Introduction

Immigrants have been found to be uninsured, and their cash expenditures for health care often have been higher than those of individuals with insurance, resulting in a further reduced ability to pay for the care they needed (Selden and Sing 2008). Other factors such as religious barriers may also affect immigrants’ healthcare access. The outcome is that some immigrants use fewer primary, preventive, and emergency medical services when compared to US citizens, even when the effects of race, income, and insurance status were controlled (Howie 2009).

An example of the medical challenges confronting immigrants is the Ghanaian population which has settled in New York City (Thompson 2009). The immigrants formed a community identity group called the Bronx Ghanaian Immigrant Muslim Community (BGIMC) which was established to provide Arabic and Islamic education and officiants to conduct traditional ceremonies for marriages and bereavement in conformity with Islamic doctrine (The Bronx Ghanaian Immigrant Muslim Community Constitution 1990).

Attention to healthcare disparities has been largely focused on race and ethnicity, and data regarding the dimensions of disparities were relatively widely available. In contrast, data on the West African immigrants health disparities are less common (Snowden and Fawley 2008). Therefore, our purpose was to determine the extent that the religion-based BGIMC members’ education, immigration status, and health insurance status predict their perceived access to healthcare services and their willingness to use healthcare services.

Methods

Research Design

The hierarchical multiple regressions were determined to be appropriate for the study because it dealt with cause and effect relationships between variables. Further, random assignment of groups was employed, especially in the strongest experimental research designs (Creswell 2008).

The hierarchical multiple regressions used in the study evaluated the relationship between independent and dependent variables by taking into account how different variables impacted on the dependent variables (Freeman 2005). The study evaluated relationships between the dependent variables, i.e., perceived healthcare access and a willingness to use health services, and the independent variables, i.e., gender, age, immigrant status, health insurance status, and educational background of the participants.

Participants

A systematic sampling approach was employed to obtain the participants. Every third member to exit the BGIMC premises was asked to first complete three qualifying questions. Both men and women exited the premises at the same point. When they passed the qualifying questions, they were asked to engage in the research. The qualifying questions were as follows:

  • Are you 18 years or older?

  • Have you lived in the United States for at least a year?

  • Do you agree to participate in the study?

Those who answered “Yes” to all three questions participated in the study and were required to read an informed consent form. Participants were recruited on a voluntary basis, were fully informed about the purpose of the study and its importance to the BGIMC, and understood that they could end their participation at any time.

G*Power was used to arrive at the minimum sample size for the hierarchical multiple regression (Erdfelder et al. 1996). Specifically, based on the assumption that the hierarchical regression will have three predictors, 0.15 effect size (medium effect), an alpha level of 0.05, and power of 0.95, the minimum sample size for this analysis was 119. The resulting sample size was 156 members from the BGIMC.

The respondents consisted of Bronx Ghanaian Muslim immigrants who had lived in the USA for at least 1 year. The respondents consisted of 54.5 % males and 45.5 % females. The average age of all respondents was 45.99 years (SD = 13.27). A third of the respondents (33.3 %) were high school graduates, while 18.6 % had at least a bachelor’s degree. Table 1 contains demographic information on the respondents.

Table 1 Frequencies: demographics (N = 156)
Full size table

Data Collection

One of the authors performed all stages of data collection, data analysis, and reporting. The survey instrument was provided in English and Hausa languages, and an author was in charge of translating the documents to respondents who needed help. The facility maintained by the BGIMC in the Bronx, New York, was the primary venue for data collection. Participant recruitment took place during community meetings, special events such as weddings and child-naming ceremonies, and at Friday and Sunday prayer sessions. Prior to the commencement of recruitment, one of the authors shared information about the study to BGIMC members during their biweekly meetings.

Instrumentation

An appropriate instrument specific to the issues of African immigrants was not found. Therefore, a new questionnaire was developed based on a focus group study of 63 Ghanaian immigrants living in New York City (Kaplan et al. 2015). Permission to adapt the focus group moderator guide was obtained. Data regarding barriers to health care from Kaplan’s research were analyzed and used to develop the survey instrument.

The instrument measured the relationship between demographic variables, insurance status variable, variables related to religious beliefs, perceived access to health care, and willingness to use health care. A test of reliability (test–retest) and an internal consistency instrument were conducted to evaluate the new instrument.

The survey instrument consisted of three separate sections, and a pilot study involving a small sample of 25 respondents was conducted to verify the content validity of the instrument.

Validity and Reliability

Expert judgment of three experts and the Content Validity Index scale were used (Wynd at al. 2003). The criteria for being an expert were as follows:

  • Someone currently working in the health field and has been in the field for at least 10 years.

  • Completion of a master’s degree from accredited college as a healthcare provider or health educator.

  • Lives in the Bronx Ghanaian Immigrant Muslim Community (BGIMC) and shares the same religious values.

  • Lived in the USA for more than 5 years.

Pilot Study

After the initial survey instrument was developed, a pilot study was conducted with 25 BGIMC members at the BGIMC facility in New York after obtaining approval from the Walden University Institutional Review Board. Nineteen males (76 %) and six females (24 %) participated in the pilot study, and 48 % had less than a high school education. The participants were at least 18 year old and had lived for more than a year in the USA in the BGIMC. The pilot study determined the clarity of the questions (Leedy and Ormrod 2013). Participants were not offered any incentive for completing the survey, and the entire survey procedure took approximately 20 min. The participants were instructed to refrain from discussion during the introduction of the study and completion of the survey to avoid a diffusion of opinions (Wilde et al. 1994). Internal consistency reliability of the survey was assessed based on feedback from the 25 participants, and no items were removed.

Most participants completed the survey instrument within 20 min. However, participants were under no time limitation. Data collection occurred on 6 days when the respondents were leaving the BGIMC premises.

One of the authors transferred responses from the survey instrument onto an Excel file and then onto the SPSS program for analysis. To minimize transfer error, the design of the Excel data sheet was based on the format of the survey instrument.

Researcher Bias

The author who collected and analyzed the data recognized the potential for biased results. Several measures dealt with the possible researcher subjectivity. Specifically, the author was constantly aware of bias and sought to be impartial throughout the study. The second author, who was not associated with the BGIMC, reviewed the data and analysis.

Data Analysis

All data were analyzed in terms of how each survey question addressed the research questions. Descriptive statistics was performed on the demographics of the respondents, including gender, age, immigration status, length of stay in the USA, and health insurance status. Logistic regression was conducted to evaluate the research questions and hypotheses.

In these analyses, perceived access to use health care and willingness to use health care were the dependent variables and education, immigration status, health insurance status, and religious beliefs were the independent variables. Specifically, the regression indicated which independent variables made a significant contribution to predicting an individual’s perceived access and willingness to use health care, along with the explanatory power of the significant independent variables.

Results

Predictors of Access to Health Care

To examine whether education level, immigration status, and health insurance status were predictors of access to health care, a logistic regression was conducted. As shown in Table 2, only one of the variables, health insurance status, made a unique statistically significant contribution to the model, having a p value of .004 and recording an odds ratio of 9.25. This indicated that those with insurance were nine times more likely to report they had access to health care than those who did not have insurance.

Table 2 Logistic regression predicting perceived access to health care
Full size table

To determine whether education, immigration status, and health insurance status were significant predictors of healthcare usage in the past 12 months, a logistic regression was performed. As shown in Table 3, health insurance status was the only independent variable that made a statistically significant contribution to the model, having an p value of .027 and odds ratio of 6.84. This indicated that those with health insurance were almost seven times more likely to report using healthcare services in the past 12 months.

Table 3 Logistic regression predicting healthcare usage
Full size table

Results of the multiple linear regressions indicated that immigration status, health insurance status, and education levels did not predict willingness to use health care for a broken arm, nor did they predict willingness to use health care for severe fever. However, immigration status, health insurance status, and education levels were able to predict willingness to use health care for dizziness.

Discussion

The authors investigated the influence of religious beliefs on BGIMC members’ perceived access and willingness to use healthcare services. It was designed to determine the extent to which the following factors influence the use of healthcare services: education, immigration status, religious beliefs, and insurance coverage. The results of the logistic regressions indicated that those with insurance were nine times more likely to report that they had access to health care than those who did not have insurance. Additionally, those with health insurance were almost seven times more likely to report using healthcare services in the past 12 months.

Further, the results of the multiple linear regressions indicated that immigration status, health insurance status, and education levels did not predict willingness to use health care for a broken arm or severe fever. However, immigration status, health insurance status, and education levels were able to predict willingness to use health care for dizziness. Finally, results indicated that religious beliefs were not a predictor of willingness to use health care.

New Contribution to the Literature

These findings indicate that there is a need for health insurance coverage for the members of the BGIMC and supports other research that immigrants working in industries such as agriculture, construction, and services did not have access to health care (National Academy of Sciences 2009). Further, 55 % of noncitizens are employed in firms with <100 employees and that these firms often do not offer health insurance. The result is that immigrants cannot afford to purchase private insurance because of cost which hinders their use of healthcare services (Siman 2009).

Significance of the Study

The findings can be applied to other immigrant groups and healthcare providers to develop recommendations and solutions for promoting better healthcare access and utilization. Although the findings may not be generalized to all immigrants in the USA, the results may be applicable to African immigrant communities from countries like Nigeria, Senegal, and other West African countries whose citizens have similar ethnic and religious similarities.

Recommendations for Action

The increased interest in health insurance reform across the country presents a timely opportunity to address health problems experienced by many immigrants. These reforms, however, will help the African population only if their health needs and other related community issues are taken into account. The following proposals serve as a core set of recommendations for action and further research.

Immigrant communities are affected by a lack of health insurance coverage, and that triggers high healthcare costs and poor-quality treatments (National Academy of Sciences 2009). Although healthcare reform aims at improving the healthcare system by providing equal access to affordable health coverage, the authors have found that actions are required to address the health-related issues of the BGIMC members with priority to increasing health insurance coverage.

Because immigrants are more likely to live below the poverty line, healthcare affordability is a serious concern (Siman 2009). Policy makers should make improving access to health care a priority by developing income-based standards for all health-related costs in addition to subsidies for people with low incomes. Specifically, the authors recommend that the BGIMC should coordinate with the community’s healthcare providers to take advantage of the affordable health insurance programs and other government subsidies and educational programs for immigrants. For example, as noted by the National Academy of Sciences (2009), there should be community-based outreach health promotion and prevention programs to address healthcare issues. Further, hospitals should address the health issues of the communities they serve by developing criteria for Disproportionate Share Hospital payments (DSH) and maintaining the tax-exempt status to benefit the community (Rittenhouse et al. 2009). Community-based programs should be evaluated periodically to ensure they address the health needs of immigrant communities and to include health insurance coverage to everyone.

A community health education promotion kit should be developed by the Ghanaian community leaders to assist in health education and health promotion strategies within the community. The kit will educate members on the process of attaining legal residence status to assist them in finding employments that provide health insurance benefits in order to have adequate access to health care. A community radio should be utilized to reach out to all community members to disseminate community health enrichment information in order to take advantage of these programs to improve their lives.

Recommendations for Further Study

Further research is necessary to understand and alleviate health disparities among immigrants in the USA. Immigrant families encounter problems using healthcare services for many reasons. Among them are a lack of religiously competent healthcare providers who understand and address their healthcare issues effectively, affordability of treatment costs, perceptions of lack of respect by healthcare providers, complexity of our healthcare system, and a lack of health insurance (Vaughn et al. 2009).

In addition, future research should study other groups to determine the effect of religious beliefs and a willingness to use healthcare services. For example, the authors studied only Ghanaian Muslims who were affiliated with the BGIMC. Therefore, the Ghanaian Muslims whose places of worship were different from the BGIMC members were not included. Also, non-Muslim Ghanaian immigrants in the Bronx were not included in the study.

Conclusion

The authors found that religious beliefs did not appear to predict willingness to use health services. Instead, respondents with insurance were nine times more likely to report that they had access to health care than those who did not have insurance. Further, those with health insurance were almost seven times more likely to report using healthcare services in the past 12 months. Religion, immigration status, health insurance status, and education levels did not predict willingness to use health care for a broken arm, nor did they predict willingness to use health care when someone had a severe fever. However, immigration status, health insurance status, and education levels were able to predict willingness to use health care when someone experienced dizziness. In summary, results indicated that religious beliefs were not a predictor of willingness to use health care.

There is a clear need to understand better how to ensure access to healthcare services and to deliver appropriate care to immigrants of all religions. A collective effort will be required to address the healthcare needs of immigrants.