This study utilized multi-stage stratified sampling. The Atwima Kwanwoma District was purposefully selected based on anecdotal evidence suggesting many single parents. The district was informally divided into Atwima and Kwanwoma and further categorized into urban and rural communities based on socioeconomic characteristics and development factors such as economic activities, road conditions, and health and educational infrastructure (11). A simple random sampling method was employed to select 12 communities (i.e., four urban and eight rural communities) from 64 communities in the district (19). The names of all communities were written on slips of paper, acknowledging the informal division of the district into Atwima and Kwanwoma, which is typically split into East and West.

Using a fishbowl selection method, where names of all communities were placed in a bowl and drawn randomly (20), six communities (four rural and two urban) were chosen from Atwima, and another six communities (four rural and two urban) were selected from Kwanwoma, making a total of 12 communities (four urban and eight rural). The larger number of rural communities (8) compared to urban ones (4) was due to the higher population in urban areas. A default prevalence rate of 50% (20), equivalent to 0.5, was used to estimate the number of single parents in the district, as specific data was unavailable. Using this prevalence rate, Fisher’s sample size estimation formula was applied to determine the overall sample size for the study (21).

The minimum sample size was calculated with the following parameters: a 95% confidence interval (∝) of 1.96, a prevalence (P) of 50% or 0.5 (since the proportion of single parents aged ≥13 years is unknown), a margin of error (d) of 5% or 0.05, and considering 5% type 1 error and 15% type 2 error. Adjustments were made to account for anticipated non-responses and outliers, aiming to enhance the generalizability and robustness of the findings (22). Consequently, the final sample size was set at N = 627. The researchers developed a face-to-face interviewer-administered questionnaire, and five research assistants were trained to administer the questionnaire. The questionnaire is part of a broader household survey covering various topics like sleep problems, depression, anxiety, financial concerns, neighborhood attachment, self-rated health and happiness, social networks, and functional limitations. Data collection occurred over 3 months, from June to August 2023. Ethical approval was obtained from relevant authorities, and written informed consent was secured from all respondents.

Though being a purely quantitative study, a convenience sampling technique was used in reaching out to the respondents. This became necessary when we observed, during the pretesting of our research instruments, that not all households in the selected communities were single-parent households. Moreover, we met some single parents at their workplaces. Therefore, using a more appropriate sampling technique, like systematic sampling, was not feasible.

The study was carried out in accordance with the World Medical Association’s Declaration of Helsinki’s code of ethics for research involving human subjects. The study protocol was reviewed and approved by the Humanities and Social Sciences Research Ethics Committee (HuSSREC) at Kwame Nkrumah University of Science and Technology in Kumasi, Ghana (Ref number: HuSSREC/AP/111/VOL.1.). Additionally, written informed consent was obtained from each respondent. Respondents were assured of confidentiality and anonymity of the information they provide. They were also assured of their right to stop or refuse to respond to any question they were not comfortable with, without being penalized. Indeed, we sought permission and approval from the parents or guardians of a few minors who had less than 18 years during data collection. Moreover, in the Ghanaian context, one is considered an adult when he or she gives birth and assumes the responsibility of childcare.

The statistical analysis was conducted using SPSS v25 (SPSS, Inc., IBM, Armonk, NY, United States). The statistical significance was determined as the p-value of <0.05. Sample characteristics are first displayed as stratified by the meeting of WHO PA guidelines (<600 MET min; ≥600 MET min) and HRQoL (low and high). The difference in sample characteristics was tested using Student’s t-tests and Chi-squared tests for continuous and categorical variables. In the second step, hierarchical OLS regression models were conducted to examine the link between meeting PA guidelines and HRQoL. We further performed interaction analysis (i.e., Sex × PA, residence × PA, and age × PA) to determine the moderation role of age, sex, and residence in the association of PA and HRQoL.

Model 1 was the unadjusted model. Model 2 included Model 1 and adjusted for age and sex. Model 3 included Model 2 and further adjusted for alcohol, residence, employment, household size, living arrangement, NHIS, education, social network, age, sex, sleep quality, depression, financial hardship, economic status, chronic conditions, and anxiety. Model 4 included Model 3 and the interaction term, PA × sex. Model 5 included Model 3 and the interaction term, PA × age. Model 6 included Model 3 and the interaction term, PA × residential status. Third, the OLS regression estimating the association between PA and the various components of HRQoL was conducted. Each component of the HRQoL variable was treated as separate dependent variables as they were individually regressed on PA and other covariates (independent variables). Lastly, OLS regression models estimating the association between meeting PA guidelines stratified by age and sex groups were also conducted. Stratification by residence was not included because the interaction analysis was not statistically significant.

A total of 627 individuals were included in the current study. The characteristics of this analytic sample are shown in Table 1. The average age was 45.0 (SD = 14.7) years, ranging from 14 to 85 years, and 67.3% were females. The majority of the sample lived in rural communities (54.5%), were gainfully employed (71.8%), and neither smoked (97.4%) or consumed alcohol (92.5%). On average, every participant had schooled for 9.5 (SD = 4.0) years while the mean monthly income was approximately 1938.28 (SD = 904.46) Ghana Cedis (US$ 140). Moreover, the mean physical activity, social network, and number of chronic physical conditions scores were 330.79 (SD = 420.83), 8.32 (SD = 6.44), and 1.07 (SD = 1.35), respectively. A Pearson’s Chi-square test and Student t-test indicated that individuals who did not meet the WHO PA guidelines with <600 MET min of PA significantly had much worse scores for functional limitations (F = 626.60, p < 0.001), chronic physical conditions (F = 49.31, p < 0.001), sleep problems (F = 10.21, p < 0.01), anxiety (F = 8.113, p < 0.005), and depression (F = 8.02, p < 0.005) compared to those who met the WHO PA guidelines with ≥600 MET min of PA. Similarly, those who had low HRQoL had worse scores for functional limitations (F = 1033.23, p < 0.001), chronic conditions (F = 40.88, p < 0.001), sleep problems (F = 10.88, p < 0.001), depression (F = 0.12, p < 0.005) and anxiety (F = 0.11, p < 0.005) compared to those who had high HRQoL. Additionally, the Chi-square results showed that there was a significant association at a 5% significance level between sex and PA (χ² = 4.076, p < 0.05), sex and HRQoL, (χ² = 11.364, p < 0.001), residence and PA (χ² = 4.708, p < 0.05). In addition, the independent t-test results revealed a statistically significant association between age and PA (F = 0.66, p < 0.001), age, and HRQoL (F = 0.031, p < 0.001).

The results of the associations between meeting WHO PA guidelines and HRQoL, estimated by hierarchical OLS regression models for the pooled sample, are detailed in Table 2. The crude model (Model 1) revealed a significant negative association between PA and low HRQoL (B = −1.57, SE = 0.14, p < 0.001). After adjusting for age and sex in Model 2, the effect size decreased, indicating a weaker negative association (B = −0.505, SE = 0.148, p < 0.001). In Model 3, which adjusted for socio-demographic, socio-economic, and health-related factors, the significant negative association persisted, though the effect size was further reduced (B = −0.298, SE = 0.132, p < 0.05). Models 4 and 5 demonstrated significant moderating effects of sex (B = −0.619, SE = 0.206, p < 0.01) and age (B = −0.062, SE = 0.008, p < 0.001) on the association between PA and low HRQoL. However, the moderating role of residence in Model 6 was not significant. Further, the OLS regression analysis of PA’s impact on various HRQoL components (Table 3) showed substantial negative associations with pain/discomfort (B = −0.143, SE = 0.053, p < 0.01), usual activities (B = −0.311, SE = 0.040, p < 0.001), mobility (B = −0.094, SE = 0.047, p < 0.05), and self-care (B = −0.109, SE = 0.028, p < 0.001). Additionally, sex and age-stratified estimations of the association between PA and HRQoL (Table 4) revealed significant differences: males (B = −0.967, SE = 0.188, p < 0.001) and females (B = −1.330, SE = 0.147, p < 0.001), as well as between age groups (14–45 years: B = −0.395, SE = 0.193, p < 0.05) and (46–90 years: B = −1.018, SE = 0.177, p < 0.001). The analysis indicated that the negative association between PA and HRQoL was more pronounced among older adults and females.

Physical activity was significantly associated with decreased poor HRQoL, even after accounting for potential confounders. Additional analysis revealed a substantial negative association between PA and specific components of HRQoL, including pain/discomfort, usual activities, mobility, and self-care. The interaction analysis revealed a modifying effect of sex differences and age differences in the association of PA with poor HRQoL. Thus, the age and sex-stratified analysis showed that the association between PA and poor HRQoL was more pronounced among older adults and females. To our knowledge, this is the first study to provide insights into the link between PA and HRQoL among single parents in a LMIC context. This adds to our current knowledge mainly based on studies using data from high-income countries USA, China, Japan etc. Moreover, public health policies may leverage the findings of this study to create targeted health promotion programs such as community-based PA initiatives (e.g., keep fit clubs), family-oriented activities and integration of PA into health education.

The results indicate a negative association between adherence to WHO PA guidelines and poor HRQoL among single parents in Ghana. Regular PA has been consistently linked to enhanced physical health, psychological well-being, and overall quality of life (3, 22, 26). Engaging in recommended PA levels appears to contribute positively to various aspects of HRQoL, including better physical functioning, reduced stress levels, and improved emotional resilience. This is consistent with previous studies investigating PA’s association with HRQoL. For instance, a study in China by Hao et al. (3) revealed PA was positively associated with HRQoL among adults. Similarly, in a large population of Tehranian adults, Jalali-Farahani et al. (26) found levels of PA were significantly associated with most subscales of HRQoL in both men (p < 0.05) and women (p < 0.01).

The association between PA and HRQoL may be explained by a range of physiological, psychological, and socio-economic mechanisms (3). Physiologically, regular PA strengthens the heart, enhances blood circulation, and lowers the risk of heart disease (10). PA may improve cardiovascular health by enhancing heart and lung function, reducing blood pressure, improving blood lipid profiles and thereby enhancing health-related quality of life (6). Regular PA may also help regulate blood sugar levels and improve insulin sensitivity, reducing the risk of type 2 diabetes and metabolic syndrome (8). Furthermore, PA builds muscle strength and endurance, making everyday activities easier, leading to high HRQoL. Additionally, regular PA aids in maintaining a healthy weight, which is associated with a reduced risk of chronic physical conditions such as diabetes and hypertension (9).

The link between regular PA and psychological health are equally compelling. PA may reduce symptoms of anxiety and depression through the release of endorphins, commonly known as “feel-good” hormones (7, 10). PA serves as a natural way to relieve stress, aiding individuals in managing daily challenges more effectively (7). Regular physical activity may mitigate the adverse effects of stress and improve mental health outcomes, which are particularly relevant for single parents who often face elevated levels of stress due to economic pressures, child-rearing responsibilities, and social isolation (12, 26, 27). Cognitively, PA may boost memory and learning abilities, which are essential for sustaining a good quality of life (10). This brain boost is partly due to its influence on mood-regulating neurotransmitters like serotonin and dopamine (25). Furthermore, participating in group fitness classes or team sports fosters social interactions, build friendships and support networks which provide a sense of belonging (26). That sense of belonging and connection is invaluable to HRQoL (26, 27).

The moderation analysis revealed that age and sex variations emerge as critical moderators that shape the magnitude and manifestation of the association between PA and HRQoL. The negative interaction term of age variations on the association of PA with poor HRQoL shows that the association was much stronger among younger single parents (14–45 years) compared with those aged (46 and above). Thus, as one ages, the effect of PA on HRQoL decreases. Age plays a crucial moderating role in the relationship between PA adherence and HRQoL. Younger single parents, typically more active and energetic, may experience greater benefits in terms of improved mood, energy levels, and stress reduction from regular physical activity, thereby leading to higher HRQoL (3). Moreover, age-related declines in muscle mass (sarcopenia), bone density (osteoporosis), cardiovascular health and the fear of injury can make certain activities more difficult or even dangerous (28). Beyond the physical, older adults might grapple with the loss of peers, chronic health conditions impacting their mental well-being, or even societal ageism (stereotypes), all of which can affect their willingness to embrace an active lifestyle (3). This aligns with existing studies. For example, Bélanger et al. (28) suggest that PA adherence is greater for young adults (16–24 years: men = 27%; women = 21%) than older adults (65+ years: 12%; 11%). Etxeberria et al. (13) also found that the oldest old had a poorer HRQoL in both physical and mental dimensions. However, an interaction analysis from Hao et al. (3) indicated that the relationship between PA and HRQoL was significantly different across young, middle-aged, and older Chinese adults (p < 0.05), revealing that older adults with the sufficient PA (coefficient = 0.090, 95%CI: [0.081, 0.100]) and active PA (coefficient = 0.057, 95%CI: [0.043, 0.072]) had significantly higher HRQoL compared with young and middle-aged groups.

The interaction analysis also revealed a negative modifying effect of sex differences in the association of PA and low HRQoL. The results suggest that PA has a differential association with HRQoL based on sex. The current study reveals that while physical activity positively influences HRQoL in both sexes, the magnitude of this benefit is higher in males. Empirical studies show that males and females have different physiological responses to physical activity, which might influence how physical activity affects their HRQoL (15). Biologically, women undergo hormonal changes throughout their lives. For women, hormonal fluctuations throughout their lifespan – menstruation, pregnancy, and menopause – can significantly influence energy levels, mood, and physical performance, thus impacting their perception of PA benefits (4). Furthermore, differences in body composition – women generally having a higher percentage of body fat and lower muscle mass than men – can influence which types of activity they find most beneficial or enjoyable (14). Again, cultural and societal expectations regarding physical activity may differ between sexes as males might be more encouraged or expected to engage in physical activity and thus might derive greater psychological and social benefits from such engagement (7). Furthermore, the type and intensity of physical activity that males and females typically engage in may differ, potentially leading to different impacts on HRQoL. For instance, males might participate more in high-intensity or team sports, which could have stronger effects on improving HRQoL (14). On the other hand, Ji et al. (15) suggest that women, compared with men, derived greater gains in all-cause and cardiovascular mortality risk reduction from equivalent doses of leisure-time PA.

The findings from this study, pending future longitudinal research, highlight the need for policies and interventions that promote the development of age and sex-specific PA programs. For example, younger single parents may benefit from high-intensity and time-efficient exercises, while older single parents might need low-impact and flexibility-focused activities like stretching, dancing and swimming. Programs may also consider the different time constraints, safety concerns, and social support systems needed for both males and females. Economic barriers may hinder single parents from engaging in regular PA. Policies by state and non-governmental agencies that provide subsidies or financial incentives for gym memberships, fitness classes, or sports participation may enhance access and adherence to PA. Again, health practitioners may promote PA options that may be easily integrated into the daily routines of single parents. Recommendations could include walking, home-based exercises, or family-oriented PA. Longitudinal research may provide insights into the long-term effects of meeting WHO PA guidelines on the HRQoL of single parents, highlighting the sustainability and effectiveness of various interventions.

This study is among the few to investigate the relationship between PA and HRQoL, and it is the first to explore how age and sex differences moderate this association, specifically among single parents in Ghana. Given the limited information on PA among single parents in Ghana, the findings from this study could serve as a valuable baseline for future research. The key variables in the study were assessed using validated scales known for their robust reliability and content validity. Additionally, various potential confounders were controlled to enhance the validity of the results. Despite these strengths, the study has limitations. The cross-sectional design of the research limits the ability to establish causal relationships. Therefore, future studies may use a longitudinal design to draw causal conclusions. The reliance on retrospective self-reporting in the questionnaire may introduce recall and reporting biases. Again, future studies may consider clinical assessments of HRQoL to strengthen the validity of the findings. Nonetheless, self-reporting remains a practical method for capturing respondents’ subjective evaluations of their HRQoL.