Abstract
Objectives: To measure expatriates’ knowledge of health insurance benefits with respect to outpatient, inpatient, prescription drug, and dental services, and to link this knowledge to sociodemographic and employment characteristic.
Methods: Cross-sectional, face-to-face interviews were conducted from March 2015 to February 2016 with a stratified random sample of 3,398 male insured expatriate workers in the private sector of Riyadh, Kingdom of Saudi Arabia. Descriptive analysis, one-way ANOVA, and linear regression were used to interpret data. Data on knowledge of health insurance benefits was compiled by adding the scores of the 4 service category items (outpatient, inpatient, prescription drug, and dental services) to those of the 5 objective knowledge items on copayments.
Results: More than 87% of the participants reported knowledge on their health insurance benefits coverage for outpatient, 62% for the inpatient, 86% for the prescription drug and 62% for the dental services. However, 7.5% knew the correct copayments for general practitioner, 64% for the outpatient (inclusive), 13% for the inpatient visits, 15% for the prescription drugs and 9.6% for the dental care. Most personal and job characteristics had associations with knowledge measures. A total of 55% of the overall knowledge variance was explained by the independent variables (R2=0.55), suggesting that other factors also influence knowledge.
Conclusion: Expatriates’ knowledge of their health insurance benefits is very limited, which indicates the importance of developing a policy for increasing this knowledge among expatriates in Saudi Arabia.
Assessing people’s knowledge of health insurance is critical for efficient delivery of care. Many studies have assessed health insurance knowledge using various measures.1,2 Studies conducted by the Medicare or Medicaid insurance programs in the United States have targeted either older or low-income individuals,3,1 which raises the question of whether these study findings can be generalized to other populations. No previous studies have assessed expatriates’ knowledge of health insurance benefits in the context of Saudi Arabia. Historically, one study examined physicians’ perceptions of health insurance in the country prior to implementation of Compulsory Employment-Based Health Insurance (CEBHI).4 Other studies in the context of Saudi Arabia assessed knowledge on topics unrelated to health insurance, such as other medical services or health care professionals.5
Various studies from Asian countries have measured awareness regarding the health insurance, but have relied on self-reports from participants.6,7 A number of studies have commented on the limitations of self-reported information, namely, that participants might overestimate their awareness.8,9 In Saudi Arabia, over 84% of private sector employees are expatriates, and they account for 56% of total employment.10,11 This increases interest in assessing what these employees know about health insurance benefits, considering that the native language of the majority of expatriates is neither Arabic nor English.12 The CEBHI scheme not only makes it mandatory for all employers to bear the full cost of health insurance premiums but also determines the unified benefits packages, which are managed by a government body called the Council of Cooperative Health Insurance (CCHI).13 The CCHI is responsible for regulating and monitoring the implementation of health insurance benefits. In addition, all necessary medical examinations, treatments, medications, outpatient and inpatient testing, as well as dental and prescription drug services are determined by the unified health insurance policy. For example, the insured maximum copayments are predetermined so as not to exceed 20% of the health care expenses or a maximum of SAR 100 (USD 26.67).14 There are no coinsurance or deductibles for inpatient services and prescription drugs, and the health policy covers up to USD 533.30 for dental treatment.14 (More details on health insurance benefits coverage can be found on the CCHI website: www.cchi.gov.sa).14
Because the health insurance plan managed by CCHI includes prescription drugs and dental services as part of its benefits, these services can be included when assessing expatriates’ perceived knowledge and validating their knowledge of copayments. Few international studies have included knowledge assessment for dental care because many health insurance plans exclude dental services,15 and Medicare and Medicaid do not cover most dental services.16 This paper aims to measure knowledge about health insurance benefits, including those for dental care and prescription drugs, among expatriates, and to link this knowledge to sociodemographic and employment characteristics.
Methods
Cross-sectional survey data were collected in Riyadh, Kingdom of Saudi Arabia. Riyadh was chosen as the study site because more than one-third of the total expatriate population in the country lives in this city.17 Face-to-face interviews were conducted among 3,398 male insured expatriate workers in the private sector, with a response rate of around 96%. The questionnaire was orally administered by trained bilingual research assistants to overcome any language and literacy barriers. These research assistants spoke the expatriates’ main languages (Arabic, Bengali, Hindi, Malayalam, Nepali, Tagalog, and Urdu). The face-to-face interview mode is suitable for increasing participants’ comprehension of survey questions and is commonly used with expatriate populations.
Data collection was conducted over an 11-month period, from March 2015 to February 2016, at expatriates’ workplaces. However, data collection was stopped during national holidays and during the holy month of Ramadan. Participants’ employing organizations were identified and randomly selected from the Ministry of Labor database and were stratified based on business sector, company size, and number of employees, similar to a previously used approach.18 All sectors associated with health care and health insurance were excluded because the knowledge of individuals working in these sectors could be influenced by external factors. Statistical Package for the Social Sciences version 23 (IBM Corp, Armonk, NY, USA) was used for randomization. Company names and other relevant information were concealed and were only known to the manager of the Statistics Department at the Ministry of Labor.
Questionnaire development
There is no current consensus on measures to assess health insurance knowledge.1 Awareness about health insurance varies based on health insurance plans.9 Therefore, certain factors were considered in this study to increase the validity and reliability of our survey questions. First, the questionnaire was developed from a comprehensive review of the literature and used the same approaches as have been used in other studies assessing knowledge of health insurance benefits. Second, the questionnaire was reviewed by 4 experts and academicians in the areas of health insurance and research methods, as well as 2 health insurance experts.
Third, cognitive interviewing was used to pretest the questions. This is a useful tool when developing survey questions, as it allows modifications that can enhance the accuracy and reliability of the final instrument.19 Cognitive interviewing was conducted in 3 stages. The questionnaire was first piloted among 10 research assistants. Each research assistant was then asked to administer the questionnaire to 10 selected participants (a total of 110 surveyed). Finally, the research team met to review the survey questions and provide their feedback.
Study variables
Independent variables in this study included questions related to expatriates’ health status, personal and job characteristics, and previous knowledge concerning the insurance in their home countries. Previous knowledge was assessed by asking several questions, which were then summarized as one independent variable denoted as “previous knowledge.” An individual was considered to have previous knowledge if the answer to any of the following questions was yes: “Have you seen any materials regarding Cooperative Health Insurance?” and “Do you know who pays for your health care costs in your home country (health insurance)? Your life insurance, auto insurance, crop insurance, pension insurance, health or medical insurance, unemployment insurance, benefits or other?”
Personal characteristics included age, nationality, educational level, marital status, length of employment in Saudi Arabia, and comfort levels for conversing in Arabic and English. Job characteristics included the respondent’s position in the company, company size and economic sector, and monthly income. Questions for both personal and job characteristics were adapted from the Medical Expenditure Panel Survey,20 with a few modifications to suit the study site. Although an employee’s position in the company determines the required level of education, most expatriates in Saudi Arabia report working below their actual educational level.11 Therefore, this study used Ministry of Labor data to determine the educational levels of respondents.21 The dependent variables were expatriates’ perceived knowledge and objective (verified) knowledge of their main health insurance benefits. Nine questions were asked to measure knowledge, as described in the next section.
Statistical analysis
To ensure the accuracy of data entry, all data were double-entered into a Microsoft Excel spreadsheet. The 2 data sets were then compared, differences examined, errors verified against the original questionnaire, and corrections made. The final data set was then exported to SPSS, where frequencies for all variables were generated. The frequency tables were then examined for anomalies.
The following 4 questions were used to obtain information on perceived knowledge: “Does your health insurance cover (1) outpatient care (such as doctor visits); (2) inpatient care (including surgery); (3) prescription drugs; and (4) dental care?” Five questions verified participants’ objective knowledge: “Do you know (1) the amount of partial payment for seeing a general practitioner as an outpatient; (2) the maximum amount for visiting a doctor as an outpatient; (3) how much you have to pay for inpatient services in case you are admitted to a hospital; (4) how much you have to pay for prescription drugs; and (5) how much your health insurance covers for dental expenses?” For every “Yes” answer to each of the 5 questions, the follow-up question, “How much?” was asked for verification.
Participants entered a value in response
There have been 2 health insurance plans, an older and a modified one. Although there are similarities between them with regard to prescription drugs, dental, and inpatient services, there are also some differences. Therefore, we asked participants which plan they had, and checked the accuracy of their responses to the objective knowledge questions against that plan. In addition, although the employer has the option of reducing the maximum copayments, these maxima are predetermined so as not to exceed 20% of the health care expenses or a maximum of SAR 100 (USD 26.67) in both the old and new plans. However, with regard to the minimum copayment, which is optional (in the new plan) based on their insurance to the employers, the participants were asked to identify their insurance classes and benefits packages, although the majority had essentially the same minimum benefits package offered in both the old and new insurance plans. However, for those who were unclear about their copayment options, a trained bilingual research assistant helped them verify this information, either by using their insurance card or by checking with the company. This assistance was important for minimizing the confusion that the participants may have had with regard to the old and new plans.
Both perceived and objective knowledge questions were scored as follows: Incorrect responses were assigned 0 points and correct responses 1 point. Responses that required participants to enter a value were accepted as correct if it came within 10% of the correct answer, as has been reported elsewhere.22 The total number of points constituted the knowledge score, with a range of 0 to 9 points (4 points available for perceived knowledge and 5 points for objective knowledge).
Data for the qualitative variables were summarized as frequencies and percentages. For knowledge variables, frequency and percentage of each score were computed. In addition, the mean score, standard deviation, median, and interquartile range were calculated for each knowledge score. Normality of the 3 knowledge scores was examined using the general rule of equality of the mean, median, and mode and considering the histogram of the knowledge score. Using this approach, we concluded that the knowledge scores were reasonably normal and would therefore be analyzed with parametric methods. To evaluate the demographic and workplace variables associated with levels of knowledge, one-way ANOVA was used. For each knowledge variable, multiple linear regression analysis was performed to examine the joint impact of demographic and job characteristic variables on scores. One-way ANOVA was used for univariate analysis because there was no need to create dummy variables. More output is produced by ANOVA than by regression, e.g., means for each level of independent variable and confidence intervals.
The results of the regression were summarized as regression confidence intervals with their standard errors. Adjusted R-squared (R2) was used as a measure of the overall model fit. All tests of significance were two-sided and p-values <0.05 were considered statistically significant. All analyses were performed with SPSS, version 23 with the exception that multiple linear regression was fitted using Stata Release 14.0 (StataCorp., College Station, TX, USA) because of its ease of creating dummy variables.
Results
Table 1 indicates that the majority of expatriates in our study were young (90.% under 45 years of age), less educated (46.5% had elementary-level education or lower), married (84.4%), from Asian countries (77.4%), comfortable speaking Arabic (81.2%), and able to converse in English (54.1%).
Personal characteristics of expatriates workers in the study group. N=3398
Table 2 shows that more than half of the expatriates interviewed worked in a large company (55.5%), had been working in Saudi Arabia for more than 5 years (57.3%), and occupied positions that required an education level lower than high school or no education at all (54.3%). The table also indicates that more than one-third (36%) of respondents worked in the construction sector, and most (80%) earned a salary below SAR 4500. Most participants (86.5%) were either in excellent or very good health, and 62.2% reported having previous knowledge about health insurance. Table 2 shows that most expatriates knew about their coverage for outpatient services (87.4%), inpatient services (62.9%), prescription drugs (86.4%), and dental care (62%).
Job characteristics of expatriates workers and their previous knowledge about health insurance. N=3398
Appendices 1 & 2 show that the variables associated with perceived, verified, and overall knowledge were similar. The Appendices also indicate that most personal and workplace characteristics were significantly associated with the 3 measures of knowledge of health insurance benefits (perceived, verified and overall knowledge). Age, educational level, marital status, nationality, native language, comfort with conversing in Arabic, length of time in Saudi Arabia, company size, economic sector, income, access to health insurance, and previous knowledge on health insurance had significant associations with perceived, verified, and overall knowledge of health insurance benefits. One interesting study result is that both the highest educational levels and the education level required to perform their jobs were significantly associated with expatriates’ knowledge of health insurance benefits. The higher the educational level, the higher the knowledge score. The mean overall knowledge score of respondents with higher levels of education (high school graduate and above) was 4.6±2.4 compared with those respondents who had no schooling with scores of 2.9±1.6, or elementary with scores of 3.8±1.8 or high school education with scores of 3.5±1.8. Appendices 1 & 2 show the significant associations between required education level and knowledge of health insurance. The higher the required education level, the higher the knowledge score. The mean overall knowledge score of expatriates who were specialists with a university-level education was 5.2±2.8.
A significant association was noted between company size and knowledge of health insurance benefits. The mean score on overall knowledge for those who worked at a company with more than 50 employees was 4.48±2.04, whereas for those who worked at a company with fewer than 50 employees, the score was 3.59±2.12. Respondents who reported previous knowledge about health insurance were more likely to have better overall knowledge about their current health insurance benefits (mean 4.3±2.16) than were those who reported they did not have previous knowledge (mean 3.7±2)
Table 3 shows the percentage of expatriates who believed they knew the copayments for general practitioner visits (66.9%), outpatient visits (inclusive) (71.2%), inpatient visits (42.2%), prescription drugs (81.6%), and dental services (43.4%). However, the percentage of respondents who actually knew the correct copayments was considerably different in most cases: general practitioner visits (7.5%), outpatient visits (64.4%), inpatient visits (13.3%), prescriptions (15.4%), and dental services (9.6%). Of those who claimed to know the copayment amounts, the percentage who gave incorrect responses were 11.3% for general practitioner, 90% for outpatient, and 31.6% for inpatient visits; 19% for prescription drugs; and 22% for dental services.
Descriptive analysis of validated knowledge about copayments. N=3398
To study the joint effect of all independent variables on knowledge scores (perceived, objective and overall), a multiple linear regression model was fit for each score. Results of the regression analyses are presented in parts 1& 2 in Table 4. In Table 4, part 1 represents the personal characteristics of workers and part 2 represents workplace characteristics of workers. In Table 4, part 2 shows that the adjusted R2 of model fit for perceived knowledge was 0.43, indicating that nearly 43% of the variation in perceived knowledge can be explained by all of the independent variables. For objective knowledge, the adjusted R2 was similar at 0.44, whereas for overall knowledge R2 was larger at 0.55.
Regression analysis results for perceived, verified and overall knowledge of health insurance for the personal characteristics of workers.
Part 1 and part 2 (Table 4) also show the regression coefficients and their standard errors, as well as p-values, for all variables. Most variables retained a significant association with perceived knowledge, as seen in the univariate analysis, except for native language and comfort level conversing in English. Since all variables were categorical, a reference group was chosen for each variable. For the variables reflecting attained educational level and required educational level of the job, the category of no (required) education was chosen as the reference. Only the category of positions requiring a university degree were significantly different from the reference. There was no significant difference between positions requiring less than a high school education and those that did not require any education. With respect to expatriates with higher levels of education, those with a high school diploma or above were the only category that differed significantly from the reference.
Interestingly previous knowledge of health insurance had no significant impact on verified knowledge (p=0.154), but was significantly associated with both perceived and overall knowledge.
Discussion
This study assesses expatriates’ knowledge of health insurance benefits by analyzing their self-reported knowledge, objectively verified knowledge about copayments, and the combination of perceived and objective knowledge.
There are similarities between our sample and the expatriate population in Saudi Arabia, as reported by the Ministry of Labor.17 For example, the average (and median) age, the 4 most common nationalities (Indian, Pakistani, Egyptian, and Bengali), and the 3 most common economic sectors are consistent with Ministry data, although some economic sectors, such as health care and health insurance, were excluded in this study.
This study found that participants were more likely to claim knowledge of insurance coverage for outpatient services and prescription drugs, with almost equal percentages (approximately 87%). However, Paez and colleagues23 found that less than 60% of people knew about their health care coverage before attending health care facilities. Benedetti et al24 reported that half of study participants knew their copayment amounts. Our study finding of 64.1% correct responses to objective knowledge items about copayments for outpatient visits (inclusive) is in line with these previous studies.
Variations between what people think they know and what they truly know also support the findings of other studies in which people tended to overestimate their knowledge on health insurance coverage.23,24 Paez et al23 found that 3 out of 4 participants think they know how to use health insurance, and the authors estimated that only one in 5 people were capable of calculating their out-of-pocket costs; however, our study found that these percentages differed from one service to another. Among those who believed they knew their health insurance copayments, we found variable percentages of those who actually knew the correct copayment for general practitioner (11%), outpatient (90%) and inpatient visits (31.6%), prescription drugs (19%) and dental services (22%). Moreover, a gap was noted between participants’ estimation of their copayments and the actual values, although a 10% margin was allowed. This gap supports the importance of verifying knowledge rather than relying on self-report.24,23
Our study revealed that knowledge with respect to dental services was low. Only 22% of respondents knew the maximum coverage for these services. There have been few studies about knowledge of dental care coverage because these services are excluded from many international health insurance plans.15 In the United States, Medicare and Medicaid programs do not cover most dental services.16
Study participants had high self-reported knowledge on prescription drug coverage (86.4%). However, their verified knowledge of copayments for medication was very low. People tended to overestimate their knowledge on prescription drugs more than any other service. This large variation between respondents’ perceived (81.6%) and actual (19%) knowledge is supported by previous research in which participants were shown to lack specific knowledge about their prescription drug copayments.24
Our study findings are inconsistent with those of other studies showing that people are well informed about their coverage for inpatient services relative to outpatient coverage.22 Our study indicates that self-reported knowledge was low for both inpatient services and copayments compared with knowledge about outpatient coverage. The expatriate population in Saudi Arabia is young and their need for major surgery is low, contrary to that among older adults covered by Medicare or low-income individuals enrolled in Medicaid. Moreover, expatriates are required to undertake 2 medical exams prior to beginning work in Saudi Arabia, thereby reducing the likelihood of major surgery.
We found not only a significant association between expatriates’ education levels and their knowledge of health insurance benefits but also an association between this knowledge and the education required to perform their jobs, a finding that has not been reported previously. The importance of educational requirements for employment may indicate that employers treat employees differently based on the importance of their position, from the employer’s perspectives. The role of employers in access to health insurance, particularly for minorities, cannot be underestimated, as has been asserted by others.25,26 Our findings are in agreement with other research reporting that minority workers in large companies have better knowledge of their health insurance benefits compared with those working in small companies.27 This finding can be explained by the fact that people obtain their knowledge about health insurance benefits from their employers as a first source,28 and larger employers have better methods of informing their employees about these benefits.
As with most research, our study has limitations. The cross-sectional study design limits our ability to derive causal inferences.29 Although the linear regression analyses increase our confidence on the association between different variables and expatriates’ knowledge, the current regression analysis cannot explain more than 55% of expatriates’ knowledge. Therefore, it is possible that influencing variables are missing such as a negative attitude toward health insurance from a religious perspective,7 differences in health insurance benefits packages,30 and the absence of basic familiarity with program terminology.31 In addition, our study did not consider that there are variations among health insurance companies with respect to informing insured participants.
Another study limitation was the fact that the sample included only males. Female expatriates and children were excluded because the number of female employees in the private sector is small (98.30% of private sector expatriates are male), thus making it difficult to obtain a sufficient sample size. Most women work in the health care sector, which was excluded from the study sample, as stated above. Moreover, not all expatriates’ families were covered under CEBHI during the study period.
Last, this study measured knowledge of health insurance benefits from the employee’s point of view. Further studies are needed to assess this knowledge from the points of view of both employers and health insurance companies because these are the main sources of information on health insurance.18,28 Moreover, further study is needed to assess whether knowledge of insurance benefits has implications for expatriates’ access to health care.
Conclusion and recommendation
In conclusion, many expatriates face challenges in understanding how health insurance works and how to estimate out-of-pocket costs. Knowledge of health insurance benefits is very limited among expatriates, with most tending to overestimate their knowledge. The factors influencing this knowledge are very complex. Our study outcomes raise questions about the effectiveness of health insurance companies in communicating with their customers. The CCHI website is only available in Arabic and English. Most expatriates speak and write in Urdu, Hindi, Malayalam, and Bengali. Expatriates who can speak Arabic and English have better familiarity with their health insurance benefits. Therefore, health insurance benefits and other insurance policy information should be made available in other languages. In addition, it should be mandated that all health insurance company policies as well as expatriates’ work contracts, which describe their right to health insurance, be translated into employees’ native languages. The CCHI must develop a national policy to help increase expatriates’ knowledge about health insurance benefits in collaboration with health insurance companies, employers, and government agencies such as the Ministry of Labor.
Acknowledgment
I would like to express my gratitude to the research assistant team and data entry team. Without their assistance, this project would have been difficult to complete. I would also like to thank Dr. Gamal Mohmed for his serving as scientific advisers and critically reviewed the study proposal and Mr. Ali Al kahtani, Head of Researches and Studies Department at Council of Cooperative Health Insurance, for serving as scientific adviser for health insurance benefits plan knowledge. Also, Ms. Ferlyn Nopre for her assistant in manuscript alignments and editing.
Appendix 1 - Result of one-way Analysis of Variance for perceived, verified, and overall knowledge of health insurance and workers personal characteristics (N=3398)
Appendix 2 - Result of one-way ANOVA for Perceived, Verified, and Overall Knowledge of Health Insurance and job/workplace characteristics.
Footnotes
Disclosure. Authors have no conflict of interests and this work was funded by King Abdullah International Medical Research Centre, Riyadh, Kingdom of Saudi Arabia (Grant number: RC13-193).
- Received December 12, 2016.
- Accepted March 15, 2017.
- Copyright: © Saudi Medical Journal
This is an open-access article distributed under the terms of the Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
In this issue
Jump to section
- Article
- Abstract
- Methods
- Results
- Discussion
- Acknowledgment
- Appendix 1 - Result of one-way Analysis of Variance for perceived, verified, and overall knowledge of health insurance and workers personal characteristics (N=3398)
- Appendix 2 - Result of one-way ANOVA for Perceived, Verified, and Overall Knowledge of Health Insurance and job/workplace characteristics.
- Footnotes
- References
- Figures & Data
- eLetters
- References
- Info & Metrics
Related Articles
Cited By...
- No citing articles found.