The Effect of Health Literacy on Health-Related Quality of Life Among Saudi Women With Chronic Diseases

ABSTRACT Background Women have unique health concerns in the context of chronic illness. Although health literacy has been linked to health-related quality of life, the pathway by which some factors mediate this relationship has not been well examined among Saudi women with chronic disease. Purpose This study was designed to examine the relationship between health literacy and health-related quality of life and determine the mediating roles of healthy lifestyle and patient-centered care in explaining this relationship. Methods A cross-sectional study was used. A random sample of 20 primary healthcare centers in Saudi Arabia was selected. A convenience sample of 300 women diagnosed with a chronic illness was recruited from these centers. Structural equation modeling was used to test the proposed relationships. Results Controlling for women's age and education, the hypothesized model adequately fit the data. Health literacy and its dimensions were found to positively and directly affect both patient-centered care (β = 0.614, B = 0.189, p < .001) and healthy lifestyle behaviors (β = 0.550, B = 1.211, p < .001). Furthermore, both patient-centered care (β = 0.390, B = 1.301, p < .001) and healthy lifestyle behaviors (β = 0.322, B = 0.150, p < .001) were found to positively and directly affect health-related quality of life, whereas health literacy had significant, direct (β = 0.285, B = 0.292, p < .001) and indirect (β = 0.417, B = 0.428, p < .001) effects on health-related quality of life through patient-centered care and healthy lifestyle behaviors. Conclusions/Implications for Practice Special consideration should be given to health literacy, patient-centered care, and healthy lifestyle behaviors to enhance health-related quality of life in women with chronic illness. Gender-specific health services and interventions are needed to improve women's health literacy.


Introduction
Chronic noncommunicable diseases (NCDs) are long-duration diseases marked by a slow progression (Sidossis & Kales, 2022). NCDs account for 71% of overall total deaths world-wide (World Health Organization [WHO], 2021). More prevalent among women than men (Quiñones et al., 2019), two thirds of all deaths and disabilities among women have been attributed to NCDs (Bonita & Beaglehole, 2014). Women's biological makeup and roles in society make their health concerns unique, increase their vulnerability, and shape their experience with NCDs (WHO, 2019). Feminine characteristics have been found to be associated with a higher rate of NCDs (Ballering et al., 2020). The health-related quality of life (HRQOL) of women is also substantially lower than that of men in the context of NCDs (Hajian-Tilaki et al., 2017).
Several studies have reported that health literacy, defined as the skills that influence an individual's capability to obtain, comprehend, appraise, and use health information to achieve well-being (Nutbeam, 2008), has been linked to HRQOL (Ehmann et al., 2020). However, some studies have found no significant relationship between health literacy and HRQOL in patients with an NCD (Couture et al., 2017). A systematic review and meta-analysis revealed that claims of a moderate relationship between health literacy and HRQOL require further supporting evidence (Zheng et al., 2018). Nutbeam (2008) highlighted three types of literacy, including functional (basic skills for reading and writing health information), interactive (advanced skills for extracting and applying health information), and critical (more advanced skills for analyzing and utilizing health information to control life events). Most prior studies have approached health literacy from the unidimensional aspect (Couture et al., 2017;Ehmann et al., 2020), that is, functional health literacy (Lastrucci et al., 2019). Although the concept health literacy may embody negative connotations such as victim blaming (Frosch et al., 2014), addressing health literacy can help individuals overcome key barriers to achieving well-being (Rowlands et al., 2017) and is one of the determinants of reaching equity in health and empowering individuals to control their health, search for reliable information, and engage in health decision making (Tufts et al., 2009). In different countries, limited health literacy is a prevalent issue, with studies finding limited/low health literacy levels ranging from 25% up to 70% (Lastrucci et al., 2019;Rajah et al., 2019;L. Zhang et al., 2020).
NCDs are also known as lifestyle diseases, which highlights the critical importance of considering lifestyle behavior, defined as the totality of an individual's life habits that influence all aspects of the health prognosis for patients (Sidossis & Kales, 2022). As skills related to obtaining and applying health-related information may help individuals adopt healthy behaviors, health literacy impacts individuals' healthy lifestyle characteristics and behaviors (Uemura et al., 2021). Inadequate health literacy may promote unhealthy choices and increase risky behaviors. Thus, health literacy mediates the relationship between an individual's socioeconomic status and their lifestyle behaviors (Adams et al., 2013). However, the evidence showing the mediating role of healthy lifestyle behaviors in the relationship between health literacy and health outcomes is unclear. Support from healthcare providers, which includes the provision of patient-centered care, is also essential in reducing health burdens. Patient-centered care aligns with a patient's desires, values, and needs and includes the patient in related discussions and decision making (Mead & Bower, 2000). Furthermore, support from healthcare providers in terms of responding to a patient's requests and preferences has a strong effect on health outcomes (Y. C. Chen et al., 2018). Health literacy may promote the ability of a patient to participate actively in care decision making and to engage with healthcare providers in conversation (Parnell et al., 2019). Patient-centered care improves health status (Stewart et al., 2000). The main components of the patient-centered care model are exploring the disease and illness experience, finding common ground, and understanding the whole person (Scambler & Asimakopoulou, 2014). Although this model refers mainly to the medical field with a focus on the physician-patient relationship, these components also align with nursing care (Kitson et al., 2012). Previous studies have examined the impact of patient-centered care on health outcomes separately from health literacy (Altin & Stock, 2016). However, the role of patient-centered care in the relationship between health literacy and patient health outcomes remains unclear. Moreover, patient-centered care has not been explicitly explored in women, although gender differences in terms of patient-provider communication have been identified (Okunrintemi et al., 2018). Prevention, management, and therapeutic approaches for care of patients with NCDs must be gender specific (Oorschot et al., 2021).
Theoretical frameworks have been developed to explain how health literacy affects individual health factors, such as disease and self-care knowledge, health risk behavior, and provider-patient interaction, as mediating factors (S. D. Lee et al., 2004;Nutbeam, 2000;Paasche-orlow & Wolf, 2007).
On the basis of the existing literature and theoretical frameworks, a conceptual model was hypothesized for this study (Figure 1). In this model, health literacy and its dimensions exert direct and positive effects on healthy lifestyle behaviors, patient-centered care, and quality of life. Moreover, healthy lifestyle behaviors and patient-centered care also exert direct and positive effects on quality of life. Thus, the relationship between health literacy and quality of life is mediated through healthy lifestyle behaviors and patient-centered care. This study was developed to examine the relationship between health literacy and HRQOL and to determine the mediating roles of healthy lifestyle and patient-centered care on this relationship among women in Saudi Arabia.

Design, Samples, and Settings
A cross-sectional design was adopted in this study, and a convenience sample of 300 Saudi women diagnosed with any NCD was recruited. The sample size was based on the results of structural equation modeling (SEM) for statistical analysis. According to Kline (2015), a minimum sample size of 200 is needed for using SEM. The participants were recruited from 20 randomly selected primary healthcare centers (PHCs) in two cities in Saudi Arabia. All of the selected PHCs were publicly funded and selected using a web-based program (https://www.random.org/). Patients with NCDs routinely visit these centers every 3-6 months. On average, 15 participants were recruited from each PHC based on the inclusion criteria of being 18-65 years old and having at least one NCD. Women were eligible if they were informed by a doctor that they had arthritis, heart disease, high blood pressure, epilepsy, obesity, stroke, cancer, diabetes, or a chronic respiratory disease (e.g., chronic obstructive pulmonary disease, asthma) and were a regular patient at the PHC. Exclusion criteria included being pregnant, not being a Saudi citizen, and not attending regular follow-ups. After being confirmed for eligibility, each woman was invited to participate in the study. Study data were collected between June 2020 and February 2021 using face-to-face, structured interviews in a private room in the PHC. WHO and Global Health Cluster colleagues have developed standard operating procedures for data collection during the COVID-19 pandemic (WHO, 2020). These protocols were followed along with Saudi Arabia's precautionary and preventive COVID-19 measures, including social distancing (maintaining at least a 1-meter distance with others), wearing masks, checking temperature every morning, and self-quarantining for 14 days after coming in contact with anyone with a confirmed or suspected case of COVID-19.

Measurements
The demographic characteristics of the participants were assessed using survey questions. Comorbidity, defined as the presence of more than two NCDs, was also assessed using disease count. The participants were also asked about their use of acute care settings (hospital admissions and emergency department visits) during the past 12 months.
The instruments used to assess the study variables were in Arabic. Three instruments without prior Arabic versions were translated from their original English using the integrated method for the adaptation and translation of measures (Sidani et al., 2010), which included an assessment of conceptual equivalence for content validity (comprehension and cultural relevance) conducted by five bilingual and bicultural experts. Some adaptations were made based on the content validity index and discussions with experts. Next, the forward translation was conducted by two bilingual translators with medical backgrounds. Pilot testing of the set of items was conducted on 30 patients with an NCD to test the reliability and assess participant understanding of the instruments.

Health literacy
The All Aspects of Health Literacy Scale (AAHLS; Chinn & McCarthy, 2013) was used to measure health literacy based on Nutbeam's framework, which defines health literacy in terms of functional, communicative, and critical health literacy (Nutbeam, 2008). The AAHLS is a 14-item self-reported measure (four functional items, three communicative items, four critical items, and three empowerment items). The first 11 items are rated on a 3-point scale, and the last two items have dichotomous choices. The first three items are reverse coded. The scale is scored by summing the items, with higher scores associated with better health literacy. This instrument is quick and easy to use in primary healthcare settings. Internal consistency was tested for the original scale, with Cronbach's α reliability ranging between .82 and .42 for the total scale and subscales (Chinn & McCarthy, 2013).When used with college students with NCDs, Cronbach's α reliability ranged from .52 to .83 (Barsell et al., 2018). In this study, on that basis of the content validity assessment and adaptation process, one of the empowerment items was removed, as reviewers suggested that the item was not appropriate to the context. Thus, only 13 items were included in the final Arabic version. Pilot testing was conducted, and then the psychometric analysis was performed to assess the scale's construct validity and internal consistency.

Healthy lifestyle behaviors
The Healthy Lifestyle and Personal Control Questionnaire (HLPCQ) is a 26-item instrument used to assess the frequency of embracing positive lifestyle habits, which is indicative of health empowerment (Darviri et al., 2014). This scale has five dimensions: healthy dietary choices (seven items), dietary harm avoidance (four items), daily routine (eight items), organized physical exercise (two items), and social and mental balance (five items). The items of each subscale are rated on a 4-point Likert scale (1 = never or rarely, 2 = sometimes, 3 = often, and 4 = always), indicating the frequency of adopting healthy lifestyle habits. The questionnaire score is calculated by summing up all items. The validity and reliability of the scale were previously tested in the Greek language, which showed satisfactory results of between .63 and .75 (Darviri et al., 2014). In this study, the scale was translated and adapted and then pilot tested. Psychometric analysis was performed to assess the scale's construct validity and internal consistency.

Patient-centered care
The Patient Perception of Patient-Centeredness (PPPC) questionnaire is a 14-item instrument used to assess patient perceptions of their physician's patient centeredness (Stewart et al., 2000). Items are rated on a 4-point Likert scale. The scale has three subscales: exploring both the disease and illness experience (four items), finding common ground (nine items), and understanding the whole person (one item). The scores for the total scale and subscales are obtained by calculating the mean. The scale was previously tested, showing good evidence of total score reliability (α = .71; Stewart et al., 2000). It has also been translated and used in different populations (Santiago & Coelho, 2021). In this study, the scale was translated and adapted and then pilot tested. Psychometric analysis was performed to assess the scale's construct validity and internal consistency.

Health-related quality of life
The WHO Quality of Life Brief (WHOQOL-BREF) is used to assess self-perceived quality of life (WHOQOL Group, 1998). The scale has 26 items with four domains: physical health (seven items), psychological health (six items), social relations (three items), and environment (eight items). In addition to the four domains, the scale includes two questions related to respondents' rated quality of and satisfaction with health. The scale is applicable cross-culturally and has been widely translated into different languages and used with different populations. The WHOQOL-BREF domains were calculated by multiplying the mean of each domain's items by four. Evidence of excellent psychometric properties of reliability and validity have been found for this measure (Kalfoss et al., 2008;Rivas et al., 2015), with internal consistency ranging from .66 to .84. The Arabic version of WHOQOL-BREF (Ohaeri & Awadalla, 2009) was used in this study. The internal consistency was .76 for the physical domain, .61 for the psychological health domain, .93 for the social relations domain, and .97 for the environmental domain.

Ethical Considerations
The study was approved by the institutional review board (IRB) of both the Ministry of Health (IRB No. 20-07E) and King Saud University . After explaining the study procedure and purpose as well as the voluntary nature of the study, informed consent was obtained from each participant before data collection. Each participant was assigned a unique ID number to ensure data were anonymous and confidential.

Data Analysis
Data analysis was performed using IBM SPSS Statistics Version 28 (IBM, Inc., Armonk, NY, USA) and Mplus Version 8.6 (Muthén & Muthén, Los Angeles, CA, USA). Missing data were assessed and found to range between 0.4% and 3.2%, so missing data were replaced with the sample mean. Descriptive statistics were used to describe the study participants. Then, the psychometric properties of the newly translated scales were assessed. Factor analysis was conducted to assess the construct validity of the scales (AAHLS, HLPCQ, and PPPC) and determine their subscales. Various indices were used to determine whether the scales' structures adequately fit the data. The fit indices were root mean square error of approximation (RMSEA), standardized root-mean-square residual (SRMR), comparative fit index (CFI), Tucker-Lewis index (TLI), and chi-squared test. Internal consistency reliability for each scale was assessed based on Cronbach's alpha.
After confirming the factor structure of the scales, descriptive statistics were computed for each variable. Next, SEM was conducted to test the hypothesized model simultaneously while controlling for participant age and education. All of the variables were latent variables. The distribution of each variable was checked based on skewness and kurtosis indices. All of the variables have skewness and kurtosis less than 2. Thus, the analysis was conducted using maximum likelihood estimation (Kline, 2015). In addition, fit indices were used to determine whether the model adequately fit the data. Standardized path coefficients (β) were reported to compare the strength of paths within the models, whereas unstandardized coefficients (B) were used to compare the strength of paths across the models (Kline, 2015).
In terms of functional health literacy, 49.3% of the participants "rarely" needed help and 41.7% "often" received the help they needed. In terms of communicative health literacy, 41.7% of the participants were "often" sure about understanding information from healthcare providers. In terms of critical health literacy, 52.3% of the participants did not really question their healthcare providers' advice. In terms of empowerment, 79.3% of the participants had not taken any action during the last year with regard to health issues related to family or the community.  Controlling for participant education and age, SEM was conducted to assess the hypothesized model, with all variables treated as latent variables. Health literacy has four indicators (functional, communicative, critical, and empowerment), lifestyle behavior has five (dietary healthy choices, dietary harm avoidance, daily routine, organized physical exercise, and social and mental balance), patient-centered care has two (exploring both the disease and illness experience and finding common ground), and quality of life has four (physical health, psychological health, social relations, and environment). The model was found to adequately fit the data: w 2 (107, N = 300) = 283.767, CFI = .964, TLI = 0.954, RMSEA = .074 (90% CI [0.064, 0.085]), and SRMR = .043. For all of the latent measures, the standardized loadings for factors were significant and ranged between .379 and .978. Health literacy was shown to positively and directly affect both patient-centered care (β = .614, B = 0.189, p < .001) and healthy lifestyle behaviors (β = .550, B = 1.211, p < .001). Moreover, both patient-centered care (β = .390, B = 1.301, p < .001) and healthy lifestyle behaviors (β = .322, B = 0.150, p < .001) were shown to positively and directly affect quality of life. Furthermore, health literacy was shown to have significant direct (β = .285, B = 0.292, p < .001) and indirect (β = .417, B = 0.428, p < .001) effects on quality of life. These findings support patient-centered care and healthy lifestyle behaviors as partially mediating the relationship between health literacy and quality of life. The standardized (β) regression coefficients for each path are shown in Figure 2.

Discussion
The results of this study contribute to the existing body of knowledge by focusing exclusively on women, reflecting that gender roles and expectations likely shape self-perceived HRQOL. This study showed that the overall hypothesized model fits with the data, suggesting its significance in the context of NCDs affecting women. Patient-centered care and healthy lifestyle mediate the relationships between the health literacy dimensions and HRQOL. The findings support the importance of considering different levels of health literacy skills, including levels of autonomy and empowerment, in medical decision making in women with chronic diseases.
The findings of a mean total health literacy score of 22.03 of 35, functional health literacy score of 8.64 of 12, communicative health literacy score of 6.04 of 9, critical health literacy score of 6.89 of 12, and empowerment score of 0.44 of 2 indicate that the participants had a relatively lower average level of health literacy than those in previous studies (X. Chen et al., 2021). However, in light of the adaption process and one item removed from consideration in this study, this comparison with previous studies may not be accurate. This study only included women, who may have a lower health literacy level than men (F. Zhang et al., 2021). Nevertheless, some studies found women had a higher health literacy level than men (H. Y. Lee et al., 2015). The lowest score in this study was for critical health literacy, which is considered the highest order of health literacy needed to appraise health information at the critical level (Sykes et al., 2013). Generally, health literacy is important for women with chronic diseases, as these skills help the individual adhere to medications and follow up with healthcare providers regularly (Karasneh et al., 2020).
In this study, the mean scores for quality of life in the physical, psychological, and social relationships and environmental dimensions were low compared with those in other studies (Strømme et al., 2020), indicating low HRQOL and highlighting the need to focus additional attention on Saudi women with NCDs. This study also supported the direct and indirect effects of health literacy on HRQOL. Given the chronicity of NCDs and the high incidence of comorbidities, these diseases have the potential to worsen patients' overall health, restrict their functional status and productivity, and limit their ability to achieve well-being. In this context, HRQOL may be considered as an indicator of best clinical practices (Megari, 2013). Thus, showing the influence of health literacy dimensions on HRQOL is an important finding that is consistent with the findings of some other studies (Ehmann et al., 2020) and inconsistent with others that found no association between health literacy and HRQOL (Couture et al., 2017). A possible explanation for these differences is that previous studies used different measurements, with some studies measuring the functional dimension of health literacy only (Couture et al., 2017). The findings of this study highlight the importance of all health literacy dimensions, including communicative, critical, and empowerment.
In addition, the findings of this study showed that receiving patient-centered care mediates the relationship between health literacy and HRQOL. This may be explained by health literacy being a dynamic concept that influences the interactions between patients and healthcare providers. Furthermore, promoting health literacy by increasing patients' literacy, communicative, and critical skills may enhance patient-centered care, which in turn can increase quality of life. To our knowledge, previous studies have not examined the mediating role of patient-centered care in the relationship between health literacy and quality of life. However, evidence has shown that patients with limited health literacy have a worse experience with patient-centered care (Rademakers & Heijmans, 2018). To address this mediating effect, changes in healthcare system norms and culture are needed to embrace patient-centered care (Bokhour et al., 2018) and promote HRQOL. In this study, patient-centered care, which reflects patients' perception of patient-centered care, was distinguished into two dimensions (exploring the disease and illness experience and finding common ground). Although the PPPC scale is designed primarily to measure the patient centeredness of physicians, its dimensions/components are highly consistent with nursing practice (Lusk & Fater, 2013).
Pursuing a healthy lifestyle as a way to promote and maintain personal health (Sidossis & Kales, 2022) is very important. This study highlighted the role of healthy lifestyle behaviors in addressing the effect of health literacy on HRQOL. Healthy people have pointed out that healthy lifestyle behaviors (e.g., physical activity, healthy dietary choices, avoiding tobacco, and controlling stress) are among the determinants of health (Centers for Disease Control and Prevention, 2020). This study highlighted that increasing health literacy influences the lifestyle of women positively, which is consistent with previous studies that found an effect of health literacy on lifestyle (Uemura et al., 2021). Although healthy lifestyle is a well-documented outcome of health literacy, previous studies have not examined the mediating role of healthy lifestyle behaviors in the relationship between health literacy and HRQOL. In the context of chronic diseases, considering the crucial impact of healthy lifestyle on HRQOL in women is highly important and deserves greater attention from healthcare providers and public health policymakers. Finally, the findings of this study also suggest the need for interventions designed to improve health literacy that can change behavioral risk factors and, ultimately, improve quality of life.

Limitations
Although the findings of this study are significant, several limitations should be considered. There may be additional, unconsidered factors that mediate the relationship between health literacy and HRQOL. Thus, future studies should examine this possibility. Moreover, the cross-sectional design used limits the ability to infer causal relationships. Thus, additional studies using a longitudinal design are needed. Although the participants were recruited using convenience sampling, which typically limits the generalizability of findings, the 20 PHCs were randomly selected from different neighborhoods, which supports the generalizability of our findings to other women with NCDs. The health literacy, perception of patient-centered care, and healthy lifestyle scales were newly used in this context in this study. They were translated and pilot tested and assessed for reliability as well as for content and construct validity. Further studies are needed to test the scales and verify the findings.

Conclusions and Implications
This study provides evidence that the multiple dimensions of health literacy influence HRQOL directly and indirectly through patient-centered care and a healthy lifestyle. Gender-specific health services are crucial to promoting women's health literacy and improving healthcare outcomes. Women's health literacy should be assessed in the context of NCDs in clinical settings. Healthcare providers, including nurses, should enhance HRQOL in women by developing health interventions such as empowering health education programs that promote advanced literacy skills and pay special attention to patient-centered care and lifestyle. These interventional programs should be designed specifically for women using a "gender lens." For example, nurses may adopt the framework for gender-transformative health promotion, which is designed to foster gender-related empowerment (Pederson et al., 2015). This framework, which employs a gender lens, may be used in tandem with integrating gender-related social factors in the context of lifestyle modification. Nurses can address barriers such as healthy dietary choices, physical Health Literacy and Chronic Disease VOL. 31, NO. 3, June 2023 exercise, and social and mental stability frequently faced by women in adopting healthy lifestyles. Patient-centered care, which is essential to nursing knowledge and practice (Ortiz, 2018), should be prioritized by nursing through designing tailored interventions for women with NCDs. Nurses can integrate patient-centered care principles such as understanding women's beliefs, requirements, preferences, and expectations in health intervention programs. In future studies, this type of intervention may be designed and tested on women. In addition, longitudinal studies will be needed to further test and verify the presence and significance of the cause-and-effect relationship between health literacy and HRQOL found in this study.