SEARCH

SEARCH BY CITATION

Keywords:

  • health literacy;
  • health-promoting behaviors;
  • multiethnic women;
  • Taiwan

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. Methods
  4. Results
  5. Discussion
  6. REFERENCES
  7. Biographies

Objective

To understand the current status of health literacy and the relationship between health literacy and health-promoting behaviors among multiethnic groups of women in Taiwan.

Design

Convenience and snowball sampling methods were used to recruit study participants. Data were collected using a cross-sectional questionnaire survey.

Setting

We recruited community female adults who lived in greater Taipei or Taoyuan areas (northern Taiwan) from January 1, 2010 through June 30, 2011.

Participants

A total of 378 female participants were contacted, of which 351 consented to participate and 347 completed valid questionnaires for analysis.

Methods

Health literacy was measured with the Taiwan Health Literacy Scale, and health-promoting behaviors were measured by the Chinese version of the Health-Promoting Lifestyle Profile.

Results

Participants had a moderate level of health literacy, and one third of them had inadequate health literacy. Participants with inadequate health literacy were more likely to be younger, not a high school graduate, and Vietnamese; to have a low monthly family income and no diagnosed diseases; to use a second language; and to regard TV/radio as the most useful source of health information. Health literacy alone could significantly predict health-promoting behaviors among the participants.

Conclusions

Our findings confirmed that low health literacy is prevalent among underprivileged women in Taiwan. Health-related programs that are literacy sensitive and culturally appropriate are needed to teach and encourage health-promoting behaviors.

Low health literacy is a major health-related concern throughout the world. The global prevalence of low health literacy is substantial. For example, 59% of Australian adults (Australian Bureau of Statistics, 2006), 60% of Canadian adults (Murray, Rudd, Kirsch, Yamamoto, & Grenier, 2007), 35% of American adults (Kutner, Greenberg, & Paulsen, 2006), and 15.5% of Japanese adults (Tokuda, Doba, Butler, & Paasche-Orlow, 2009) are not health literate. In Taiwan, nearly 30% of adults are health illiterate, based on a national survey conducted in 2008 (S. Lee, Tsai, Tsai, & Kuo, 2010). Low health literacy has been found to be related to increased rates of hospitalization, use of emergency services, difficulty in taking medications and understanding labels/health messages, poorer general health status, all-cause death rates of elderly people, and decreased use of mammography screenings and influenza vaccinations (Berkman, Sheridan, Donahue, Halpern, & Crotty, 2011). Mothers with low health literacy have been associated with increased rates of infant mortality (Macinko, Guanais, & de Souza, 2006) and poor health behaviors in their young children, such as nighttime bottle use and no daily teeth cleaning (Vann, Lee, Baker, & Divaris, 2010). In contrast, researchers have found that women with higher health literacy tend to have better oral health status (J. Y. Lee, Divaris, Baker, Rozier, & Vann, 2012) and more accurately perceive recurrent breast cancer risk (Brewer et al., 2009). They also engage more often in preventive health behaviors, such as having a mammography (Bennett, Chen, Soroui, & White, 2009), checking food expiration dates, and monitoring physical changes (S. Lee, Tsai, Tsai, & Kuo, 2011). Although a consensus has not been reached regarding the relationship between health literacy and gender differences, women's health literacy is especially important because it affects not only a woman's personal health, but also the health of her family in her role as the primary caregiver.

One third of Taiwan's multiethnic women are inadequate in health literacy. Immigrant women from Vietnam had the lowest health literacy scores.

Numerous demographic factors are related to low health literacy. For instance, old age (Kutner et al., 2006; Osborn, Paasche-Orlow, Baily, & Wolf, 2011), low education attainment (Howard, Sentell, & Gazmararian, 2006; Kim, 2009; Osborn et al., 2011), low-income status (Kim; von Wagner, Knight, Steptoe, & Wardle, 2007), ethnic minority (Bennett, Chen, Soroui, & White, 2009; Chang, 2010; Howard et al., 2006), the language that is spoken is other than mother language (Todd, Harvey, Hoffman-Goetz, 2011; Zanchetta & Poureslami, 2006), and the presence of chronic illness (Kim, 2009) have been found to be related to low levels of health literacy. Although numerous studies have investigated the related factors of health literacy, few of these studies have targeted multiethnic women, especially multiethnic women in Taiwan.

Taiwan is a state in East Asia. From a Western perspective, people in Taiwan share the same physical characteristics and are often categorized as Asian; however, in reality, they are quite different culturally. Although not officially recognized, it is generally accepted that except for new immigrants from other countries, four major ethnic groups exist in Taiwan: Hokkien (67.5%), Hakka (18.1%), Waishengren (7.1%), and indigenous peoples (2%) (Hakka Affairs Council, 2011). Indigenous peoples, or Austronesians, first inhabited Taiwan 12,000 to 15,000 years ago and, in the 17th century, were joined by large numbers of Han peoples, mostly Hokkien and Hakka, who migrated to Taiwan from the southeastern provinces of the Chinese mainland (Government Information Office, 2012; Tsai & Chiu, 1993). In 1949, 1.5 million people from mainland China fled to Taiwan due to civil war and were called Waishengren by early Hokkien and Hakka immigrants (Government Information Office, 2012; Tsai & Chiu, 1993). In addition to these four major groups, new immigrants have formed a fifth ethnic group in Taiwan (Sandel & Liang, 2010). The population of the fifth group (3.2%) has increased rapidly and outnumbers the indigenous peoples. The majority of new immigrants came to Taiwan through international marriage, and most of them are female. The largest group of new immigrants via marriage is from mainland China (64.22%), followed by immigrants from Vietnam (18.9%), Indonesia (6.06%), and other countries (National Immigration Agency, 2012).

Multiethnic groups are substantial not only in Taiwan but also in the United States. The ethnic population of the United States consists of non-Hispanic Whites, African Americans, Native Americans, and Asian Americans. According to the 2010 census, Asian Americans are the fastest-growing population and increased from 10.2 million in 2000 to 14.7 million in 2010 (U.S. Department of Commerce, 2011). Cultural and language differences may create barriers for ethnic minority women such as Asians, especially due to their levels of health literacy to “obtain, process, and understand basic health information and services” (U.S. Department of Health and Human Services, 2000, pp. 11–20) and in their abilities to engage in health-promoting behaviors (Shaw, Huebner, Armin, Orzech, & Vivian, 2008; Zanchetta & Poureslami, 2006). Although some scholars have acknowledged the importance of culture in health literacy among indigenous peoples (Smylie, Williams, & Cooper, 2006) and immigrants (Zanchetta & Poureslami, 2006), the status of health literacy and health-promoting behaviors among multiethnic groups of women has not been widely discussed in either Taiwan or other countries. The International Council of Nurses (ICN; 2009) emphasizes that health care providers have to understand the status of a client's health literacy and the influence of health literacy on health behaviors. With thorough understanding of the crucial impact of health literacy on health status and health behaviors, the goal of improvement of the level of health literacy and promotion of health behaviors could be achieved ultimately. Therefore, it is critical to explore the status of health literacy and the relationship between health literacy and health-promoting behaviors.

The aim of this study was to investigate not only the present status of health literacy, but also the relationship between health literacy and health-promoting behaviors among multiethnic women in Taiwan. For the purpose of this study, we investigated three research questions: How prevalent is low health literacy among multiethnic women in Taiwan? What demographic characteristics are associated with low health literacy among multiethnic women in Taiwan? What is the relationship between health literacy and health-promoting behaviors among multiethnic women in Taiwan?

Methods

  1. Top of page
  2. ABSTRACT
  3. Methods
  4. Results
  5. Discussion
  6. REFERENCES
  7. Biographies

Study Design

We used convenience and snowball sampling methods for data collection in this cross-sectional study. These two sampling methods were employed because of difficulties in recruiting ethnic minorities, including indigenous peoples and new immigrants.

Setting and Participants

Prior to recruiting participants and collecting data, we received ethical approval to conduct this study from the institutional review board of Kaohsiung Medical University. From January 1, 2010, through June 30, 2011, we recruited and collected data from community-dwelling female adults who lived in the greater Taipei or Taoyuan areas located in northern Taiwan. A variety of participant recruitment methods were used. In addition to distributing fliers, posting announcements, and providing research information to community dwellers, we contacted government organizations, local health departments, churches, private companies, and community women's groups. Eligible participants met the following inclusion criteria: were at least age 18; self-identified as Taiwanese (Hokkien, Hakka, or Waishengren), Aborigine, Chinese mainlander, or Vietnamese; lived in the greater Taipei or Taoyuan areas; and were capable of reading or writing, or of communicating with the interviewers. We obtained each participant's consent before administrating the questionnaire. For those who could read or write Chinese, or could communicate with the interviewers but could not understand the questionnaires fully, a trained interviewer completed the questionnaire for them based on their verbal answers and asked the participants to reconfirm their answers.

Sample size for this study was calculated based on results of a prior study about health literacy of women in Taiwan (data not published). In that prior study, correlation coefficient between health literacy and health-promoting behaviors was .20 (N = 330). We used G*Power, a statistical power analysis program (Faul, Erdfelder, Buchner, & Lang, 2009). With correlation coefficient of .20 and power of .80, a minimum of 280 participants was required for this study. Among the 378 women who met the inclusion criteria and were contacted to participate in the study, 351 consented to their participation, reflecting a response rate of 92.9%. The total number of valid questionnaires that were analyzed was 347.

Measures

Two questionnaires in three language versions were used in this study: traditional Chinese, simplified Chinese, and Vietnamese. These two questionnaires were developed in traditional Chinese by Taiwanese scholars. Because traditional Chinese and simplified Chinese differ only in written characters and can be converted directly, we asked one language expert to convert the questionnaires in traditional Chinese form to simplified Chinese. For Vietnamese version, we had the questionnaires translate into Vietnamese by a translation company who followed a standard procedure for translating and validating instruments. Three bilingual experts in Chinese and Vietnamese were invited to confirm the content validity of the translated instruments after the instruments were translated.

Health Literacy

Health literacy in this study was defined as a way to read and understand the relevant information about health or medical knowledge (American Medical Association, 1999). We used a modified version of the Taiwan Health Literacy Scale (THLS) to assess health literacy. The THLS contains 66 terminologies categorized into nine health-related factors: pharmaceutical, top 10 death causes, general diseases, organs, physiological, physical examination, medical treatment, disease symptoms, and superficial characteristics of diseases (Pan, Su, & Chen, 2010). The THLS measures the participant's health literacy by asking if she has ever heard the terminology and how well she understands it. The original version of the THLS is a 5-point Likert-type scale, graded as 1 (have never heard the term), 2 (have heard but do not understand the term), 3 (have heard and understand the term a little), 4 (have heard and understand the term a lot), and 5 (have sufficient knowledge of the term; equivalent to a health professional's knowledge of the term) (Pan et al., 2010). Because the meanings of options 3 and 4 are similar in the scale's original Chinese version, we merged the two options into one; therefore, the version of the THLS used in this study became a 4-point scale, graded as 1 (have never heard the term), 2 (have heard but do not understand the term), 3 (have heard and understand the term), and 4 (have sufficient knowledge of the term).

The THLS has been proven to have satisfactory construct and criterion-related validity in scale development research (Pan et al., 2010). In this study, the Cronbach's alpha for the scale was .99, and the mean item-total correlation was .74 (ranging from .61 – .84). Construct validity of the modified THLS was supported by factor analysis, which showed that 52.6% of the variance of health literacy could be explained by one factor. According to the scale developers, individuals who score lower than 3.0 (out of 5) on the THLS may encounter a certain degree of difficulty in understanding health-related information (Pan et al., 2010); therefore, we used < 2.5 (out of 4) as the cutoff point to distinguish inadequate from adequate health literacy.

Health-Promoting Behaviors

To assess the participants’ adult health-promoting behaviors, we used an inventory developed by Taiwanese scholars who modified the Health-Promoting Lifestyle Profile created by Walker, Sechrist, and Pender (1987) and adapted it for people in Taiwan (Chen et al., 1997). This Chinese version of the Health-Promoting Lifestyle Profile (HPLP-Chinese) contains 40 items and is categorized into the following six health behaviors: nutrition, health responsibility, self-actualization, interpersonal support, exercise, and stress management. The instrument is a Likert-type scale ranging from 1 (never) to 4 (routinely). The total score ranges from 40 to 160; the higher the score, the greater the intent of the participant to adopt a healthier behavior.

The HPLP-Chinese has been used in the assessment of health-promoting behaviors of new immigrant women (Wei, Chen, Chen, Chang, & Chen, 2010) and adolescents (Chen, Wang, Yang, & Liou, 2003) in Taiwan, with good internal consistency, reliability, and validity as a categorized and continuous measure of health-promoting behavior (Cronbach's alpha = .90 to .94). In this study, the Cronbach's alpha for the HPLP-Chinese was .93, and the mean item-total correlation was .49 (ranging from .33 – .64).

Demographic Variables

Demographic variables included age, education, ethnicity, language, monthly family income, employment, marital status, diagnosed diseases, and the most useful sources of health information. Based on the participants’ self-identification, ethnicity was divided into four groups: Taiwanese, Aborigine, mainland Chinese, and Vietnamese. Taiwanese participants included those who considered themselves Hokkien, Hakka, or Waishengren. The official language in Taiwan is Mandarin Chinese, but other dialects such as Minnan-yu and Hakka-fa are also frequently used by lay people; therefore, we categorized languages into a dominant language (Mandarin, Minnan-yu, & Hakka-fa) and a second language (Aboriginal language, Vietnamese, English, & other languages).

For the demographic variable of diagnosed diseases, participants were asked whether they had any chronic disease, such as hypertension, diabetes, heart disease, and liver disease. We also asked the participants to identify the following type of resource that provided them with the most useful health information: TV/radio, books/magazines/brochures, Internet, or health institution (e.g., health-related workshops held at hospitals, clinics, or community health stations).

Statistical Methods

Descriptive statistics were used to analyze the participants’ demographic information and level of health literacy. Demographic data were analyzed by means and standard deviation. Cronbach's alpha was used for internal validity. Normality of all data was tested before conducting inferential statistics. Scatter plots of the data were examined to confirm linearity. Missing data were examined, and the expectation maximization algorithm was applied to manage missing data if they were missing at random.

The Mann-Whitney U test and the Kruskal-Wallis test were used to compare health literacy by categorical variables because health literacy was not normally distributed. ANOVA was used to test differences between the various ethnic groups on age and health promoting behaviors. Spearman's correlation was used to analyze relationships between continuous variables. Multiple regression analysis was conducted for predictive effect of health literacy on health-promoting behaviors. Demographic variables were included in multiple regression models only if they were associated with health-promoting behaviors. Two-tailed test and α level of .05 were used. Data analysis was conducted using SPSS version 15.

Results

  1. Top of page
  2. ABSTRACT
  3. Methods
  4. Results
  5. Discussion
  6. REFERENCES
  7. Biographies

Descriptive Analysis

Table 1 presents an overall statistical description of the study population. The mean age of the participants was 37.42  (range 15–85 years, SD = 11.34). Approximately one half of the participants were Taiwanese and the other one half were ethnic minorities including Aborigines, mainland Chinese, and Vietnamese. Education attainment and monthly family income of the participants were almost equally distributed among all the ethnic groups. Most of the participants were employed, were married, had no diagnosed diseases, and spoke dominant languages including Mandarin, Minnan-yu, and Hakka-fa. The majority of the participants deemed TV and radio as their most useful sources of health information.

Table 1. Descriptive Analysis of Variables (N = 347)
 M ± SDRangen%
Note
  1. A mean health literacy score equal to or greater than 2.5 was categorized as adequate, whereas a score less than 2.5 was categorized as inadequate.

Health literacy2.60 ± 0.551.17–3.91  
Adequate  23266.8
Inadequate  11533.2
Health-promoting behaviors103.18 ± 17.6958–154  
Age (in years)37.42 ± 11.3415–85  
Education    
< High school  10730.8
High school  10931.4
> High school  12536.1
Missing  61.7
Ethnicity    
Taiwanese  17049.0
Aborigine  4412.7
Mainland Chinese  4011.5
Vietnamese  9326.8
Family monthly income (New Taiwan Dollar)    
< 30,000  12034.6
30,000–50,000  10831.1
> 50,000  10430.0
Missing  154.3
Language mostly used    
Dominant (Mandarin, Minnan-yu, and Hakka-fa) 27679.5 
Second (Aboriginal, Vietnamese, English, and other) 7020.2 
Missing  10.3
Employment    
Employed  21562.0
Unemployed  12636.3
Missing  61.7
Marital status    
Single  8524.5
Married  21762.5
Other  3510.1
Missing  102.9
Diagnosed diseases    
No  24370.0
Yes  10430.0
Most useful sources of health information    
TV/radio  12937.2
Books/magazines/brochures  318.9
Internet  5114.7
Health institution  4513.0
Missing  9126.2

One third of the participants were inadequate in health literacy. Their mean scores in health literacy and health-promoting behaviors were moderate. Table 2 outlines differences in demographic variables, health literacy, and health-promoting behaviors by ethnicity (Taiwanese, Aborigine, mainland Chinese, & Vietnamese) and language (dominant & second). Differences in levels of health literacy by ethnicity and language were significant. The majority of Taiwanese participants had adequate health literacy, whereas most Vietnamese participants had inadequate health literacy. Participants using a dominant language were more likely to be adequate in health literacy, compared to those using a second language. No matter what ethnic group the participants belonged to or what languages they used, no significant differences of health-promoting behaviors were identified between groups.

Table 2. Ethnic Characteristics of Participants
 TaiwaneseAborigineMainland ChineseVietnameseχ2p
 n (%)n (%)n (%)n (%)  
Note
  1. Chi-squared test was not done for variables that had expected count less than 5. For variables education, monthly family income, employment, marital status, and most useful sources of health information, Chi-squared test was done by excluding category of missing.

  2. a

    Dominant languages included Mandarin, Minnan-yu, and Hakka-fa; secondary languages included Aboriginal language, Vietnamese, English, and other language.

Sample size170 (49.0)44 (12.7)40 (11.5)93 (26.8)  
Health literacy    85.08<.001
Adequate150 (43.2)22 (6.3)28 (8.1)32 (9.2)  
Inadequate20 (5.8)22 (6.3)12 (3.5)61 (17.6)  
Education    105.98<.001
<High school19 (5.5)24 (6.9)11 (3.2)53 (15.3)  
High school49 (14.1)10 (2.9)14 (4.0)36 (10.4)  
> High school100 (28.8)9 (2.6)12 (3.5)4 (1.2)  
Missing2 (.6)1 (.3)3 (.9)0 (0)  
Languagea    
Dominant169 (48.8)43 (12.4)40 (11.6)24 (6.9)  
Second1 (.3)1 (.3)0 (0)69 (19.9)  
Monthly family income (new Taiwan dollar)    41.95<.001
<30,00045 (13.0)23 (6.6)17 (4.9)35 (10.1)  
30,000–50,00043 (12.4)12 (3.5)11 (3.2)42 (12.1)  
> 50,00076 (21.9)6 (1.7)10 (2.9)12 (3.5)  
Missing6 (1.7)3 (.9)2 (.6)4 (1.2)  
Employment    8.18.04
Employed115 (33.1)25 (7.2)27 (7.8)48 (13.8)  
Unemployed52 (15.0)18 (40.9)12 (30.0)44 (47.3)  
Missing3 (1.8)1 (.3)1 (.3)1 (.3)  
Marital status    53.94<.001
Single68 (19.6)10 (2.9)4 (1.2)3 (.9)  
Married85 (24.5)24 (6.9)29 (8.4)79 (22.8)  
Other16 (4.6)7 (2.0)6 (1.7)6 (1.7)  
Missing1 (.3)3 (.9)1 (.3)5 (1.4)  
Diagnosed diseases    25.57<.001
No104 (30.0)28 (8.1)27 (7.8)84 (24.2)  
Yes66 (19.0)16 (4.6)13 (3.7)9 (2.6)  
Most useful sources of health information    
TV/radio48 (13.8)17 (4.9)25 (7.2)39 (11.2)  
Books/magazines/brochure24 (6.9)2 (.6)5 (1.4)0 (0)  
Internet40 (11.5)3 (.9)3 (.9)5 (1.4)  
Health institution15 (4.3)12 (3.5)1 (.3)17 (4.9)  
Missing43 (12.4)10 (2.9)6 (1.7)32 (9.2)  
Age (in years)37.96 ± 11.9542.76 ± 12.2741.36 ± 12.9132.01 ± 5.2911.91<.001
Health-promoting behaviors104.90 ± 18.80102.32 ± 20.15101.97 ± 15.79100.97 ± 14.861.12.34

Relationships between Measured Variables

As presented in Table 3, age had a positive relationship with health literacy (r = .21), indicating that older participants were more health literate. With the exceptions of employment and marital status, all demographic variables were related to health literacy. Data showed that participants with low health literacy were more likely to be younger, Vietnamese, and not a high school graduate; to use a second language; to have a low monthly family income and no diagnosed diseases; and to regard TV and radio as the most useful sources of health information. The participants’ health-promoting behaviors had no relationship with any demographic variables, except for age (r = .18, < .01) and the most useful sources of health information (F = 3.74, p < .01). Participants who regarded TV and radio as the most useful sources of health information had the lowest score of health-promoting behaviors. Health literacy was statistically significantly correlated with health-promoting behaviors (r = .29).

Table 3. Relationships between Measured Variables
 Health literacyHealth-promoting behaviors
  rp rp
Note
  1. The Spearman correlation was used for continuous variables. The Mann-Whitney U test was used for variables that had two categories whereas the Kruskal-Wallis test was used for variables that had more than two categories.

Health-promoting behaviors .29< .001 --
Age .21< .001 .15.01
 M ± SDχ2pM ± SDχ2p
Education 61.65<.001 2.93.23
<High school2.48 ± .86  102.45 ± 17.28  
High school2.97 ± .73  101.44 ± 18.18  
> High school3.24 ± .65  105.22 ± 17.80  
Ethnicity 97.94<.001 2.98.40
Taiwanese3.26 ± .63  105.03 ± 18.79  
Aborigine2.84 ± .73  102.32 ± 20.15  
Mainland Chinese2.99 ± .67  101.97 ± 15.79  
Vietnamese2.25 ± .78  100.97 ± 14.86  
Monthly family income (NT$) 18.22<.001 1.44.49
<30,0002.82 ± .88  101.69 ± 18.23  
30,000–50,0002.76 ± .80  102.06 ± 16.76  
> 50,0003.17 ± .65  105.02 ± 17.71  
Language mostly used 8.22<.001 −1.01.31
Dominant3.09 ± .72  103.75 ± 18.33  
Second2.19 ± .73  101.06 ± 14.97  
Employment 1.21.23 −.44.66
Employed2.91 ± .80  102.83 ± 17.72  
Unemployed2.88 ± .81  103.40 ± 17.78  
Marital status 3.58.17 5.98.05
Single3.08 ± .66  99.73 ± 18.85  
Married2.86 ± .85  104.58 ± 17.08  
Other2.84 ± .86  102.61 ± 18.35  
Diagnosed diseases 3.16.002 −1.47.14
No2.81 ± .81  102.16 ± 17.35  
Yes3.13 ± .76  105.57 ± 18.34  
Most useful sources of health information 15.82.001 10.64.01
TV/radio2.69 ± .81  99.52 ± 17.96  
Books/magazines/brochures3.26 ± .67  107.70 ± 17.04  
Internet3.11 ± .74  106.08 ± 15.70  
Health institution3.12 ± .80  107.06 ± 18.74  

Those individuals with inadequate health literacy are likely to be younger, have no diagnosed diseases, and regard TV/radio as the most useful health source.

To test the predictive effect of health literacy on health-promoting behaviors, demographic variables related to health-promoting behaviors were controlled for by entering them as the first set of regression, and health literacy was entered as the second set of regression. The first model, which included age and most useful sources of health information, explained 7% of the variance of health-promoting behaviors (Model 1). The addition of health literacy to the second model (Model 2) accounted for an additional 6% of the variance, and health literacy alone could significantly predict health-promoting behaviors (Table 4).

Table 4. Regression Analysis of Health Literacy on Health-Promoting Behaviors
 Model 1Model 2
 BtpBtp
Most useful sources of health information      
Health institution (referent)   
TV/radio−6.64−2.16.03   
Books/magazines/brochures2.50.59.55   
Internet1.35.37.72   
Age.252.49.01   
R2 = .07, F = 4.33, p = .002      
Most useful sources of health information      
Health institution (referent)   
TV/radio   −4.80−1.59.11
Books/magazines/brochures   1.33.32.75
Internet   .20.06.96
Age   .171.77.08
Health literacy   8.524.08<.001
    R2 = .13, F = 7.02, p < .001
    Fchange = 16.65, pchange < .001

Discussion

  1. Top of page
  2. ABSTRACT
  3. Methods
  4. Results
  5. Discussion
  6. REFERENCES
  7. Biographies

One third of the multiethnic women who participated in the study had inadequate levels of health literacy. This finding is similar to results in previous research conducted by S. Lee et al. (2011), who found that 29% of women in Taiwan had inadequate and marginal health literacy. However, the percentage of ethnic minority participants in this study far exceeded the proportion of the ethnic minority population in S. Lee et al.'s (2011) nationally representative female sample, in that their ethnic proportion included approximately 93% of Taiwanese women (Hokkien, Hakka, & Waishengren), 2% of Aboriginal women, and 3% of women belonging to other ethnic groups. In contrast, we had an equal number of ethnic minority women (Aborigines, Chinese mainlanders, & Vietnamese) and ethnic majority women (Taiwanese). Therefore, we might speculate that no matter what proportions the ethnic groups are, approximately one third of all women in Taiwan are inadequate in health literacy. In addition, if we compare ethnic minority women (Aborigines, Chinese mainlanders, & Vietnamese) with ethnic majority women (Taiwanese), more than one half of ethnic minority women in Taiwan are inadequate in health literacy.

Among the ethnic minority women in this study, new immigrants from Vietnam had the lowest mean score of health literacy, whereas new immigrants from China scored the second highest in health literacy. This finding is consisted with results in previous research conducted by Wang, Lin, Yang, Tsai, and Huang (2012) who found that Asian immigrant women are likely to have low levels of health literacy. These differences might be attributed to cultural disparities. Although both ethnic groups are influenced by Chinese culture, Chinese mainlanders share a much more similar culture with Taiwanese than do Vietnamese. Moreover, Chinese mainlanders and Taiwanese in Taiwan use a dominant language (Minnan-yu, Hakka-fa, or Mandarin Chinese), whereas Vietnamese in Taiwan mostly use a second language (Vietnamese). Our finding that multiethnic women who use a second language tend to have inadequate health literacy confirms and is consistent with results from previous research (Todd et al., 2011; Zanchetta & Poureslami, 2006). Language has been identified as an issue relative to health literacy among Asian Americans. Researchers indicate that Asian Americans with limited English proficiency tend to have limited health literacy (Kim & Keefe, 2010).

Our findings also confirm other researchers’ reports that inadequate health literacy is more prevalent among women without a high school diploma (Howard et al., 2006; Kim, 2009; Osborn et al., 2011) and with a low family income (Kim, 2009; von Wagner et al., 2007). From the above-mentioned results, it is easy to conclude that people who are from ethnic minority groups, who do not use the dominant language of the host country, and who have lower education attainment and lower income are susceptible to low health literacy. Individuals with these demographic characteristics are often deemed underprivileged or underserved. Caring for underprivileged people is a nation's responsibility. Improving their level of health literacy is a way of empowering underprivileged people to be equally treated by the society (Kickbusch, 2001).

Contrary to results from other studies, we found that individuals without a diagnosed disease had lower health literacy than those with a diagnosed disease. Results from other studies indicated that patients with chronic diseases, such as hypertension or arthritis (rheumatic pain), lack the interest or mobility to actively search for health information (Kim, 2009; Koo, Krass, & Aslani, 2006); therefore, the researchers concluded that patients with chronic diseases are less likely to be health literate. These differences between findings from previous studies and this study might be attributed to a lack of awareness among this study's participants. Individuals without a chronic disease may be less likely to be aware of the importance of health care than those with a chronic disease. Furthermore, such a lack of awareness may have prevented the participants without a chronic disease from hearing or even understanding the meaning of the terminology assessed in this study's survey questionnaire, which in turn lowered their scores of health literacy.

A lack of awareness about the importance of health care might also explain our other finding that younger women tended to be health illiterate. This finding is contrary to previous research suggesting that older people who may also be vulnerable are more likely to have low health literacy (Kutner et al., 2006; Paasche-Orlow, Parker, Gazmararian, Nielsen-Bohlman, & Rudd, 2005). When Pan et al. (2010) developed the THLS, they found two reversed U-shape relationships between age and health literacy: (a) People who scored the highest on the THLS were age 20 to 29, and the score dropped after age 30 to 39; (b) the second peak appeared at age 50 to 59, and the score downturned afterward. The mean age of this study's participants was approximately 37, which is the age bracket that marks the first downturn point in health literacy identified in Pan et al.'s study. The rationale for older people having low health literacy can be attributed to the physical deterioration of their cognitive abilities (Federman, Sano, Wolf, Siu, & Halm, 2009), whereas the reason why younger people have lower scores in health literacy might be social-cultural. We speculated two possible explanations. First, the Vietnamese participants in this study were the youngest, and most of their health literacy was inadequate. Second, like the relationship between individuals with diagnosed diseases and a higher level of health literacy, 30-something-year-old women may not have developed any chronic diseases and, moreover, may tend not to pay much attention to health-related information. Further research is needed to investigate the relationships among age, having a diagnosed disease, and health literacy.

This study's results showed that individuals with the lowest health literacy score considered TV and radio programs as the most useful sources of health-related information, whereas individuals with the highest score of health literacy considered books, magazines, or brochures as the most useful sources of health information. This finding is similar to results yielded from past research in which individuals with low health literacy tended to seek health information more frequently from nonprint media (TV/radio) than from printed media (books, magazines, or brochures) (Kutner et al., 2006). It is understandable that people with lower literacy abilities might encounter difficulties when reading print media because a higher level of reading ability is required, whereas watching or listening to TV and radio programs requires less literacy abilities. However, based on the findings generated from this study, it seems that nonprint media is not very effective in delivering health-related information.

Although the education value of TV or radio programs cannot be overlooked, the programs must be well designed to deliver not only correct but also efficient health information during a limited time frame. Even though print media provide more in-depth details of health information than broadcast media, reading materials should be simplified and comprehensible to fit the need of people with lower literacy abilities. It is important to acknowledge that no single medium that delivers health messages is equally effective for all audiences. Some people comprehend best through visual media, whereas others learn best through audio media (P. P. Lee, 1999).

The relationship between health-promoting behaviors and ethnicity seems to be varied across studies. We found that health-promoting behaviors had no significant differences among multiethnic women in Taiwan. Wang's (1999) study yielded the same result, in which no differences in health-promoting lifestyles were found across three ethnic groups (Hokkien, Hakka, and Aborigine) of elderly females in Taiwan. Similarly, Bennett et al. (2009) found that among older adults, race/ethnicity was not related to utilizing mammogram screenings; however, race/ethnicity was related to older adults utilizing influenza vaccinations and dental checkups. Other studies have provided evidence that certain healthy behaviors have significant differences across female ethnic groups. For example, Gavin, Fox and Grandy (2011) found that, among female adults with type 2 diabetes, Hispanics had the lowest rate of physical activity. Asian Americans had lower screening rates for breast cancer compared to White women (Kim & Keefe, 2010). The discrepancy in various studies’ findings related to the relationship between ethnicity and healthy behaviors might be due to the differences of sample populations and the health behaviors measured. More research is needed to reach a convincing result of the relationship between ethnicity and health-promoting behaviors.

Data from this study not only revealed a positive relationship between health literacy and health-promoting behaviors, but also indicated that health literacy can predict health-promoting behaviors independently. Cho, Lee, Arozullah and Crittenden (2008) also documented a positive relationship between health literacy and health behaviors with the same measuring tool that we used in this study; however, their study results yielded a stronger relationship coefficient (r = .42) compared to our results (r = .29). This difference might be attributed to Cho et al.'s sample population, which included both genders, whereas this study assessed only females. Other studies have also demonstrated a positive relationship between health literacy and certain healthy behaviors, such as monitoring physical changes and checking food expiration dates (S. Lee et al., 2011), utilizing vaccinations (Howard et al., 2006), eating at least five daily servings of fruit and vegetables, and exercising in the last 7 days (von Wagner et al., 2007). However, not all healthy behaviors are associated with health literacy. For example, S. Lee et al. (2011) found that individuals’ level of health literacy was not associated with utilizing physical checkups and Pap smear screenings. Also, in a review of published studies that examined the relationship between health literacy and health, Easton, Entwistle and Williams (2010) did not find a clear association between health literacy and preventive health or health-risk behaviors. Therefore, more research is needed to confirm the relationship between health literacy and health-promoting behaviors.

Limitations

This study has some limitations. First, we only surveyed women in the greater Taipei and Taiyuan areas located in northern Taiwan. Nevertheless, these areas are ideal for recruiting ethnic minority participants in Taiwan because they contain 31.5% and 41.8% of the total population of Vietnamese and mainland Chinese immigrants, respectively (Government Information Office, 2012). Second, by using nonprobability sampling methods (snowball and convenience), the generalizability of this study's findings is limited; however, the prevalence of low health literacy identified in this study is similar to S. Lee et al.'s (2011) research, which used a nationally representative sample. Third, the THLS does not fully capture all dimensions of the concepts of health literacy; moreover, it cannot assess a variety of other aspects of health literacy, such as abilities in computing and communicating.

Conclusions and Implications

This study is the first to investigate the health literacy status and the relationship between health literacy and health-promoting behaviors among multiethnic groups of women in Taiwan. We found that (a) one third of Taiwan's multiethnic women are inadequate in health literacy; (b) multiethnic women in Taiwan who have a low level of health literacy are more likely to be younger, be Vietnamese immigrants, use a second language, lack a high-school diploma, earn a low monthly family income, have no diagnosed diseases, and regard TV and radio as the most useful sources of health information; and (c) health literacy explains a small fraction of the differences in health-promoting behaviors among multiethnic women in Taiwan.

Health literacy explains a small fraction of the difference in health-promoting behaviors.

In general, we found that low health literacy could be more prevalent among underprivileged women who are mostly of ethnic minority and use a second language. Underprivileged women in the United States, such as Asian immigrants whose primary language is not English and who may have limited English proficiency with insufficient English reading and writing skills, contributes to their lower health literacy levels. Statistics showed that Asian Americans have the highest rates of limited English proficiency in the United States (Yip, 2012). It is crucial to increase their health-related reading and writing abilities so that to increase their capacity to use health information effectively. In other words, it is important to empower this group of people by means of multiple teaching activities. For instance, individual tutoring by healthcare providers can be a way to provide information of health literacy for immigrant women. A group learning program may also be effective to motivate immigrant women's learning interest and enhance interactive communication. Social welfare departments and health authorities should work together to help underprivileged women gain such capabilities.

Furthermore, female new immigrants from Vietnam have the lowest health literacy scores compared to other ethnic women in Taiwan. We recommend that social welfare departments and health authorities work together to help underprivileged women develop and use such abilities. We also suggest that health authorities design culturally appropriate health-promotion programs for new immigrants, including media-related health messages that fit the needs of women with inadequate health literacy. Other materials such as computer games, pocket books, audio or video disks, and mobile phone applications that teach health-related terminologies and applications are worth developing to facilitate women in learning and increasing knowledge about health.

REFERENCES

  1. Top of page
  2. ABSTRACT
  3. Methods
  4. Results
  5. Discussion
  6. REFERENCES
  7. Biographies

Biographies

  1. Top of page
  2. ABSTRACT
  3. Methods
  4. Results
  5. Discussion
  6. REFERENCES
  7. Biographies
  • Hsiu-Min Tsai, RN, PhD, FAAN, is a professor in the Department of Nursing, Chang Gung University of Science and Technology, Tao-Yuan, Taiwan.

  • Ching-Yu Cheng, RN, PhD, is an associate professor in the Department of Nursing, Chang Gung University of Science and Technology, Chiayi, Taiwan.

  • Shu-Chen Chang, RN, MSN, is the director of the Department of Nursing, Changhua Christian Hospital, Changhua, Taiwan.

  • Yung-Mei Yang, RN, PhD, is an assistant professor in the College of Nursing, Kaohsiung Medical University, Kaohsiung, Taiwan.

  • Hsiu-Hung Wang, RN, PhD, FAAN, is a professor in the College of Nursing, Kaohsiung Medical University, Kaohsiung, Taiwan.