RURAL residence is a noteworthy social determinant of poor health outcomes, both in the United States and internationally.1–3 Residents of rural areas are more likely to report fair to poor health than urban individuals (19.5% vs 15.6%).3,4 Furthermore, residents of rural areas tend to have higher rates of obesity, diabetes, hypertension, high cholesterol level, and asthma than their urban counterparts.2–8 In comparison with their urban counterparts, rural residents are older, poorer, and sicker (more affected by chronic health conditions).7,9
In Florida (where the present study was conducted), rural counties fare worse than urban ones in measures of morbidity and mortality, engagement in health-promoting behaviors, and health care access and utilization.10 Rural residents in Florida, for example, have high rates of diabetes and arthritis when compared with national data.11 There are disproportionately higher death rates in rural areas than in urban areas for 20 of the 25 leading causes of death.12
When compared with urban residents, residents of rural areas, in particular racial/ethnic minorities, face a unique combination of sociodemographic variables (eg, lower household incomes, higher rates of poverty, lower rates of health insurance, lower levels of education) that put them at higher risk for morbidity and mortality than urban residents.3,7,9,11 Factors such as inadequate health facilities, shortage of health care providers, and lower access to preventive and specialized health care, which have a negative impact on residents of rural areas, also lead to health disparities.2,3,12,13 In sum, rural communities have sociodemographic characteristics that help explain the health disparities they experience.14–16 In face of these demographic disadvantages, rural residents must commit to consistently engaging in behavioral health practices such as treatment adherence and engagement in health-promoting behaviors (eg, exercising and healthy eating) to stay healthy.
However, rural patients have lower rates of treatment adherence and engagement in health-promoting behaviors than their urban counterparts.17,18 Exact rates of treatment adherence are inconsistently depicted in the health care literature.19 However, research consistently shows that rural individuals' demographic characteristics (eg, lower levels of education and literacy) may put them at risk for lower treatment adherence and therefore at increased risk for poor health outcomes.
Rural residents also have lower rates of engagement in health-promoting behaviors than their urban counterparts. Rural residents have a higher percentage of sedentary lifestyle than urban residents.11,20–22 Leisure time inactivity is most common for men and women in rural counties.7 Residents of rural areas are less likely to have nutritional diets and more likely to be current or former smokers than their urban counterparts.3,7,23–25 The largest urban-rural increases in smoking are seen in the south of the United States.7
PRECURSORS OF TREATMENT ADHERENCE AND ENGAGEMENT IN HEALTH-PROMOTING BEHAVIORS
Health Self-Empowerment Theory (HSET) is a literature-based theory that is useful in understanding treatment adherence and engagement in health-promoting behaviors.26,27 HSET recognizes the effect of social, environmental, and economic conditions on health behaviors.26,27 These conditions could partially explain the differences in health outcomes between residents of rural areas and residents of urban areas. Yet, given that many of these variables are intractable, research is needed that identifies modifiable psychological and knowledge variables that may empower rural individuals to engage in health-promoting behaviors (such as healthy eating and physical activity) and to adhere to treatment under whatever environmental, cultural, social, and economic conditions that may exist in their lives.
HSET asserts that self-empowerment–oriented, cognitive-behavioral variables (ie, motivation to engage in health-promoting behaviors such as eating a healthy diet and exercising, self-praise of health-promoting behaviors, active coping strategies/skills for managing stress, taking responsibility for one's health/health responsibility, health knowledge, and health self-efficacy) influence the occurrence of health-promoting behaviors.26 These variables become key to understanding and modifying the health behaviors of racial/ethnic minorities and individuals with low incomes and limited health resources (such as minorities residing in rural areas), who often have low actual or perceived power over their health and many other aspects of their lives.27
Health self-efficacy is one of the self-empowerment–oriented variables included in HSET. In urban individuals, health self-efficacy (ie, one's perceived capability of engaging in mental and physical health-promoting behaviors and healthy lifestyles, and the expectations that personal effort can lead to these healthy behaviors and a healthy lifestyle)26,28,29 has been identified as a determinant of both treatment adherence30 and engagement in health-promoting behaviors such as engagement in physical activity, health eating, and smoking.26,31–33 Literature on these associations is scarce for rural patients.
PRESENT STUDY HYPOTHESIS
Given the research linking health self-efficacy to engagement in health-promoting behaviors and treatment adherence mostly with urban individuals, this study examines the links between health self-efficacy and treatment adherence and engagement in health-promoting behaviors in a sample of adult rural patients in North Florida. This study intentionally includes an overrepresentation of racial/ethnic minorities (in particular, African American/black individuals), as they are most affected by health disparities in the area where this study was conducted. Using a cross-sectional design, the following research hypothesis was investigated: health self-efficacy will predict levels of treatment adherence and engagement in health-promoting behaviors (ie, individuals with lower health self-efficacy will exhibit lower rates of treatment adherence and engagement in health-promoting behaviors; Figure 1).
Participants in this study were a convenience sample of 273 patients from 2 clinics in North Central Florida. Both clinics serve predominately indigent and low-income rural patients. To be enrolled in the proposed study, patients had to (a) be at least 18 years old, (b) be patients at one of these 2 health care centers in the 12 months prior to the study, (c) be able to communicate either verbally or in writing in English or in Spanish, and (d) be able to read and sign an informed consent form that documents agreement to participate in the study.
The race/ethnicity distribution among participants was as follows: 22 (8.1%) self-identified as Hispanic; 3 (1.1%) as American Indian or Alaska Native; 90 (33.0%) as black or African American; 129 (47.3%) as Caucasian/white/European American; and 22 (8.1%) as other. Of the 273 participants, 175 (64.1%) self-identified as female, 73 (26.7%) self-identified as male, and 25 (9.2%) did not report their sex. Participants were primarily low income. For additional demographic information, see Table 1.
The sample included an overrepresentation of individuals who self-identified as African American/black, given that (1) they are the largest racial/ethnic minority in North Florida, in particular in the area where the study was conducted (with a population of 21.7% black vs 57.2% white vs 10% Hispanic)34; and (2) they are the racial/ethnic minority group with the highest rate of overweight and obesity (70.3% vs 60.6% in non-Latino whites) in Florida.35 Females were overrepresented in this sample (64.1% vs 26.7% males).
Participation involved anonymously completing a research participation packet that included (1) 2 copies of the informed consent form—one for participants to keep and one for the researchers to keep, and (2) an assessment battery. The following 4 questionnaires were used in this study: (a) Demographic Data Questionnaire; (b) Self-Rated Abilities for Health Practices Scale; (c) Health Promoting Lifestyle Profile II; and (d) General Adherence Measure.
The Demographic Data Questionnaire was constructed for the proposed study by the principal investigators. It was used to obtain information about each patient participant's sex, age, marital status, race/ethnicity, level of education, employment status, generation status, and household income.
The Self-Rated Abilities for Health Practices Scale36 is a 28-item inventory that assesses patients' self-perceived ability to implement health-promoting behaviors. The inventory contains 4 subscales: Exercise, Nutrition, Responsible Health Practice, and Psychological Well-being. Each subscale comprises 7 items. Instructions ask respondents to rate the extent to which they are able to perform health practices related to these 4 subscales (listed earlier). Sample items are as follows: “I am able to find healthy foods that are within my budget” (Nutrition), and “I am able to do exercises that are good for me” (Exercise). Items are scored on a 5-point Likert scale (with 0 = not at all, and 4 = completely). There are no reverse scored items. Total scores range from 0 to 112, with higher scores indicating higher self-efficacy for health practices. In a sample similar to the sample of the current study, the Self-Rated Abilities for Health Practices Scale demonstrated high reliability and validity.36 The Cronbach α for participants in this study was 0.94 for the total scale and 0.92 and 0.83 for the Exercise and Nutrition subscales, respectively.
The General Adherence Measure is a 5-item measure of treatment adherence, developed during the Medical Outcomes Study to assess patients' tendency to follow medical recommendations from their health care providers.37 The instructions on the General Adherence Measure asks participants to rate adherence to medical treatment in the prior 12 months using a 4-point Likert scale, where 1 = none of the time and 4 = all of the time. Sample items are as follows: “I had a hard time doing what my provider suggested I do,” and “I followed my provider's suggestions exactly.” Two of the items on the scale (items 1 and 3) are reversed scored. The 5 items on this measure can be averaged to yield a general adherence score. Higher scores mean higher treatment adherence. Internal consistency reliability for this scale is acceptable (Cronbach α = 0.81), and it has a 2-year stability of 0.41.37 The Cronbach α for this study's sample was 0.49. No particular item seemed to be driving this Cronbach α (which would increase only up to 0.57 by removing scale items).
The 52-item Health Promoting Lifestyle Profile II (S. N. Walker and D. M. Hill-Polerecky, Unpublished data, 1996) is a self-report inventory that assesses participants' level of engagement in an overall health-promoting lifestyle. Participants are asked to indicate how frequently they engage in specific health-promoting behaviors (eg, “choose a diet low in fat, saturated fat, and cholesterol” and “follow a planned exercise program”). Items are rated on a 4-point Likert scale (1 = never to 4 = routinely). Higher scores indicate a lifestyle with self-reported higher health-promoting behaviors. The instrument has 6 different subscales. Healthy eating (nutrition) and physical activity were used in the present study. Walker and Hill-Polerecky (Unpublished data, 1996) have reported a Cronbach α of 0.94 for the overall measure. The Cronbach α for the overall measure for participants in this study was 0.91.
Before the start of the study, the 2 principal investigators (coauthors of this study) met with the directors of the 2 identified rural health care clinics for the purpose of obtaining their permission to conduct the study at their respective health care clinics. Once this permission was obtained from the directors of the clinics and the institutional review board (IRB) at the university where principal investigators conducted this project, study implementation occurred in 3 phases.
Phase I: Training research team members
Prior to the launch of the study, the principal investigators trained undergraduate research assistants on the specifics of study implementation. This training lasted 1 hour. The study implementation training covered the following topics: (a) the purpose of the study; (b) potential benefits to patients due to participation in the study; (c) culturally sensitive strategies for recruiting culturally diverse, mostly rural, low-income individuals who may or may not speak English (eg, addressing patients with a title such as Mr or Mrs unless otherwise requested as a sign of respect, speaking assertively but slowly, etc); and (d) culturally sensitive strategies for collecting data from culturally diverse adults (eg, administering the language-appropriate battery, assisting with reading and completing questionnaires as needed, allowing participants the time that they need to complete the questionnaires, encouraging participants to take breaks when completing the assessment battery, and pleasantly answering any questions that may come up for participants). This training encompassed mock participant recruitment and data collection sessions that included having research assistants practice the learned study-related behaviors and skills with their peers and then ask questions to the principal investigators. The principal investigators observed these mock sessions and provided feedback to the trainees. The principal investigators also conducted external control monitoring during some of the actual recruitment and data collection sessions at the health care clinics.
Phase 2: Recruitment of patient participants
Once training was complete, the trained, culturally diverse research assistants met patients at the 2 aforementioned clinics during normal patient care office hours and invited them to participate in the study after they saw their health care providers. Research assistants approached patients and verbally explained the purpose of the study (ie, to understand what might influence their health outcomes and to use the results of the proposed study to develop interventions that help rural patients reach optimal health). Research assistants also handed patients a recruitment flyer that included the patient participation criteria, the purpose of the study, and the principal investigators' contact information. Research assistants explained to participants that they would receive a $15 visa gift card for enrolling in the study. This information was also included on the aforementioned recruitment flyer.
Research assistants explained participation criteria to patients. Patients who expressed interest in participating and met participation criteria were ask to read (or have someone read to them) the informed consent form (ICF) and then sign this form in front of a witness. Each patient participant was given a copy of the ICF to keep.
Phase 3: Data collection
Data collection took place in the waiting room of the clinics. After being enrolled in the study, participants completed the assessment battery, which took approximately 1 hour. Participants could involve the help of a trained research team member to complete the questionnaires, as needed. Payment forms and completed questionnaires were stored in separate envelopes. To protect patient confidentiality, no identifying information was written on the participants' assessment batteries.
To protect patients' confidentiality, ICFs and assessment batteries were kept separately during the data collection process and later in the principal investigators' laboratory. All data were processed in accordance with the ethical IRB standards at the university where the study was conducted. The overall study (data collection at the clinics) lasted 2 months.
The research team requested participating patients to help recruit additional patients for the study by asking adults that they knew who used either of the 2 participating health care clinics. Patients who agreed to invite other patients were given flyers to help with this recruitment.
Data from the measures of health self-efficacy, treatment adherence, and engagement in health promoting behaviors showed multivariate and univariate normality. In this way, it fit the assumptions of the General Linear Model and univariate and multivariate analyses. The means and standard deviations are presented in Table 2.
Bivariate correlations were conducted to examine the associations among the major variables of interest in this study among the total sample of patient participants. Results are presented in Table 3.
Analyses to test the study hypothesis
A structural equation model with maximum likelihood estimation (Figure 1) was conducted using SPSS AMOS 22 to investigate the study hypothesis: health self-efficacy will predict levels of treatment adherence and engagement in health-promoting behaviors (ie, individuals with lower health self-efficacy will exhibit lower rates of treatment adherence, healthy eating, and engagement in physical activity).
Overall, the model was a good fit. The χ2 test was nonsignificant and less than twice the degrees of freedom (= 3.080, P = .214), suggesting that the data do not significantly depart from the model. Furthermore, absolute and incremental fit indices were used as adjuncts to assess model fit. RMSEA = 0.045 pointed to excellent fit. Other indicators also showed excellent fit (CFI = 0.99; TLI = 0.97; IFI = 0.99; NFI = 0.98). Health self-efficacy significantly predicted levels of treatment adherence (β = .187, P = .005). In addition, health self-efficacy significantly predicted levels of engagement in health-promoting behaviors (β = .548, P < .001). This model explained 39.8% of the variance in health-promoting behaviors and 3.5% of the variance in treatment adherence (Figure 2).
The health disparities experienced by rural Americans have been extensively documented. However, while rural health disparities are one of 14 disparity concerns present in Healthy People 2020, efforts to support focused research to understand the nature of these disparities and possible avenues of repair have been limited.30 The present study responds to a need in the health care literature for research that attempts to understand what promotes healthier lifestyles among rural residents.
This study tested the hypothesis that health self-efficacy would predict levels of treatment adherence and engagement in health-promoting behaviors (ie, healthy eating and physical activity) in a group of culturally diverse patients in North Florida. Overall, the model was a good fit. Health self-efficacy significantly predicted levels of treatment adherence and engagement in health-promoting behaviors.
Interpretations and implications
Health self-efficacy is increasingly receiving recognition as a precursor to positive health behaviors. This study shows that research linking health self-efficacy, treatment adherence, and health-promoting behaviors in urban patients can, to an extent, be generalized to rural patients, too. When specific health practices are believed to lead to desired health outcomes but patients struggle to adjust their behavior, taking health self-efficacy into consideration is key.38
Boosting patients' health self-efficacy could potentially be a way of increasing their treatment adherence and engagement in healthy eating and physical activity and thus of improving their health outcomes. Specific suggestions for increasing patient's self-efficacy may include (a) breaking down the target behavior into smaller components; (b) coming up with a plan including specific behavioral strategies with the patient; (c) allowing patients to make their own choices (grounded on their cultural beliefs/practices and developmental level); and (d) giving patients' consistent, focused feedback.
Despite its several methodological strengths, this study has 4 main limitations. First, the participating health care clinics and individual participants were not randomly selected. The sample of individuals participating in this study was from only 2 rural health care clinics in North Central Florida. Thus, generalizability of findings from this study to other rural patients in other parts of Florida or the United States is limited. In addition, patients who participated were active in receiving health services (were targeted at clinics where they had received care for at least 1 year) and also expressed interest in participating. This form of participant self-selection may further limit generalizability of the present findings to patients who are accessing and actively utilizing health services (vs patients who may not be adherent enough to be in care or to agree to participate). The present study should be replicated with a larger and randomly selected sample.
Second, the measure of treatment adherence had relatively poor internal consistency. This may limit interpretation of results. Future studies may employ other measures of treatment adherence and even combine sources of adherence information (eg, medical record data) to overcome the limitations of any single approach. Multiple measures of adherence have been used in health care research (eg, self-reports, practitioner reports, physiological parameters), and each of these measures yields different nonadherence rates (even for the same participants in the same study).39 Because health care research on residents of rural areas, in particular racial/ethnic minorities, is limited, it is unclear what measures might be more appropriate for assessing treatment adherence in this population.
A third limitation is that, given the cross-sectional nature of this study, it is not possible to establish whether changes in the precursor set in this study (ie, health self-efficacy) lead to changes in levels of treatment adherence and engagement in health-promoting behaviors. Longitudinal research could provide a more reliable picture of the relationship between health self-efficacy and treatment adherence/engagement in health-promoting behaviors.
A final limitation of this study is that it used a mono-method approach to data collection, relying only on self-report measures. While self-report instruments have been found to be reliable in health care, they may encourage “socially desirable” responses in patients.39 Future research similar to the present study ideally should include data retrieved from multiple sources (eg, health care records of appointment keeping) and measures of social desirability.
The present study is an effort to meet the calls to investigate health disparities among culturally diverse rural patients—a population experiencing high rates of morbidity/mortality and typically underserved in health care and underrepresented in health care research. Rural residence is a social determinant of poor health outcomes and yet precursors of treatment adherence and engagement in health behaviors in rural patients are poorly understood.
Findings from this study show that health self-efficacy predicts engagement in health-promoting behaviors and adherence to treatment. For patients to engage in health-promoting behaviors such as healthy eating, they need to feel empowered to act on their knowledge and believe that their actions will bring about the desired results. Future research should concentrate on understanding specifically how to best meet the needs of patients with low health self-efficacy so that they too can engage in health-promoting behaviors and lead healthy lifestyles.
If the findings of future similar studies without the limitations of the present study provide support for the findings in the present study, the need to develop interventions to promote health efficacy among rural patients will be supported. Such research is critical for developing effective strategies to reduce health disparities that disproportionately impact racial/ethnic minorities in rural communities in the United States.
1. Crosby RA, Casey BR, Wendel ML, Mills LA, Vanderpool RC. Rural populations and health: determinants, disparities, and solutions. In: Crosby RA, Wendel ML, Vanderpool RC, Casey BR, eds. Rural Populations and Health: Health Determinants, Disparities, and Solutions. San Francisco, CA: Jossey-Bass; 2012:3–23.
2. Gamm LD, Hutchison LL, Dabney BJ, Dorsey AM. Rural Healthy People 2010: a companion document to Healthy People 2010. Volume 1. http://www.srph.tamhsc.edu/centers/rhp2010
. Published 2013. Accessed March 17, 2017.
3. Sparks PJ. Rural health
disparities. In: Kulcsar LJ, Curtis KJ, eds. International Handbook of Rural Demography. College Station, TX: Springer; 2012:255–271.
5. Carman AL, Scutchfield FD. Public health systems, health policy, and population-level prevention in rural America. In: Crosby RA, Wendel ML, Vanderpool RC, Casey BR, eds. Rural Populations and Health: Health Determinants, Disparities, and Solutions. San Francisco, CA: Jossey-Bass; 2012:3–23.
9. Agency for Healthcare Research and Quality (AHRQ). 2014 National Healthcare Quality and Disparities Report Chartbook on Rural Health
Care. Rockville, MD: US Department of Health and Human Services. http://www.ahrq.gov/research/findings/nhqrdr/index.html
. Published 2015. Accessed February 1, 2019.
11. Logan H, Guo Y, Dodd VJ, Muller K, Riley J III. The burden of chronic disease in a rural North Florida sample. BMC Public Health. 2013;13:906.
14. Beard JR, Tomaska N, Earnest A, Summerhayes R, Morgan G. Influence of socioeconomic and cultural factors on health. Aust J Rural Health
15. Bellamy GR, Bolin JN, Gamm LD. Rural healthy people. Fam Community Health. 2011;34:182–188.
16. Probst JC, Moore CG, Glover SH, Samuels ME. Person and place: the compounding effects of race/ethnicity and rurality on health. Am J Public Health. 2004;94:1695–1703.
17. Martin MY, Kim Y, Kratt P, et al Medication adherence among rural, low-income hypertensive adults: a randomized trial of a multimedia community-based intervention. Am J Health Promot. 2001;25(6):372–378.
18. Wurimi MW, Harpel T. Practicing preventive health: the underlying culture among low-income rural populations. J Rural Health
19. Cummings DM, Wu JR, Cene C, et al Perceived social standing, medication nonadherence, and systolic blood pressure in the rural south. J Rural Health
20. Joens-Matre RR, Welk GJ, Calabro MA, Russell DW, Nicklay E, Hensley LD. Rural-urban differences in physical activity, physical fitness, and overweight prevalence of children. J Rural Health
21. Patterson PD, Moore CG, Probst JC, Shinogle JA. Obesity and physical inactivity in rural America. J Rural Health
22. Pullen C, Noble Walker S, Flandt K. Determinants of health-promoting behaviors
in rural older women. Fam Community Health. 2001;24:49–72.
23. Hartley D. Rural health
disparities, population health, and rural culture. Am J Public Health. 2004;94:1675–1678.
24. Schootman M, Homan S, Weaver KE, Jeffe DB, Yun S. The health and welfare of rural and urban cancer survivors in Missouri. Prev Chronic Dis. 2013;10:130052. doi:10.5888/pcd10.130052.
26. Tucker CM, Butler AM, Loyuk IS, Desmond FF, Surrency SL. Predictors of a health-promoting lifestyle and behaviors among low-income African American mothers and white mothers of chronically ill children. J Natl Med Assoc. 2009;101:103–110.
27. Tucker CM, Butler AM, Kaye LB, et al Impact of a culturally sensitive health self-empowerment workshop series on health behaviors/lifestyles, BMI, and pressure of culturally diverse overweight/obese adults. Am J Lifestyle Med. 2014;8(2):122–132.
28. Bandura A. Self-efficacy: toward a unifying theory of behavioral change. Psychol Rev. 1997;84:191–215.
29. Bandura A. Guide for constructing self-efficacy scales. In: Pajares F, Urdan T, eds. Self-efficacy Beliefs of Adolescents. Vol 5. Greenwich, CT: Information Age Publishing; 2006:307–337.
30. Dunbar-Jacob J, Mortimer-Stephens MK. Treatment adherence
in chronic disease. J Clin Epidemiol. 2001;54(S1):S57–S60.
31. Bosworth HB. Improving Patient Treatment Adherence
: A Clinician's Guide. New York, NY: Springer; 2010.
33. O'Hea EL, Boudreaux ED, Jeffries SK, Carmack Taylor CL, Scarinci IC, Brantley PJ. Stage of change movement across three behaviors: the role of self-efficacy. Am J Health Promot. 2004;19:94–102.
36. Becker H, Stuifbergen A, Oh H, Hall S. Self-rated abilities for health practices: a health self-efficacy
measure. Health Values. 1993;17:42–50.
37. DiMatteo MR, Hays RD, Sherbourne CD. Adherence to cancer regimens: implications for treating the older patients. Oncology. 1992;6(2)(suppl):50–57.
38. Bolin JN, Bellamy GR, Ferdinand AO, et al Rural Healthy People 2020: new decade, same challenges. J Rural Health
39. Dunbar-Jacob J, Mortimer-Stephens MK. Treatment adherence
in chronic disease. J Clin Epidemiol. 2001;54(Suppl 1):S57–S60.