.05) | 0.22 (3.91, 6.78) | 0.06 (−0.53, 3.31) |
| Daily social media time | 0.01 (−0.04, 0.05) | −0.00 (−0.01, 0.01) | 0.03 (−0.26, 0.58) | −0.01 (−0.55, 0.41) |
| Internet gamer | ||||
| No | Ref. | Ref. | Ref. | Ref. |
| Yes | 0.11 (0.22, 0.75) | 0.01 (−0.02, 0.03) | 0.13 (2.40, 6.22) | 0.06 (−0.16, 4.04) |
| Daily internet game time | 0.03 (−0.07, 0.10) | −0.01 (−0.01, 0.01) | 0.10 (−0.36, 1.70) | 0.04 (−0.42, 1.70) |
| MVPA | ||||
| None at all | Ref. | Ref. | Ref. | Ref. |
| Less than 50 min | 0.05 (−0.01, 0.04) | 0.08 (0.00, 0.04) | 0.07 (−0.02, 3.78) | 0.11 (0.77, 4.39) |
| 50 to 100 min | 0.10 (0.01, 0.06) | 0.08 (0.01, 0.05) | 0.13 (1.67, 6.02) | 0.13 (1.76, 5.93) |
| 100-150 min | 0.07 (0.00, 0.08) | 0.05 (−0.01, 0.07) | 0.11 (2.34, 9.39) | 0.10 (2.03, 8.78) |
| More than 150 min | 0.11 (0.02, 0.07) | 0.08 (0.01, 0.06) | 0.12 (1.57, 6.34) | 0.11 (1.31, 5.94) |
| Loneliness | −0.07 (−0.01, −0.00) | −0.05 (−0.01, 0.00) | −0.13 (−1.15, −0.43) | −0.13 (−1.11, −0.42) |
Abbreviations: HRQoL, health-related quality of life; MVPA, moderate-to-vigorous physical activities; βu, univariable regression; βm, multivariable regression; Ref., reference group; HKD, Hong Kong Dollar.
Bold values indicate P < .05.
Path Analysis
The mediation model of status of social media users and internet gamers showed excellent model fit (χ2/df = 1.23, CFI = 1.00, NNFI = 1.00, RMSEA = 0.02; Figure 1; Table 3). MVPA significantly and positively mediated the relationships between social media users and EQ-5D index (β = 0.01; 95% CI, 0.00 to 0.01) and EQ VAS (β = 0.01; 95% CI, 0.10 to 0.61). Social media users were directly and positively associated with EQ-5D index (β = 0.27, P < .001) and EQ VAS (β = 0.18, P < .001). As the path from social media users to loneliness was nonsignificant, loneliness was not a significant mediator between social media users and HRQoL. Internet gamers were positively associated with loneliness (β = 0.07, P <.05) and EQ VAS (β = 0.08, P <.05). Loneliness was negatively associated with EQ-5D index (β = −0.07, P <.05) and EQ VAS (β = −0.14, P <.001). However, the mediation role of loneliness was not significant (P > .05). Internet gamers were not significantly associated with MVPA or EQ-5D index (P > .05). The total effects of social media users on EQ-5D index (β = 0.28, P < .05) and EQ VAS (β = 0.20, P < .05) were significant, while the total effect of internet gamers on EQ VAS (β = 0.07, P < .05) was significant but not on EQ-5D index (β = 0.02, P > .05). Supplemental Figure S1 shows the model by age. MVPA significantly fully mediated the effect of social media users on EQ VAS in younger adults (β = 0.01; 95% CI, 0.10 to 0.59), and partially mediated its effect on EQ-5D index in older adults (β = 0.02; 95% CI, 0.01 to 0.31). However, internet gamers were only associated with more loneliness and better EQ VAS in younger adults but not older adults.
Figure 1.
Path analysis of association between screen time and HROoL with MVPA and loneliness.
Significant standardized regression coefficients are presented as association; dash line represents effect that was not significant. Significant socio-demographic variables of HROoL were controlled. HRQoL, health-related quality of life; MVPA, moderate-to-vigorous physical activities
*P < .05, **P < .01, ***P < .001.
Table 3.
Associations Among Status of Social Media Users and Internet Games, MVPA, Loneliness, and HRQoL (N = 1070).
| Paths | Unstandardized coefficients | Standardized coefficients | ||
|---|---|---|---|---|
| Estimate | SE | Estimate | P | |
| Internet gamer → MVPA | 0.17 | 0.11 | 0.05 | .12 |
| Internet gamer → Loneliness | 0.39 | 0.17 | 0.07 | .03 |
| Social media user → MVPA | 0.31 | 0.09 | 0.12 | <.001 |
| Social media user → Loneliness | −0.06 | 0.13 | −0.02 | .63 |
| MVPA → EQ-5D | 0.01 | 0.00 | 0.08 | .01 |
| MVPA → EQ VAS | 0.89 | 0.27 | 0.10 | <.01 |
| Loneliness → EQ-5D | −0.01 | 0.00 | −0.07 | <.01 |
| Loneliness → EQ VAS | −0.85 | 0.18 | −0.14 | <.001 |
| Internet gamer → EQ-5D | 0.01 | 0.01 | 0.03 | .41 |
| Internet gamer → EQ VAS | 2.54 | 0.99 | 0.08 | .01 |
| Social media user → EQ-5D | 0.08 | 0.01 | 0.27 | <.001 |
| Social media user → EQ VAS | 4.46 | 0.76 | 0.18 | <.001 |
Abbreviations HRQoL, health-related quality of life; MVPA, moderate to vigorous physical activities
Bold values indicate P < .05.
The mediation model of social media time among social media users showed good model fit (χ2/df = 2.50, CFI = 0.98, NNFI = 0.90, RMSEA = 0.05; Figure 2; Table 4). Loneliness significantly and totally mediated the relationships between social media time and EQ-5D index (β = −0.02; 95% CI, −0.03 to −0.00) and the EQ VAS (β = −0.02; 95% CI, −0.25 to −0.05). MVPA was not a significant mediator between social media time and HRQoL. The total effects of social media time on EQ-5D index (β = 0.02, P > .05) and EQ VAS (β = 0.03, P > .05) were not significant. In subgroup analyses, the results were only significant in younger adults (aged 18-65 years; Supplemental Figure S2). Loneliness significantly and completely mediated the effect of social media time on EQ-5D index only in younger adults (β = −0.02; 95% CI, −0.04 to −0.01). The mediation model of internet gaming time among gamers showed poor model fit and nonsignificant model paths.
Figure 2.
Path analysis of association between social media time and HROoL with MVPA and loneliness.
Significant standardized regression coefficients are presented as association; dash line represents effect that was not significant. Significant socio-demographic variables of HROoL were controlled.
*P < .05, **P < .01, ***P < .001.
Table 4.
Associations Among Social Media Time, MVPA, Loneliness, and HRQoL (n = 658).
| Paths | Unstandardized coefficients | Standardized coefficients | ||
|---|---|---|---|---|
| Estimate | SE | Estimate | P | |
| Social media time → MVPA | −0.05 | 0.03 | −0.08 | .06 |
| Social media time → Loneliness | 0.14 | 0.04 | 0.14 | <.001 |
| MVPA → EQ-5D | 0.00 | 0.00 | 0.06 | .10 |
| MVPA → EQ VAS | 0.96 | 0.31 | 0.12 | <.01 |
| Loneliness → EQ-5D | −0.00 | 0.00 | −0.09 | <.05 |
| Loneliness → EQ VAS | −0.45 | 0.21 | −0.09 | <.05 |
| Social media time → EQ-5D | 0.00 | 0.00 | 0.04 | .32 |
| Social media time → EQ VAS | 0.27 | 0.22 | 0.05 | .21 |
Abbreviations HRQoL, health-related quality of life; MVPA, moderate-to-vigorous physical activities.
Bold values indicate P < .05.
Discussion
The current study represents the first attempt to investigate the relationships among screen time, physical activity, loneliness, and HRQoL in a population-based sample of Chinese adults in Hong Kong. Different roles of internet gamers and social media users in these relationships were found.
Social media users reported more frequent MVPA and better HRQoL than nonusers, with MVPA significantly mediating this positive association. This pattern was robust across both age groups. This finding aligns with ICT theory rather than displacement theory: social media may serve as a source of fitness information, workout inspiration, and social modeling,32,33 and may enhance e-health literacy. 34 Rather than displacing physical activity, social media use in this sample co-occurred with higher activity levels.
While being a user was beneficial, greater daily time on social media showed negative indirect effects on HRQoL through increased loneliness. This pattern was driven entirely by younger adults. This supports the displacement hypothesis: longer screen time may reduce face-to-face interaction, fostering superficial online connections that fail to meet genuine social needs.35,36 The age-specific finding likely reflects generational differences: younger adults often develop and maintain relationships online, whereas older adults typically established meaningful offline relationships before adopting social media. 37 Together, these results move beyond a simplistic “good/bad” narrative, showing that the relationship depends on both the dimension of use (user status vs time) and age.
Internet gamers showed more loneliness but better EQ VAS in younger adults. The results may stem from a complex interplay of factors related to gaming behavior. On the one hand, younger internet gamers may be less likely to participate in real-life activities and social interactions outside of the gaming environment. On the other hand, they may perceive their overall well-being more positively due to factors such as satisfaction with gaming experiences, social connectedness within gaming communities, and the sense of achievement and competence derived from gaming accomplishments, which could reflect on the self-perceived quality of life and health status on the visual analog scale. These results, combined with the nonsignificant effect of time spent on internet games, are different from the findings in children and adolescents. 38 This may suggest that internet games may have different roles in youth and adults. Another explanation is that youth gamers may be more likely to develop internet gaming disorder due to their underdeveloped self-regulation capacity, while adults are less likely to develop addictive symptoms. 20 Understanding multifaceted influences of internet games requires a comprehensive examination of individual motivations, self-management ability, emotional experiences, and social dynamics within the gaming context. Qualitative studies in different age groups may provide valuable insights and comprehensive information on the role of internet games.
The study had several limitations. Due to the cross-sectional design, the direction of these associations cannot be confirmed. For example, individuals with better HRQoL may be more likely to be active on social media. Second, the study is susceptible to recall bias due to the reliance on self-reported data. Objective assessments, such as clinical evaluations or physiological markers, can provide a more comprehensive understanding of the impact of internet gaming and social media use on various dimensions of health. Third, data were collected during the early phase of the COVID-19 pandemic. Screen time behaviors during the pandemic may not generalize to nonpandemic periods. With limited opportunities for outdoor activities and in-person socializing, individuals may have relied more heavily on digital platforms for social connection, entertainment, and information. Thus, the positive associations observed between social media use and HRQoL may be partly attributable to the pandemic context, where screen time served as a substitute for unavailable in-person interactions. In nonpandemic times, the same patterns may differ. In addition, loneliness levels were likely elevated during the study period due to social distancing measures. The finding that greater social media time was associated with higher loneliness may reflect reverse causality (lonely individuals turning to social media) or cyclic effect rather than a purely causal effect of screen time. The generalizability of our findings to postpandemic settings is unknown. Screen time behaviors that became normalized during the pandemic (eg, video conferencing, online social events, remote work) may persist, but the psychological and social context has changed. Future studies should replicate our mediation models in postpandemic cohorts. Fourth, although we controlled for key sociodemographic factors, residual confounding (eg, by underlying personality traits, general health consciousness) could influence the observed associations. Fifth, the exclusion of individuals who solely use mobile phones or who were not at home during the survey period may introduce bias into the results. Younger individuals and heavy screen time users might be underrepresented in the sample. This may limit the generalizability of our findings to these specific subgroups. Sixth, MVPA was measured using ordinal categories rather than continuous minutes and was treated as a continuous variable in path analysis, which assumes equal spacing between categories and linearity of effects. This approach may have reduced statistical sensitivity (eg, true differences between “100-150 minutes” and “>150 minutes” might be missed if the effect is non-linear) and potentially biased estimates. Future studies should use continuous minute-level MVPA data (eg, accelerometry or detailed recall) to validate these findings and detect potential nonlinear thresholds (eg, diminishing returns beyond 150 min/week).
Conclusions and Implications
The findings highlight the need to deeply and comprehensively understand the role of internet games and social media in lifestyles and health-related quality of life in adult populations. Screen time can be a double-edged sword. Social media users had better HRQoL even after controlling for socio-demographic variables. Mediation analyses further revealed that MVPA was a mediating factor in the association. However, prolonged social media time might reduce HRQoL via loneliness in younger adults. Younger internet gamers reported increased loneliness but better HRQoL. These findings offer several practical implications for nurses and other healthcare professionals working with community-dwelling adults. First, nurses should assess not only how much time patients spend on social media and gaming but also whether they use these platforms and how they use them. Distinguishing user status from time spent can identify different risk and benefit profiles. For younger adults, asking about social media time (eg, >2-3 h/day) may help detect those at risk for loneliness. Given that social media users reported higher MVPA, nurses can leverage social media to deliver physical activity interventions, such as by recommending evidence-based fitness channels, online walking groups, or condition-specific exercise content on platforms like YouTube, Facebook, or WhatsApp. This is particularly relevant for older adults who may benefit from digital health resources but need guidance on safe use. Nursing curricula should include digital health literacy training, enabling nurses to guide patients on healthy screen time behaviors. Healthcare organizations should consider developing evidence-based guidelines for screen time in adults that move beyond a one-size-fits-all approach to age- and purpose-specific recommendations.
Supplemental Material
Supplemental material, sj-pdf-1-wjn-10.1177_01939459261467805 for How Screen Time May Affect Health-Related Quality of Life in Community-Dwelling Adults in Hong Kong: The Mediation Roles of Physical Activity and Loneliness by Xue Yang, Chang Wang, Benjamin H. K. Yip and Samuel Y. S. Wong in Western Journal of Nursing Research
ORCID iD: Xue Yang 
https://orcid.org/0000-0001-7892-2994
Ethical Considerations: The study procedures were carried out in accordance with the Declaration of Helsinki. Ethics approval was obtained from the Survey and Behavioral Research Ethics Committee of The Chinese University of Hong Kong (Ref.# SBRE-19-645).
Consent to Participate: All participants provided verbal informed consent.
Authors Contributions: Conceptualization, Methodology, Writing–Original draft: XY; Writing – Review & Editing: SW, BY; Formal analysis: XY, CW.
Funding: The authors received no financial support for the research, authorship, and/or publication of this article.
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Supplemental Material: Supplemental material for this article is available online.
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author.
References
- 1. Guo N, Luk TT, Wang MP, et al. Self-reported screen time on social networking sites associated with problematic smartphone use in Chinese adults: a population-based study. Front Psychiatry. 2021;11:614061. doi: 10.3389/fpsyt.2020.614061 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Li X, Liu Q. Social media use, eHealth literacy, disease knowledge, and preventive behaviors in the COVID-19 pandemic: cross-sectional study on Chinese netizens. J Med Internet Res. 2020;22(10):e19684. doi: 10.2196/19684 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Liao Y, Shibata A, Ishii K, Koohsari MJ, Oka K. Cross-sectional and prospective associations of neighbourhood environmental attributes with screen time in Japanese middle-aged and older adults. BMJ Open. 2018;8(3):e019608. doi: 10.1136/bmjopen-2017-019608 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Sanders JL, Williams RJ, Damgaard M. Video game play and internet gaming disorder among Canadian adults: A national survey. Can J Addict. 2017;8(2):6-12. doi: 10.1097/CXA.0000000000000006 [DOI] [Google Scholar]
- 5. Exelmans L, Van den Bulck J. Sleep quality is negatively related to video gaming volume in adults. J Sleep Res. 2015;24(2):189-196. doi: 10.1111/jsr.12255 [DOI] [PubMed] [Google Scholar]
- 6. Torrance GW. Utility approach to measuring health-related quality of life. J Chronic Dis. 1987;40(6):593-603. doi: 10.1016/0021-9681(87)90019-1 [DOI] [PubMed] [Google Scholar]
- 7. Kaplan RM, Hays RD. Health-related quality of life measurement in public health. Annu Rev Public Health. 2022;43(1):355-373. doi: 10.1146/annurev-publhealth-052120-012811 [DOI] [PubMed] [Google Scholar]
- 8. Karimi M, Brazier J. Health, health-related quality of life, and quality of life: what is the difference? Pharmacoeconomics. 2016;34(7):645-649. doi: 10.1007/s40273-016-0389-9 [DOI] [PubMed] [Google Scholar]
- 9. Lacy KE, Allender SE, Kremer PJ, et al. Screen time and physical activity behaviours are associated with health-related quality of life in Australian adolescents. Qual Life Res. 2012;21(6):1085-1099. doi: 10.1007/s11136-011-0014-5 [DOI] [PubMed] [Google Scholar]
- 10. Motamed-Gorji N, Qorbani M, Nikkho F, et al. Association of screen time and physical activity with health-related quality of life in Iranian children and adolescents. Health Qual Life Outcomes. 2019;17(1):2. doi: 10.1186/s12955-018-1071-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Davies CA, Vandelanotte C, Duncan MJ, van Uffelen JG. Associations of physical activity and screen-time on health related quality of life in adults. Prev Med. 2012;55(1):46-49. doi: 10.1016/j.ypmed.2012.05.003 [DOI] [PubMed] [Google Scholar]
- 12. Barr-Anderson DJ, Hazzard VM, Hahn SL, Folk AL, Wagner BE, Neumark-Sztainer D. Stay-at-home orders during COVID-19: The influence on physical activity and recreational screen time change among diverse emerging adults and future implications for health promotion and the prevention of widening health disparities. Int J Environ Res Public Health. 2021;18(24):13228. doi: 10.3390/ijerph182413228 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Bennasar-Veny M, Yañez AM, Pericas J, et al. Cluster analysis of health-related lifestyles in university students. Int J Environ Res Public Health. 2020;17(5):1776. doi: 10.3390/ijerph17051776 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Runacres A, Mackintosh KA, Knight RL, et al. Impact of the COVID-19 pandemic on sedentary time and behaviour in children and adults: a systematic review and meta-analysis. Int J Environ Res Public Health. 2021;18(21):11286. doi: 10.3390/ijerph182111286 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Bize R, Johnson JA, Plotnikoff RC. Physical activity level and health-related quality of life in the general adult population: a systematic review. Prev Med. 2007;45(6):401-415. doi: 10.1016/j.ypmed.2007.07.017 [DOI] [PubMed] [Google Scholar]
- 16. Qin F, Song Y, Nassis GP, et al. Physical activity, screen time, and emotional well-being during the 2019 novel coronavirus outbreak in China. Int J Environ Res Public Health. 2020;17(14):5170. doi: 10.3390/ijerph17145170 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Tan SS, Fierloos IN, Zhang X, et al. The Association between Loneliness and Health Related Quality of Life (HR-QoL) among Community-Dwelling Older Citizens. Int J Environ Res Public Health. 2020;17(2):600. doi: 10.3390/ijerph17020600 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Park C, Majeed A, Gill H, et al. The Effect of Loneliness on Distinct Health Outcomes: A Comprehensive Review and Meta-Analysis. Psychiatry Res. 2020;294:113514. doi: 10.1016/j.psychres.2020.113514 [DOI] [PubMed] [Google Scholar]
- 19. MacDonald KB, Patte KA, Leatherdale ST, Schermer JA. Loneliness and screen time usage over a year. J Adolesc. 2022;94(3):318-332. doi: 10.1002/jad.12024 [DOI] [PubMed] [Google Scholar]
- 20. Yang X, Yip BHK, Lee EKP, Zhang D, Wong SYS. The relationship between technology use and problem technology use and potential psychosocial mechanisms: population-based telephone survey in community adults during COVID-19. Front Psychol. 2021;12:696271. doi: 10.3389/fpsyg.2021.696271 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Neuman SB. The displacement effect: Assessing the relation between television viewing and reading performance. Read Res Q. 1988;23(4):414-440. doi: 10.2307/747641 [DOI] [Google Scholar]
- 22. Mutz DC, Roberts DF, van Vuuren DP. Reconsidering the displacement hypothesis: Television’s influence on children’s time use. Communic Res. 1993;20(1):51-75. doi: 10.1177/009365093020001003 [DOI] [Google Scholar]
- 23. Casanova G, Zaccaria D, Rolandi E, Guaita A. The effect of information and communication technology and social networking site use on older people’s well-being in relation to loneliness: review of experimental studies. J Med Internet Res. 2021;23(3):e23588. doi: 10.2196/23588 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Christensen MA, Bettencourt L, Kaye L, et al. Direct measurements of smartphone screen-time: relationships with demographics and sleep. PloS One. 2016;11(11):e0165331. doi: 10.1371/journal.pone.0165331 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Ping W, Zheng J, Niu X, et al. Evaluation of health-related quality of life using EQ-5D in China during the COVID-19 pandemic. PLoS One. 2020;15(6):e0234850. doi: 10.1371/journal.pone.0234850 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. World Health Organization. WHO Guidelines on Physical Activity and Sedentary behaviour. World Health Organization; 2020. [PubMed] [Google Scholar]
- 27. Qi M, Li P, Moyle W, Weeks B, Jones C. Physical activity, health-related quality of life, and stress among the Chinese adult population during the COVID-19 pandemic. Int J Environ Res Public Health. 2020;17(18):6494. doi: 10.3390/ijerph17186494 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Leung GT, de Jong Gierveld J, Lam LC. Validation of the Chinese translation of the 6-item De Jong Gierveld Loneliness Scale in elderly Chinese. Int Psychogeriatr. 2008;20(6):1262-1272. doi: 10.1017/s1041610208007552 [DOI] [PubMed] [Google Scholar]
- 29. EuroQoL Research Foundation. EQ-5D-5L User Guide, 2019. Accessed December 1, 2025. https://euroqol.org/publications/user-guides
- 30. Hooper D, Coughlan J, Mullen MR. Structural equation modelling: guidelines for determining model fit. Electronic J Bus Res Methods. 2008;6(1):53-60. [Google Scholar]
- 31. Kline RB. Principles and Practice of Structural Equation Modeling. Guilford publications; 2023. [Google Scholar]
- 32. Liu J, Wang C, Meng Z, Yu C. Digital interaction and active aging: the impact of social media use on physical activity behavior mediated by social capital and self-efficacy. Front Psychol. 2025;16:1580936. doi: 10.3389/fpsyg.2025.1580936 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Zhang N, Ding K, Park C, Vo J, Cantos KM. The effect of social support on social media on Asian college students’ intention to participate in physical activity in the United States. Health Equity. 2023;7(1):735-742. doi: 10.1089/heq.2022.0208 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Lissak G. Adverse physiological and psychological effects of screen time on children and adolescents: literature review and case study. Environ Res. 2018;164:149-157. doi: 10.1016/j.envres.2018.01.015 [DOI] [PubMed] [Google Scholar]
- 35. Subrahmanyam K, Frison E, Michikyan M. The relation between face-to-face and digital interactions and self-esteem: a daily diary study. Hum Behav & Emerg Tech. 2020;2(2):116-127. doi: 10.1002/hbe2.187 [DOI] [Google Scholar]
- 36. Twenge JM, Spitzberg BH, Campbell WK. Less in-person social interaction with peers among US adolescents in the 21st century and links to loneliness. J Soc Pers Relat. 2019;36(6):1892-1913. doi: 10.1177/0265407519836170 [DOI] [Google Scholar]
- 37. Benvenuti M, Giovagnoli S, Mazzoni E, Cipresso P, Pedroli E, Riva G. The relevance of online social relationships among the elderly: how using the web could enhance quality of life? Front Psychol. 2020;11:551862. doi: 10.3389/fpsyg.2020.551862 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Li Q, Chan KK-y, Ramos MR, Wang X, Wong SY-s, Yang X. Non-linear relationships and interaction effects between different types of screen time and mental health issues among Hong Kong adolescents. Psychiatry Res Commun. 2025;5(1):100205. doi: 10.1016/j.psycom.2025.100205 [DOI] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental material, sj-pdf-1-wjn-10.1177_01939459261467805 for How Screen Time May Affect Health-Related Quality of Life in Community-Dwelling Adults in Hong Kong: The Mediation Roles of Physical Activity and Loneliness by Xue Yang, Chang Wang, Benjamin H. K. Yip and Samuel Y. S. Wong in Western Journal of Nursing Research
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author.