- Email: [email protected]
AIDS
JOGNN
IN FOCUS
Results from a Secondary Data Analysis Regarding Satisfaction with Health Care among African American Women Living with HIV/AIDS Jillian Lucas Baker, Caryn R. R. Rodgers, Zupenda M. Davis, Edward Gracely, and Lisa Bowleg
Correspondence Jillian Lucas Baker, DrPH, EdM, La Salle University, School of Nursing & Health Sciences, Department of Urban Public Health & Nutrition, 1900 W. Olney Ave. St. Benilde Tower, 3rd Floor, Office 3334, Philadelphia, PA 19144. [email protected]
ABSTRACT
Keywords African American women HIV AIDS health care providers patient satisfaction quantitative
Results: A majority of women reported satisfaction with medical services (88%, n = 140). Communication with health care providers, detectable viral load, education, income, self-reported health status, and sexual orientation were significantly bivariately associated with satisfaction with healthcare (all ps < .05). In the multivariate models, no variables significantly predicted satisfaction with healthcare.
Objective: To analyze satisfaction with health care among African American women living with HIV/AIDS. Design: Secondary analysis of baseline data of African American women who participated in Protect and Respect, a sexual risk reduction program for women living with HIV/AIDS Setting: HIV Care Clinic in an urban city in the northeast United States. Participants: One hundred fifty-seven (157) African American women living with HIV/AIDS. Methods: Regression analyses were used to examine the relationships between demographic variables, self-reported health characteristics, communication with health care providers, and satisfaction with health care provider.
Conclusion: Because satisfaction with health care can influence the quality of care received, health outcomes, and adherence to provider recommendations among patients living with HIV/AIDS, health care providers’ ability to elicit satisfaction from their patients is just as important as the services they provide. This project is one of the first studies to find high rates of satisfaction with health care among African American women living with HIV/AIDS. Further examination of satisfaction with health care among African American women living HIV/AIDS may help in narrowing health care disparities and negative treatment outcomes.
JOGNN, 43, 664-676; 2014. DOI: 10.1111/1552-6909.12491 Accepted May 2014
Jillian Lucas Baker, DrPH, EdM, is an assistant professor of public health, La Salle University, School of Nursing & Health Sciences, Department of Urban Public Health & Nutrition, Philadelphia, PA.
(Continued)
The authors report no conflict of interest or relevant financial relationships.
664
n the United States, one fourth of the population living with HIV infection are women (Centers for Disease Control and Prevention [CDC], 2013). African American women are disproportionally affected: 64% of new infections in women occur in African Americans, 18% in Whites, and 15% in Hispanic/Latinas (CDC, 2012a). Although the CDC (2001) reported a decrease in new incidence of HIV infections among African American women, 1 in 32 African American women will be diagnosed with HIV infection in their lifetimes.
I
African American women are disproportionately vulnerable not only to HIV infection but also to disparities in the health care system that affect the quality of care they receive and their access
to health care. Overall, African American women are less likely to use health care services despite need (Copeland, Scholle, & Binko, 2003). African American women also face particularly high rates of poverty, with one in four living in poverty (U.S. Department of Health & Human Services, 2010). Poverty is linked to lower quality and limited access to education and health care for African American women (El-Bassel, Caldeira, Ruglass, & Gilbert, 2009). In addition to these disparities, HIV-positive African Americans experience prejudicial treatment by health care providers (LaVeist & Carroll, 2002; Sohler, Li, & Cunningham, 2007), have delayed access to care (Cargill & Stone, 2005), and lower rates of antiretroviral adherence (Bogart, Bird, Walt, Delahanty, & Figler, 2004;
C 2014 AWHONN, the Association of Women’s Health, Obstetric and Neonatal Nurses
http://jognn.awhonn.org
Baker, J. L., Rodgers, C. R. R., Davis, Z. M., Gracely, E., and Bowleg, L.
Kleeberger, Buechner, & Palella, 2004) in comparison to Whites. Currently, several opportunities are missed for linking women with HIV to health care. Among HIVpositive women, 85% have been diagnosed, however, only 70% have been linked to care (Kaiser Family Foundation, 2013). Further, 41% are in regular care, 36% are prescribed antiretroviral therapy (ART), and 26%are virally suppressed (CDC, 2012b). An alarming concern is that these factors are associated with poorer clinical outcomes and higher mortality among African Americans living with HIV/AIDS (Bogart, Wagner, Galavan, & Klein, 2010; Wong, Shapiro, Boscardin, & Ettner, 2002). Consequently, African American women have the highest AIDS death rate (23.8 per 100,000 population) among all races/ethnicities, which is 22 times as high as the rate of death for Whites (1.1) and almost 5 times as high as the rate for Hispanics/Latinos (4.5) (CDC, 2011). Examining factors that result in treatment nonadherence is crucial for enhancing health care outcomes among African Americans living with HIV/AIDS (Bogart et al., 2010). The purpose of this study was to describe patient reports of satisfaction with healthcare among African American women living with HIV/AIDS (AAWLWH/A). In addition to reporting satisfaction with health care in general, determinants of satisfaction with health care were also examined (e.g., sensitive communication with health care providers and certain psychosocial and health characteristics).
Among women of color in the United States, African American women are disproportionately affected by HIV/AIDS.
volvement in health care decision making (Apollo, Golub, Wainberg, & Indyk, 2006; Marelich & Murphy, 2003). Variables that can influence patient satisfaction include race, socioeconomic status, cultural mistrust, experiences of racism and discrimination, trust in health care providers (Benkert, Peters, Clark, & Keeves-Foster, 2006), patient–provider communication (O’Malley & Forrest, 2002), and health status (Burke et al., 2003; O’Malley, Sheppard, Schwartz, & Mandelblatt, 2004). Overall, African Americans have been found to be less satisfied with their health care (Campbell, Ramsay, & Greene, 2001; LaVeist & Nuru-Jeter, 2002). However, existing literature regarding AAWLWH/A and their access to and participation in the health care system is extremely limited. In one study of dissatisfaction with health care among women living with HIV/AIDS, the authors found that women who reported being Latina, having a lower health status, not receiving ART, with depressive symptomatology, and no consistent health care provider were more likely to be dissatisfied with their health care (Burke et al., 2003). These factors are all relevant for the health outcomes of AAWLWH/A, as dissatisfaction with health care among PLWH/A can result in negative health outcomes such as treatment underutilization and discontinuity of care (Abel & Painter, 2003).
Satisfaction with Health Care among People Living with HIV/AIDS
Gaps in the Literature
Patient satisfaction research began in the 1960s (Hall & Dornan, 1988) and has been the most analyzed independent variable linked to health outcomes. However, patient satisfaction research was limited in the field of HIV due to confidentiality limitations, the short existence of the disease, and the stigma attached (Emlet, 2007). Thus, examination of patient satisfaction among persons living with HIV/AIDS (PLWH/A) is a very new field of inquiry. Some researchers have examined satisfaction or dissatisfaction with health care as a contributing factor to health care utilization and adherence to provider recommendations. Recent authors also found that satisfaction with health care has a positive effect on adherence to ART (Burke-Miller et al., 2006; Heckman, Catz, Heckman, Miller, & Kalichman, 2004; Schneider, Kaplan, Greenfield, Li, & Wilson, 2004) and in-
Although researchers have repeatedly demonstrated health disparities in access and quality of care overall affecting the African American community (Cohen, 2008; Institute of Medicine, 2003), few researchers have examined patient satisfaction specifically among AAWLWH/A. Some qualitative researchers have examined patient– provider interactions, social support, the role of the caregiver, and stigma among AAWLWH/A (Black & Miles, 2002). For example, the quality of the patient–provider relationship has been noted as influential toward the active participation of women living with HIV/AIDS in their own health care (Marelich & Murphy, 2003). In a quantitative study, Benkert and colleagues (2006) found that perceptions of racism and mistrust had a negative effect on satisfaction with health care among a sample of low-income African American
JOGNN 2014; Vol. 43, Issue 5
IN FOCUS
Caryn R. R. Rodgers, PhD, is an assistant professor at the Preventive Intervention Research Center, Albert Einstein College of Medicine, Department of Pediatrics, Bronx, NY. Zupenda M. Davis, DrPH, MPH, MCHES, is an assistant professor and the director of the Bachelor of Science of Public Health Program, La Salle University, School of Nursing & Health Sciences, Department of Urban Public Health & Nutrition, Philadelphia, PA. Edward Gracely, PhD, is an associate professor in the Drexel University School of Public Health, Department of Epidemiology and Biostatistics, Philadelphia, PA. Lisa Bowleg, PhD, is a professor of applied social psychology at the George Washington University, Columbian College of Arts & Sciences, Washington, DC.
665
IN FOCUS
Satisfaction with Health Care among African American Women Living with HIV/AIDS
patients. Although the majority of the participants were women, their HIV status was not stated in the article. In a study specifically examining gender disparities in HIV health care utilization, the researchers concluded that women were significantly less likely to have optimal HIV health care services utilization, which included lower use of HIV primary care services and greater use of the emergency department (Sohler et al., 2007). In addition, the authors concluded that they could not identify the factors that explained the gender differences but that more qualitative and quantitative research are necessary to better understand the complexities of factors associated with these differences (Sohler et al., 2007). Further, a majority of patient satisfaction studies among PLWH/A have involved primarily African American men. In two recent studies, researchers examined health care utilization among African American men living with HIV (Bodenlos et al., 2007; Buseh, Kelber, Stevens, Poedel, & Park, 2009). Dang, Westbrook, Rodriguez-Barradas, and Giordano (2012) examined drivers of patient satisfaction among a sample of HIV primary care patients and found that evaluations of health care providers (HCPs) correlated the most with overall satisfaction with health care. However, the majority of the participants were African American men. Given these gaps in the literature and that satisfaction with health care can influence the quality of care received and health outcomes of PLWH/A, it is important to increase our understanding of what AAWLWH/A experience within the health care system. Examining factors that may enhance satisfaction with health care among AAWLWH/A may contribute to the enhancement of clinical care, overcoming barriers to utilization of care and treatment, and potentially decrease disparities in mortality rates. Further, it may also lend to the development of interventions around increasing access and follow through with health care among AAWLWH/A. Thus, the goal of this study was to examine possible determinants of satisfaction with health care among AAWLWH/A. Findings will contribute to a limited body of knowledge on factors that influence satisfaction with health care among this population.
Hypotheses The secondary data analysis tested the following hypotheses: (a) Participants who report more positive evaluations of their health, current use of HIV drug therapy, an undetectable viral load, higher CD4 counts, and lower rates of injection drug use will report higher levels of satisfaction with health care than participants who rate their health more negatively, are not currently taking HIV drug therapy, have a detectable viral load, have lower CD4 counts, and have higher rates of injection drug use and (b) participants who reported higher levels of communication with their HCPs about risk transmission will report higher levels satisfaction with health care in comparison to women who reported less communication with HCPs.
Participants A total of 184 women were enrolled in the Protect and Respect program. At enrollment all participants were HIV positive and diagnoses ranged between 0.3 (3.6 months) to 20.1 years (M = 8.7, SD = 5.0). The sample for this study consisted of the African American female participants who accounted for 85% (N = 157) of the total sample size of the parent study. The majority reported living with HIV for at least 6 months (M = 5.74, SD = 3.33) and ranged in age from 20 to 70 years (M = 39.85, SD = 8.72) (see Table 1).
Conceptual Models
A secondary analysis was conducted in 2009 from data collected during the Protect and Respect program, a randomized, controlled, intervention
Drawing on theoretical and empirical relationships highlighted in the patient satisfaction literature, two conceptual models were developed for this study: a theoretical model and a testable model. Variables needed to test the theoretical model were not included in the original study, however; a limitation of the data. Therefore, a testable model was developed. The testable model is the model
JOGNN, 43, 664-676; 2014. DOI: 10.1111/1552-6909.12491
http://jognn.awhonn.org
Methods Design
666
program for HIV-positive women. The program was designed to increase HIV/AIDS knowledge, protective attitudes, and risk-reduction selfefficacy and was implemented for 3 years starting in 2004. Women who were at least age 18 years, English speaking, and HIV-positive for at least 6 months were eligible to participate in the program. All women enrolled in Protect and Respect completed confidential questionnaires at four time periods: at baseline, 6 months, 12 months, and 18 months. All questionnaires were administered via Audio Computer-Assisted Self-Interviewing (ACASI) procedures. Only baseline data were utilized for the analyses in this study. The Institutional Review Board approved all study procedures.
IN FOCUS
Baker, J. L., Rodgers, C. R. R., Davis, Z. M., Gracely, E., and Bowleg, L.
Table 1: Summary of Participant Demographic Variables (N = 157) Variable Age
M(SD) N
%
Did not graduate from high school (HS)
73
46.4
HS Graduate/Graduate Equivalency Diploma
53
33.8
Some college/associate’s degree
24
15.3
39.85(8.72)
Level of Education
College graduate (Bachelor of Arts/Bachelor of Science)
5
3.2
Some graduate school
2
1.3
Yes
33
21.2
No
124
78.8
Heterosexual
130
82.8
Gay/Homosexual
9
5.7
Bisexual
14
8.9
Other
2
1.3
Not Sure
2
1.3
87
55.4
20
12.7
Employed
Sexual Identification (n = 156)
Relationship Status Single/Never Married Committed Relationship Domestic Partnership
10
7.0
Married
11
7.0
Separated
6
3.8
Divorced
11
7.0
Widowed
9
5.7
Other
3
1.4
$10,000 or less
115
73.2
$10,001 to $20,000
22
14.0
$20,001 to $40,000
13
8.2
$40,001 to $60,000
4
2.3
$60,001 to $80,000
2
1.2
Over $80,000
1
1.1
Annual Personal Income
Mode of HIV Transmission Sex with HIV+ man
129
82.2
Sex with HIV+ woman
2
1.3
Sharing needles
23
14.6
Needle Stick
3
1.9
JOGNN 2014; Vol. 43, Issue 5
667
IN FOCUS
Satisfaction with Health Care among African American Women Living with HIV/AIDS
that includes the questions and concepts that were available for statistical analyses from the baseline data. This model was tested in the final analysis. Theoretical Model. The theoretical model (see Figure 1) encompasses findings in the literature that demonstrated relationships between participants’ characteristics and satisfaction with health care. It is also an expansion of Gerbert, Love, Caspers, Linkins, and Burack’s (1999) theoretical model of the relationship between the characteristics of PLWH/A, the characteristics of physicians, patient involvement in health care, and improved health outcomes. Because satisfaction with health care has been shown to be significantly associated with involvement in health care among PLWH/A (Apollo et al., 2006; Roberts, 2002; Sullivan, Stein, Savetsky, & Samet, 2000), it preceded involvement in health care in the adaptation of the Gerbert model (1999). This model would have allowed an examination of whether participant and HCP characteristics influenced levels of satisfaction with health care and whether higher levels of satisfaction lead to enhanced patient involvement in health care. Testable Model. The testable model (see Figure 2) includes the questions that were available for statistical analysis from the original study’s questionnaire. Due to the small number of HCPs in the original study (N = 10), there was not enough variability in the available sample for a meaningful analysis of the effects of HCP characteristics such as gender and race/ethnicity on satisfaction with health care levels. Thus, HCP characteristics were not tested in the analysis. Further, there were no questions in the available data that encompassed patient involvement in health care thereby limiting the ability to test whether higher levels of satisfaction health care would be associated to enhanced involvement in health care. Thus, the testable model was used in this study to explore the relationships between patients’ characteristics and satisfaction with health care.
Cluster of differentiation 4 (CD4) count was selfreported in response to the following question: “What was the count of your most recent CD4 T-cell?” Participants provided a numerical response. Viral load was self-reported in response to the following question: “Was the result of your most recent viral load test detectable?” (response = yes/no). With regards to the reliability and validity of self -reported CD4 count and viral load, Kinsler and colleagues (2008) found significant correlations between health status markers provided by participants and those abstracted from paper medical records for CD4 cell counts (Kappa coefficient = .85, p < .001) and viral load (Kappa coefficient = .56, p < .001).
Predictor variables included HIV health status, injection drug use, and communication with HCPs about risk transmission. HIV health status consisted of four separate items that were used to assess the HIV health status of participants. Participants self-reported health status using a 5-point Likert-type scale in response to the following state-
Injection drug use (IDU) was the self-report variable of participant drug use in the past 30 days. The item that was included in the analysis was the following: “In the last 30 days, have you injected any drugs?” (response = yes/no). With regard to the reliability and validity of self- reported IDU measures, Johnson and colleagues (2000) examined the reliability and validity of selfreported IDU among a sample of 259 drug users not in treatment. Self-report of injection behavior in the last 30 days was relatively reliable (Pearson r(127) = .88). Validity of self-reported drug use was determined by comparing responses to questions about drug use in the last 48 hours to urinalysis results. Total agreement between self-report and urinalysis results was over 84% with comparable Kappa coefficients for cocaine (Kappa coefficient = .671) and opiates (Kappa coefficient = .58). Thus, pursuant to Johnson et al. (2000), the self-reported injection drug use items were considered reliable and valid and utilized in the analysis. In a recent study among a sample of out-of-treatment drug users the authors also found high validity of self-report of injection drug use with a 95% accuracy rate (Kappa coefficient = .54) and high reliability for self-report of amphetamine use in the past 30 days (Kappa coefficient = .75; Napper, Fisher, Johnson, & Wood, 2010).
JOGNN, 43, 664-676; 2014. DOI: 10.1111/1552-6909.12491
http://jognn.awhonn.org
Measures Predictor Variables
668
ment: “In general, would you say your health is . . . ” (1 = poor to 5 = excellent). The item current medication use was self-reported in response to the following question: “Are you currently using HIV drug therapy?” (response = yes/no). In a recent metaanalysis, Simoni and colleagues (2006) found that self-reported adherence was significantly associated to viral load, CD4 count, and other indirect measures such as pill count in 79% of studies comparing measurement approaches (Simoni et al., 2006).
IN FOCUS
Baker, J. L., Rodgers, C. R. R., Davis, Z. M., Gracely, E., and Bowleg, L.
Patient characteristics Medication use (ART) Self-rated health Substance use
Increased satisfaction with health care
Undetectable viral load
Active involvement in health care: Complying with HCP recommenddations
CD4 count
Less perceived discrimination
Trust of HCP
Enhanced Communication with HCP
HCP & patient gender concordance
HCP & patient race concordance
HCP
Patient race
HCP
Patient gender
Figure 1. Theoretical model. ART = antiviral therapy; HCP = health care provider.
In the baseline questionnaire, all items relating to communication with HCPs focused specifically on communication about HIV risk transmission. Thus, communication with HCPs was assessed with three separate items: (a) I discuss sex with my primary health care provider; (b) During my office visits, my primary health care provider and I usually talk about the risk of my transmitting HIV to others; and (c) The subject of using condoms comes up during my regular HIV care visits. Participants rated their agreement with these statements using a 7-point Likert-type scale (1 = strongly disagree – 7 = strongly agree). The communication items had low internal consistency (Cronbach’s α = .44). As a result, the three communication items were treated as distinct variables for purpose of analysis as there were no other items from the
JOGNN 2014; Vol. 43, Issue 5
questionnaire that measured communication with HCPs.
Outcome Variables Satisfaction with health care was assessed with three items: (a) I am satisfied with the medical services I received at this clinic, (b) I am satisfied with the social support services I receive at this clinic, and (c) I am satisfied with the HIV prevention services that I receive from the staff at this clinic. Participants rated their agreement with these items using a 7-point Likert-type scale (1 = strongly disagree – 7 = strongly agree). The satisfaction items had low internal consistency (Cronbach’s α = .33), indicating that they are not measuring the same underlying trait. Therefore, the satisfaction items were treated as separate items and not as a scale.
669
IN FOCUS
Satisfaction with Health Care among African American Women Living with HIV/AIDS
Participant characteristics
• • • • •
Medication use (ART) Self-rated health Substance use Undetectable viral load CD4 count
Satisfaction with health care
• • •
Communication with health care provider
• • •
Medical services Social support services HIV prevention services
Condom use HIV transmission Sex
Figure 2. Testable model. ART = antiviral therapy.
Data Analysis Descriptive statistics and frequency counts were used to characterize the study sample, and hypotheses were tested using methods appropriate to the level of the variables. Chi-squared tests were utilized for the bivariate analyses to compare proportions and t tests (or Mann-Whitney U tests) were used to compare means. Pearson’s (or Spearman’s depending on the data distribution) correlations were used to determine the relationship between the study variables. Regression analyses were used to determine the relationships between the predictor variables and the outcome variables. All statistical analyses were conducted using SPSS statistical software version 16. Hypothesis 1
To address and test Hypothesis #1, bivariate analyses between the predictor variables and the outcome variables were conducted using MannWhitney U tests for dichotomous predictor variables (e.g., detectable viral load and current drug therapy use) and correlation analysis for continuous variables (e.g., CD4 count).
670
count, current medication use, injection drug use, self-reported health status, viral load) and the outcome variables (i.e., satisfaction with medical services, social support services, and HIV prevention services). Hypothesis 2
To address and test Hypothesis #2, bivariate relationships were examined between the communication with HCP scores and the satisfaction with health care scores using correlation analysis. Linear regression procedures were utilized to determine the relationship between the communication with HCP measures (i.e., communication about condom use, HIV transmission and sex) and the outcome variables (i.e., satisfaction with medical services, social support services, and HIV prevention services).
Results Sample Description
Following these bivariate comparisons, a regression model (De Vincenzi, 1994) was constructed using the predictor variables and the satisfaction with health care items as the outcome variables. Linear regression was used to determine the relationship between the predictor variables (i.e., CD4
Participants were 157 African American women from the Protect and Respect study. Characteristics of study participants are displayed in Table 1. The mean age of the participants was 39.85 years (SD = 8.72). Approximately one half of the sample reported being single. More than 73% of participants reported an annual income of less than $10,000 and 78% were unemployed. Thirty-five percent had obtained a high school diploma. The
JOGNN, 43, 664-676; 2014. DOI: 10.1111/1552-6909.12491
http://jognn.awhonn.org
IN FOCUS
Baker, J. L., Rodgers, C. R. R., Davis, Z. M., Gracely, E., and Bowleg, L.
majority of participants (83%) identified as heterosexual, and 82% reported heterosexual transmission as their primary exposure category.
Among racial/ethnic minority women living with HIV, health care disparities lend to dissatisfaction with health care, which can result in treatment underutilization and discontinuity of care.
Description of Study Variables Participant Health Characteristics Several psychosocial and health variables were considered potential predictors of satisfaction with health care. These variables included participant self-reported health status, CD4 count, viral load, injection drug use, and ART use. Descriptive statistics for these variables are reported in Table 2. As shown in Table 2, injection drug use among the sample was relatively low, with only 26% of participants reporting drug use in the past 30 days. With regards to HIV health status, 65% of participants reported having a detectable viral load. The mean CD4 count was 459 (SD = 333). Considering that a CD4 count of less than 200 signifies an AIDS diagnosis (Wu et al., 2000), the mean CD4 count of the participants indicated that a majority of the women were in good health; more specifically, the AIDS virus was not progressing and their immune systems were functioning well. Lastly, almost one half of the participants (49%) reported current medication therapy use at baseline.
Satisfaction with Health Care Overall, the women reported satisfaction with their health care based on the three satisfaction variables. Eighty-eight percent of the participants (n = 140) reported being satisfied with medical services. A majority of participants (65%, n = 101) also reported satisfaction with HIV prevention services; 62% (n = 95) reported satisfaction with social support services.
Bivariate Correlations Significant bivariate correlations among the study variables are presented in Tables 3 and 4. Selfreported health status and a detectable viral load were significantly correlated with satisfaction with HIV prevention services (p < .05) and satisfaction with medical services (p < .05). Also, communication about sex and communication about condom use were significantly correlated with satisfaction with social support services (p < .05).
Regression Analyses Communication with Health Care Providers about Risk Transmission Overall, participants reported some degree of communication with their HCPs based on the three communication variables. The majority of the women (63%, n = 96) reported discussing sex with their HCPs and discussing condom use with their HCPs (63%, n = 95). A majority of the sample (66%, n = 101) also reported communicating with their HCPs about transmitting HIV to others.
To test Hypothesis 1, regression models were used to assess whether participant characteristics had an effect on satisfaction with health care. A separate regression model was used to assess the effect of each participant characteristic on each of the three satisfaction variables (see columns 1–3 in Table 5). Table 5 shows the effect of the dependent variables (satisfaction with health care items) regressed on the independent variables (participant characteristics). None of the participant characteristics significantly predicted participant
Table 2: Summary of Baseline Means of Participant Characteristics (N = 157) Variable
n (%)
Mean
SD
Median
Minimum
Maximum
Viral load detectable – yes
82(65.1) 459.36
332.63
400.0
0
2,000
3.12
1.14
3.0
1
5
How long at this clinic (years)
12.00
82.56
4
0
998
First learned of HIV + status (months)
5.74
3.33
6
0.3
240
Most recent CD4 count Current medication use- yes
77(63.1)
Self-reported health status Illegal drug use in past 30 days – yes
JOGNN 2014; Vol. 43, Issue 5
38(25.9)
671
IN FOCUS
Satisfaction with Health Care among African American Women Living with HIV/AIDS
Table 3: Summary of Spearman Correlations between Numeric Predictor Variables and Satisfaction Scales Satisfaction with
Medical Services
HIV Prevention Services
−.15
.06
Self-reported health status
.16∗
−.03
.17∗
Communication with HCP about sex
.05
−.21 ∗∗
−.06
−.07
−.15
Most recent CD4 count
Communication with HCP about HIV transmission Communication with HCP about condom use
−.02
Social Support Services
−.17 −.06
∗∗
−.24
−.03
Note. CD4 count = CD4 + T cells; HCP = health care provider. ∗ Correlation significant at p < .05 (two-tailed). ∗∗ Correlation significant at p < .01 (two-tailed)
Contrary to existing literature, African American women living with HIV can be satisfied with their health care.
satisfaction with medical services (p > .05, see column 1 in Table 5), satisfaction with social support services (p > .05, see column 2 in Table 5), or satisfaction with HIV prevention services (p > .05, see column 3 in Table 5). Thus, Hypothesis 1 was not supported. To test Hypothesis 2, regression models were generated to examine the effect of the communication variables on each satisfaction variable. A separate regression model was used to assess the effect of each communication variable on each of the three satisfaction variables (see columns 1–3 in Table 6). Table 6 shows the effect of the dependent variables (satisfaction with health care) regressed on the independent variables (communication with HCPs). None of the communication variables significantly predicted participant satisfaction with medical services (p > .05, see column 1 in Table 6), satisfaction with social support services (p > .05, see column 2 in Table 6), or satisfaction with HIV prevention services (p > .05, see column 3 in Table 6). Therefore, this hypothesis is not supported.
Discussion Although AAWLWH/A account for a sizable majority of people in need of health care services, and research documents that PLWH/A who are satisfied with their HIV care experience more optimal health outcomes, there is a significant gap in the literature on their health care experiences including satisfaction with health care. Thus, a key goal of this study was to report on rates of satisfaction with health care among AAWLWH/A and provide new information about what factors influence satisfaction with health care. We found that the majority of participants reported being satisfied with health care, specifically their medical, social support, and HIV prevention services. However, we did not observe any evidence that satisfaction with these aspects of health care were significantly correlated with reported health status, medication therapy use, illegal drug use, CD4 count or viral load.
Results from this study demonstrated that the participants were satisfied with health care in all three aspects of care analyzed: medical services, social support services, and HIV prevention services. The women also reported high ratings of communication with their HCPs about condom use,
The nonsignificant findings in the relationship between satisfaction with health care and psychosocial and health characteristics are not consistent with previous literature, which suggests a relationship between psychosocial and health characteristics and satisfaction with health care (Burke et al., 2003; Burke-Miller et al., 2006; Sullivan et al., 2000). It is possible that the analyzed variables (e.g., CD4 count, viral load, medication therapy use) may not be the only factors that influence
JOGNN, 43, 664-676; 2014. DOI: 10.1111/1552-6909.12491
http://jognn.awhonn.org
Summary of Results
672
sex, and transmission of HIV. However, we did not find any significant relationships between psychosocial/health characteristics and satisfaction with health care.
IN FOCUS
Baker, J. L., Rodgers, C. R. R., Davis, Z. M., Gracely, E., and Bowleg, L.
Table 4: Summary of Means ± SD for Satisfaction Scales as a Function of Categorical Variables Satisfaction with
n
Medical Services
Social Support Services
HIV Prevention Services
73
6.49 ± 1.2∗
4.36 ± 2.3∗
6.27 ± 1.5
HS/Graduate Equivalency Diploma
53
∗
4.36 ± 2.4
4.91 ± 2.1
6.16 ± 1.1
More than HS
31
6.55 ± 0.9
5.78 ± 1.6
6.59 ± 0.63
$10,000 Or Less
115
6.39 ± 1.2
4.84 ± 2.3
6.3 ± 1.2
$10,001 to $20,000
22
6.57 ± 0.7
5.14 ± 1.8
6.48 ± 0.9
$20,001 to $40,000
13
5.77 ± 1.8
4.84 ± 1.7
5.77 ± 1.7
Education Less than high school (HS)
Income
∗
1.0 ± 0
6.25 ± 1.5
$40,001 to $60,000
4
5.55 ± 2.4
$ 60,001 to $80,000
2
7.0 ± 0
6.5 ± 0.7
6.5 ± 0.7
Over $80,000
1
7.0 ± 0
1.0 ± 0∗
7.0 ± 0
Current Medication -yes
77
6.39 ± 0.9
4.56 ± 2.3
6.42 ± 1.1
Drug use in last 30 days -yes
38
6.31 ± 1.34
4.84 ± 2.2
6.36 ± 1.2
82
6.37 ± 1.3
4.89 ± 2.2
6.32 ± 1.1
130
6.3 ± 1.3
4.63 ± 2.2a
6.3 ± 1.2
Health status
Viral Load Detectable – yes
a
Sexual orientation Heterosexual – yes
Note. ∗ Correlation significant at p < .05 (two-tailed).
patient satisfaction with health care among this sample. Although it was hypothesized that communication with HCPs would be significantly correlated with satisfaction with health care, results indicate there was no statistically significant relationship. These findings are also not consistent with previous studies that suggest relationships between communication with HCPs and satisfaction with health care among PLWH/A (Apollo et al., 2006; Marelich & Murphy, 2003; Wilson & Kaplan, 2000). However, ratings of communication with HCPs were higher than expected based on the current literature base. Discrimination continues to be a reality for PLWH/A (Bogart et al., 2010) in ways that may prevent some, particularly AAWLWH/A who encounter judgment and stigma (Black & Miles, 2002), from discussing sensitive topics with their HCPs. Thus, this finding is encouraging as it demonstrates that AAWLWH/A can feel comfortable discussing very personal and stigmatized
JOGNN 2014; Vol. 43, Issue 5
issues including sex and condom use with their HCPs.
Limitations Although this study provides meaningful information related to satisfaction with health care among AALWH/A, there are limitations. One limitation of this study was the reliability of the measures. A major focus of this study was to demonstrate the relationships between participant health characteristics and satisfaction with health care. Because this was a secondary data analysis, a valid measure of patient satisfaction could not be used for analyses. Further, the satisfaction items were not reliable as a scale. Instead, three satisfaction items were used separately to assess different aspects of satisfaction with health care. None of the psychosocial and health characteristics was significantly associated with satisfaction with health care. Thus, reliability of self-reports of the measures should be considered.
673
IN FOCUS
Satisfaction with Health Care among African American Women Living with HIV/AIDS
Table 5: Regression Analysis for Participant Characteristics as Predictors of Satisfaction with Healthcare (N = 157) Satisfaction with
Satisfaction with Social
Satisfaction with HIV
Variable
Medical Services
Support Services
Prevention Services
Self-Report
β SE F p
β SE F p
β SE F p
Health Status
.09 0.1 1.0.31
−.06 0.2 .15.69
.13 .09 2.1 .15
Current Medication Use
.54 .22 .06 .81
−.25 0.4 .34 .56
−.15 .22 .45 .50
Detectable Viral Load
0.2 0.2 0.4 .52
−0.2 0.4 15 0.7
−.12 .24.22 .63
CD4 count
.06 .21 .39.53
−.001 .001 1.9.17
.09 .02.84.35
Drug Use
−.07 .24 .09 .77
−.14 .44 .09 .76
−.12 23 .32 .57
R2
0.004
0.020
0.009
Note. Each column provides results from the set of regressions using one of the three satisfaction variables as the dependent variable; SE = standard error; F = f distribution; CD4 count = CD4 + T cells.
This study was a secondary data analysis, which limited the extent to which other variables that may be associated with satisfaction with health care, such as time spent with HCP, discrimination from HCPs, and trust (Benkert et al., 2006; O’Malley et al., 2004; Stone, 2004; Warde, 2001), could be analyzed. The quantitative design of this study is another possible limitation. A mixed methods study that involved the use of qualitative methods may have been a more appropriate design for investigating new concepts particularly associations between health characteristics and satisfaction with health care. Lastly, study results could be confounded by the relationships between the HCPs and the study participants, as most of the women were patients of the HIV care clinic where the Protect and Respect program was located for an average of 13 years (see Table 1). Patient–provider relationships have been found to be more important than other com-
ponents of care in its association with overall satisfaction (Dang et al., 2012). Thus, it is possible that long-term relationships between the women and their HCPs may have inflated their ratings of satisfaction. Given that a majority of women did indicate satisfaction with the health care services provided, it would appear that the HCPs are providing a very high level standard of care to their patients. However, HCP characteristics, provider engagement, or specific clinic attributes were not available to be analyzed to assess their relationship to satisfaction with health care among the sample.
Conclusion Given the disparities of HIV infection among women, health care providers will continue to be sought after to provide HIV care to African American women. Because satisfaction with health care can influence the quality of care received, health
Table 6: Summary of Regression Analysis for Communication with Health Care Provider as Predictor of Satisfaction with Health Care (N = 157) Satisfaction with
Satisfaction with Social
Satisfaction with HIV
Variable
Medical Services
Support Services
Prevention Services
Discuss sex with provider
β SE F p
β SE F p
β SE F p
−.06 .047 1.6 .20
−.13 .09 2.3 .13
−.006 .05 .02 .90
Discuss risk of HIV transmission to others
−.07 .049 1.9 .17
−.06 .09 .39 .54
−.06 .05 1.6 .21
Subject of condoms use comes up in HIV care visits
−.08 .047 2.8 .09
−.15 .09 3.1 .08
−.03 .05 .47 .50
R2
.018
0.021
0.003
Note. Each column provides results from the set of regressions using one of the three satisfaction variables as the dependent variable; SE = standard error; F = f distribution.
674
JOGNN, 43, 664-676; 2014. DOI: 10.1111/1552-6909.12491
http://jognn.awhonn.org
IN FOCUS
Baker, J. L., Rodgers, C. R. R., Davis, Z. M., Gracely, E., and Bowleg, L.
outcomes, and adherence to provider recommendations among people living with HIV/AIDS, health care providers’ ability to elicit satisfaction from their patients is just as important as the health care services they provide. We contribute to the patient satisfaction literature by specifically focusing on satisfaction with health care among AAWLWH/A. However, the results may be reflective of care provided by the health care providers of the specific clinic they attended. To enhance the understanding of satisfaction among AAWLWH/A, additional research should utilize qualitative methods to explore how the context of the patient– provider relationship influences satisfaction with health care. Further analyses with a larger sample size, more reliable and validated measures, and a mixed methods approach may also improve assessments of satisfaction with health care among AAWLWH/A.
Burke-Miller, J., Cook, J., Cohen, M., Hessol, N., Wilson, T., Richardson, J., & Gange, S. (2006). Longitudinal relationships between use of highly active antiretroviral therapy and satisfaction with care among women living with HIV/AIDS. American Journal of Public Health, 96(6), 1044–1051. doi:10.2105/ajph.2005.061929 Buseh, A., Kelber, S., Stevens, T., Poedel, R., & Park, C. G. (2009). Interpersonal and attitudinal factors in healthcare utilization among HIV-infected urban African American men. Challenge, 5(2), 1– 18. Campbell, J., Ramsay, J., & Greene, J. (2001). Age, gender, socioeconomic, and ethnic differences in patients’ assessments of primary health care. Quality in Health Care, 10, 90–95. Cargill, V. A., & Stone, V. E. (2005). HIV/AIDS: A minority health issue. Medical Clinics of North America, 89(4), 895–912. Centers for Disease Control and Prevention. (2011). HIV surveillance among women. Atlanta, GA: Author. Retrieved from http://www. cdc.gov/hiv/topics/surveillance/resources/slides/women/index. htm Centers for Disease Control and Prevention. (2012a). CDC fact sheet: New HIV infections in the United States. Atlanta, GA: Author. Retrieved from http://www.cdc.gov/nchhstp/newsroom/ docs/2012/hiv-infections-2007--2010.pdf Centers for Disease Control and Prevention. (2012b). HIV in the United
Acknowledgement The authors thank Jennifer A. Faerber, PhD, Division of Rheumatology, Children’s Hospital of Philadelphia and Thomas R. Tenhave, PhD (deceased), former Director of the Center for Clinical Epidemiology and Biostatistics at the University of Pennsylvania.
States: The stages of care. Atlanta, GA: Author. Retrieved from http://www.cdc.gov/hiv/pdf/research_mmp_stagesofcare.pdf Centers for Disease Control and Prevention. (2013). Fact sheet: HIV among women. Atlanta, GA: Author. Retrieved from http://www.cdc.gov/hiv/topics/women/pdf/women.pdf Cohen, J. (2008). Challenges and successes in reducing health disparities: Workshop summary. Washington, DC: National Academies Press. Copeland, V. C., Scholle, S. H., & Binko, J. A. (2003). Patient satisfaction. Journal of Health & Social Policy, 17(2), 35–48.
REFERENCES Abel, E., & Painter, L. (2003). Factors that influence adherence to HIV
Dang, B., Westbrook, R., Rodriguez-Barradas, M., & Giordano, T. (2012). Identifying drivers of overall satisfaction in patients re-
medications: Perceptions of women and health care providers.
ceiving HIV primary care: A cross-sectional study. PLOS One,
Journal of the Association of Nurses in AIDS Care, 14(4), 61–69.
7(8), 1–7.
Apollo, A., Golub, S. A., Wainberg, M. L., & Indyk, D. (2006). Patient-
De Vincenzi, I. (1994). A longitudinal study of human immunodeficiency
provider relationships, HIV, and adherence: Requisites for a part-
virus transmission by heterosexual partners. New England Jour-
nership. Social Work in Health Care, 42(3/4), 209–224.
nal of Medicine, 331(6), 341–346.
Benkert, R., Peters, R., Clark, R., & Keeves-Foster, K. (2006). Effects of
El-Bassel, N., Caldeira, N., Ruglass, L., & Gilbert, L. (2009). Addressing
perceived racism, cultural mistrust, and trust in providers on sat-
the unique needs of African American women in HIV prevention.
isfaction with care. Journal of the National Medical Association, 98(9), 1532–1540. Black, B., & Miles, M. (2002). Calculating the risks and benefits of disclosure in African American women who have HIV. Journal of Obstetric, Gynecologic, & Neonatal Nursing, 31(6), 688–697.
American Journal of Public Health, 99, 996–1001. Emlet, C. (2007). Experiences of stigma in older adults living with HIV/AIDS: A mixed methods analysis. AIDS Patient Care & STDs, 21(10), 740–752. Gerbert, B., Love, C., Caspers, N., Linkins, K., & Burack, J. H.
Bodenlos, J., Grothe, K., Whitehead, D., Konkle-Parker, D., Jones, G.,
(1999). “Making all the difference in the world”: How physi-
& Brantley, P. (2007). Attitudes toward health care providers and
cians can help HIV-seropositive patients become more involved
appointment attendance in HIV/AIDS Patients. Journal of the
in their healthcare. AIDS Patient Care & STDs, 13(1), 29–
Association of Nurses in AIDS Care, 18(3), 65–73.
39.
Bogart, L., Bird, S., Walt, L., Delahanty, D., & Figler, J. (2004). Associa-
Hall, J. A., & Dornan, M. C. (1988). Meta-analysis of satisfaction with
tion of stereotypes about physicians to health care satisfaction,
medical care: Description of research domain and analysis of
help-seeking behavior, and adherence to treatment. Social Sci-
overall satisfaction levels. Social Science & Medicine, 27, 637–
ence & Medicine, 58, 1049–1058.
644.
Bogart, L., Wagner, G., Galavan, F., & Klein, D. (2010). Longitudinal re-
Heckman, B. D., Catz, S. L., Heckman, T. G., Miller, J. G., & Kalichman,
lationships between antiretroviral treatment adherence and dis-
S. C. (2004). Adherence to antiretroviral therapy in rural persons
crimination due to HIV-serostatus, race, and sexual orientation
living with HIV disease in the United States. AIDS Care, 16(2),
among African-American men with HIV. Annals of Behavioral Medicine, 40(2), 184–190. Burke, J., Cook, J., Cohen, H., Wilson, T., Anastos, K., Young, M., & Gange, S. (2003). Dissatisfaction with medical care among women with HIV: Dimensions and associated factors. AIDS Care, 15(4), 451–462. doi:10.1080/0954012031000134692
JOGNN 2014; Vol. 43, Issue 5
219–230. doi:10.1080/09540120410001641066 Institute of Medicine. (2003). Unequal treatment: Confronting racial and ethnic disparities in health care. Washington, DC: National Academies Press. Johnson, M., Fisher, D., Montoya, I., Booth, R., Rhodes, F., Andersen, M., & Williams, M. (2000). Reliability and validity of
675
IN FOCUS
Satisfaction with Health Care among African American Women Living with HIV/AIDS
not-in-treatment drug users’ follow-up self-reports. AIDS and Behavior, 4, 373–380.
STDs, 16(1), 43–50.
Kaiser Family Foundation. (2013). Women and HIV fact sheet. Retrieved
Schneider, J., Kaplan, S. H., Greenfield, S., Li, W., & Wilson, I. B. (2004).
from http://kff.org/hivaids/fact-sheet/women-and-hivaids-in-the-
Better physician-patient relationships are associated with higher
united-states/
reported adherence to antiretroviral therapy in patients with HIV
Kinsler, J., Cunningham, C., Mohanty, S., & Wong, M. (2008). Are medical records a more reliable and valid source of CD4 count, viral
infection. Journal of General Internal Medicine, 19(11), 1096– 1103. doi:JGI30418
load, and outpatient visit data than self-reports? A comparison
Simoni, J., Kurth, A., Pearson, C., Pantalone, D., Merrill, J., & Frick, P.
with electronic medical records as the gold standard. Health
(2006). Self-report measures of antiretroviral therapy adherence:
Services and Outcomes Research Methodology, 8, 187–192.
A review with recommendations for HIV research and clinical
Kleeberger, C., Buechner, J., & Palella, F. (2004). Changes in adherence to highly active antiretroviral therapy medications in the Multi-center AIDS Cohort Study. AIDS, 18, 683–688. LaVeist, T., & Carroll, T. (2002). Race of physician and satisfaction with care among African American patients. Journal of the National Medical Association, 94, 937–943.
management. AIDS Behavior, 10, 227–245. Sohler, N., Li, X., & Cunningham, C. (2007). Perceived discrimination among severely disadvantaged people with HIV infection. Public Health Reports, 122(3), 347–355. Stone, V. E. (2004). Optimizing the care of minority patients with HIV/AIDS. Clinical Infectious Diseases, 38(3), 400–404.
LaVeist, T., & Nuru-Jeter, A. (2002). Is doctor-patient race concordance
Sullivan, L. M., Stein, M. D., Savetsky, J. B., & Samet, J. H. (2000).
associated with greater satisfaction with care? Journal of Health
The doctor-patient relationship and HIV-infected patients’ satis-
and Social Behavior, 43, 296–306.
faction with primary care physicians. Journal of General Internal
Marelich, W. D., & Murphy, D. A. (2003). Effects of empowerment among HIV-positive women on the patient-provider relationship. AIDS Care, 15(4), 475–481. Napper, L., Fisher, D., Johnson, M., & Wood, M. (2010). The reliability and validity of drug users’ self reports of amphetamine use among primarily heroin and cocaine users. Addictive Behaviors, 35, 350–354.
Medicine, 15(7), 462–469. U.S. Department of Health and Human Services. (2010). U.S. Department of Health and Human Services: National HIV/AIDS strategy.
Retrieved
from
http://aids.gov/federal-resources/
national-hiv-aids-strategy/nhas.pdf Warde, C. (2001). Time is of the essence. Journal of General Internal Medicine, 16, 712–713.
O’Malley, A., & Forrest, C. (2002). Beyond the examination room: Pri-
Wilson, I. B., & Kaplan, S. (2000). Physician-patient communication in
mary care performance and the patient-physician relationship
HIV disease: The importance of patient, physician, and visit char-
for low-income women. Journal of General Internal Medicine,
acteristics. Journal of Acquired Immune Deficiency Syndrome,
17(1), 66–74.
25(5), 417–425.
O’Malley, A., Sheppard, V., Schwartz, M., & Mandelblatt, J. (2004). The
Wong, M. D., Shapiro, M. F., Boscardin, W. J., & Ettner, S. L. (2002).
role of trust in use of preventive services among low-income
Contribution of major diseases to disparities in mortality. New
African-American women. Preventive Medicine, 38(6), 77– 85. Roberts, K. J. (2002). Physician-patient relationships, patient satisfaction, and antiretroviral medication adherence among HIV in-
676
fected adults attending a public health clinic. AIDS Patient Care
JOGNN, 43, 664-676; 2014. DOI: 10.1111/1552-6909.12491
England Journal of Medicine, 347(20), 1585–1592. Wu, A. W., Gifford, A., Asch, S., Cohn, S. E., Bozzette, S. A., & Yurk, R. (2000). Quality-of-care indicators for HIV/AIDS. Disease Management & Health Outcomes, 7(6), 315–330.
http://jognn.awhonn.org
IN FOCUS
Results from a Secondary Data Analysis Regarding Satisfaction with Health Care among African American Women Living with HIV/AIDS Jillian Lucas Baker, Caryn R. R. Rodgers, Zupenda M. Davis, Edward Gracely, and Lisa Bowleg
Correspondence Jillian Lucas Baker, DrPH, EdM, La Salle University, School of Nursing & Health Sciences, Department of Urban Public Health & Nutrition, 1900 W. Olney Ave. St. Benilde Tower, 3rd Floor, Office 3334, Philadelphia, PA 19144. [email protected]
ABSTRACT
Keywords African American women HIV AIDS health care providers patient satisfaction quantitative
Results: A majority of women reported satisfaction with medical services (88%, n = 140). Communication with health care providers, detectable viral load, education, income, self-reported health status, and sexual orientation were significantly bivariately associated with satisfaction with healthcare (all ps < .05). In the multivariate models, no variables significantly predicted satisfaction with healthcare.
Objective: To analyze satisfaction with health care among African American women living with HIV/AIDS. Design: Secondary analysis of baseline data of African American women who participated in Protect and Respect, a sexual risk reduction program for women living with HIV/AIDS Setting: HIV Care Clinic in an urban city in the northeast United States. Participants: One hundred fifty-seven (157) African American women living with HIV/AIDS. Methods: Regression analyses were used to examine the relationships between demographic variables, self-reported health characteristics, communication with health care providers, and satisfaction with health care provider.
Conclusion: Because satisfaction with health care can influence the quality of care received, health outcomes, and adherence to provider recommendations among patients living with HIV/AIDS, health care providers’ ability to elicit satisfaction from their patients is just as important as the services they provide. This project is one of the first studies to find high rates of satisfaction with health care among African American women living with HIV/AIDS. Further examination of satisfaction with health care among African American women living HIV/AIDS may help in narrowing health care disparities and negative treatment outcomes.
JOGNN, 43, 664-676; 2014. DOI: 10.1111/1552-6909.12491 Accepted May 2014
Jillian Lucas Baker, DrPH, EdM, is an assistant professor of public health, La Salle University, School of Nursing & Health Sciences, Department of Urban Public Health & Nutrition, Philadelphia, PA.
(Continued)
The authors report no conflict of interest or relevant financial relationships.
664
n the United States, one fourth of the population living with HIV infection are women (Centers for Disease Control and Prevention [CDC], 2013). African American women are disproportionally affected: 64% of new infections in women occur in African Americans, 18% in Whites, and 15% in Hispanic/Latinas (CDC, 2012a). Although the CDC (2001) reported a decrease in new incidence of HIV infections among African American women, 1 in 32 African American women will be diagnosed with HIV infection in their lifetimes.
I
African American women are disproportionately vulnerable not only to HIV infection but also to disparities in the health care system that affect the quality of care they receive and their access
to health care. Overall, African American women are less likely to use health care services despite need (Copeland, Scholle, & Binko, 2003). African American women also face particularly high rates of poverty, with one in four living in poverty (U.S. Department of Health & Human Services, 2010). Poverty is linked to lower quality and limited access to education and health care for African American women (El-Bassel, Caldeira, Ruglass, & Gilbert, 2009). In addition to these disparities, HIV-positive African Americans experience prejudicial treatment by health care providers (LaVeist & Carroll, 2002; Sohler, Li, & Cunningham, 2007), have delayed access to care (Cargill & Stone, 2005), and lower rates of antiretroviral adherence (Bogart, Bird, Walt, Delahanty, & Figler, 2004;
C 2014 AWHONN, the Association of Women’s Health, Obstetric and Neonatal Nurses
http://jognn.awhonn.org
Baker, J. L., Rodgers, C. R. R., Davis, Z. M., Gracely, E., and Bowleg, L.
Kleeberger, Buechner, & Palella, 2004) in comparison to Whites. Currently, several opportunities are missed for linking women with HIV to health care. Among HIVpositive women, 85% have been diagnosed, however, only 70% have been linked to care (Kaiser Family Foundation, 2013). Further, 41% are in regular care, 36% are prescribed antiretroviral therapy (ART), and 26%are virally suppressed (CDC, 2012b). An alarming concern is that these factors are associated with poorer clinical outcomes and higher mortality among African Americans living with HIV/AIDS (Bogart, Wagner, Galavan, & Klein, 2010; Wong, Shapiro, Boscardin, & Ettner, 2002). Consequently, African American women have the highest AIDS death rate (23.8 per 100,000 population) among all races/ethnicities, which is 22 times as high as the rate of death for Whites (1.1) and almost 5 times as high as the rate for Hispanics/Latinos (4.5) (CDC, 2011). Examining factors that result in treatment nonadherence is crucial for enhancing health care outcomes among African Americans living with HIV/AIDS (Bogart et al., 2010). The purpose of this study was to describe patient reports of satisfaction with healthcare among African American women living with HIV/AIDS (AAWLWH/A). In addition to reporting satisfaction with health care in general, determinants of satisfaction with health care were also examined (e.g., sensitive communication with health care providers and certain psychosocial and health characteristics).
Among women of color in the United States, African American women are disproportionately affected by HIV/AIDS.
volvement in health care decision making (Apollo, Golub, Wainberg, & Indyk, 2006; Marelich & Murphy, 2003). Variables that can influence patient satisfaction include race, socioeconomic status, cultural mistrust, experiences of racism and discrimination, trust in health care providers (Benkert, Peters, Clark, & Keeves-Foster, 2006), patient–provider communication (O’Malley & Forrest, 2002), and health status (Burke et al., 2003; O’Malley, Sheppard, Schwartz, & Mandelblatt, 2004). Overall, African Americans have been found to be less satisfied with their health care (Campbell, Ramsay, & Greene, 2001; LaVeist & Nuru-Jeter, 2002). However, existing literature regarding AAWLWH/A and their access to and participation in the health care system is extremely limited. In one study of dissatisfaction with health care among women living with HIV/AIDS, the authors found that women who reported being Latina, having a lower health status, not receiving ART, with depressive symptomatology, and no consistent health care provider were more likely to be dissatisfied with their health care (Burke et al., 2003). These factors are all relevant for the health outcomes of AAWLWH/A, as dissatisfaction with health care among PLWH/A can result in negative health outcomes such as treatment underutilization and discontinuity of care (Abel & Painter, 2003).
Satisfaction with Health Care among People Living with HIV/AIDS
Gaps in the Literature
Patient satisfaction research began in the 1960s (Hall & Dornan, 1988) and has been the most analyzed independent variable linked to health outcomes. However, patient satisfaction research was limited in the field of HIV due to confidentiality limitations, the short existence of the disease, and the stigma attached (Emlet, 2007). Thus, examination of patient satisfaction among persons living with HIV/AIDS (PLWH/A) is a very new field of inquiry. Some researchers have examined satisfaction or dissatisfaction with health care as a contributing factor to health care utilization and adherence to provider recommendations. Recent authors also found that satisfaction with health care has a positive effect on adherence to ART (Burke-Miller et al., 2006; Heckman, Catz, Heckman, Miller, & Kalichman, 2004; Schneider, Kaplan, Greenfield, Li, & Wilson, 2004) and in-
Although researchers have repeatedly demonstrated health disparities in access and quality of care overall affecting the African American community (Cohen, 2008; Institute of Medicine, 2003), few researchers have examined patient satisfaction specifically among AAWLWH/A. Some qualitative researchers have examined patient– provider interactions, social support, the role of the caregiver, and stigma among AAWLWH/A (Black & Miles, 2002). For example, the quality of the patient–provider relationship has been noted as influential toward the active participation of women living with HIV/AIDS in their own health care (Marelich & Murphy, 2003). In a quantitative study, Benkert and colleagues (2006) found that perceptions of racism and mistrust had a negative effect on satisfaction with health care among a sample of low-income African American
JOGNN 2014; Vol. 43, Issue 5
IN FOCUS
Caryn R. R. Rodgers, PhD, is an assistant professor at the Preventive Intervention Research Center, Albert Einstein College of Medicine, Department of Pediatrics, Bronx, NY. Zupenda M. Davis, DrPH, MPH, MCHES, is an assistant professor and the director of the Bachelor of Science of Public Health Program, La Salle University, School of Nursing & Health Sciences, Department of Urban Public Health & Nutrition, Philadelphia, PA. Edward Gracely, PhD, is an associate professor in the Drexel University School of Public Health, Department of Epidemiology and Biostatistics, Philadelphia, PA. Lisa Bowleg, PhD, is a professor of applied social psychology at the George Washington University, Columbian College of Arts & Sciences, Washington, DC.
665
IN FOCUS
Satisfaction with Health Care among African American Women Living with HIV/AIDS
patients. Although the majority of the participants were women, their HIV status was not stated in the article. In a study specifically examining gender disparities in HIV health care utilization, the researchers concluded that women were significantly less likely to have optimal HIV health care services utilization, which included lower use of HIV primary care services and greater use of the emergency department (Sohler et al., 2007). In addition, the authors concluded that they could not identify the factors that explained the gender differences but that more qualitative and quantitative research are necessary to better understand the complexities of factors associated with these differences (Sohler et al., 2007). Further, a majority of patient satisfaction studies among PLWH/A have involved primarily African American men. In two recent studies, researchers examined health care utilization among African American men living with HIV (Bodenlos et al., 2007; Buseh, Kelber, Stevens, Poedel, & Park, 2009). Dang, Westbrook, Rodriguez-Barradas, and Giordano (2012) examined drivers of patient satisfaction among a sample of HIV primary care patients and found that evaluations of health care providers (HCPs) correlated the most with overall satisfaction with health care. However, the majority of the participants were African American men. Given these gaps in the literature and that satisfaction with health care can influence the quality of care received and health outcomes of PLWH/A, it is important to increase our understanding of what AAWLWH/A experience within the health care system. Examining factors that may enhance satisfaction with health care among AAWLWH/A may contribute to the enhancement of clinical care, overcoming barriers to utilization of care and treatment, and potentially decrease disparities in mortality rates. Further, it may also lend to the development of interventions around increasing access and follow through with health care among AAWLWH/A. Thus, the goal of this study was to examine possible determinants of satisfaction with health care among AAWLWH/A. Findings will contribute to a limited body of knowledge on factors that influence satisfaction with health care among this population.
Hypotheses The secondary data analysis tested the following hypotheses: (a) Participants who report more positive evaluations of their health, current use of HIV drug therapy, an undetectable viral load, higher CD4 counts, and lower rates of injection drug use will report higher levels of satisfaction with health care than participants who rate their health more negatively, are not currently taking HIV drug therapy, have a detectable viral load, have lower CD4 counts, and have higher rates of injection drug use and (b) participants who reported higher levels of communication with their HCPs about risk transmission will report higher levels satisfaction with health care in comparison to women who reported less communication with HCPs.
Participants A total of 184 women were enrolled in the Protect and Respect program. At enrollment all participants were HIV positive and diagnoses ranged between 0.3 (3.6 months) to 20.1 years (M = 8.7, SD = 5.0). The sample for this study consisted of the African American female participants who accounted for 85% (N = 157) of the total sample size of the parent study. The majority reported living with HIV for at least 6 months (M = 5.74, SD = 3.33) and ranged in age from 20 to 70 years (M = 39.85, SD = 8.72) (see Table 1).
Conceptual Models
A secondary analysis was conducted in 2009 from data collected during the Protect and Respect program, a randomized, controlled, intervention
Drawing on theoretical and empirical relationships highlighted in the patient satisfaction literature, two conceptual models were developed for this study: a theoretical model and a testable model. Variables needed to test the theoretical model were not included in the original study, however; a limitation of the data. Therefore, a testable model was developed. The testable model is the model
JOGNN, 43, 664-676; 2014. DOI: 10.1111/1552-6909.12491
http://jognn.awhonn.org
Methods Design
666
program for HIV-positive women. The program was designed to increase HIV/AIDS knowledge, protective attitudes, and risk-reduction selfefficacy and was implemented for 3 years starting in 2004. Women who were at least age 18 years, English speaking, and HIV-positive for at least 6 months were eligible to participate in the program. All women enrolled in Protect and Respect completed confidential questionnaires at four time periods: at baseline, 6 months, 12 months, and 18 months. All questionnaires were administered via Audio Computer-Assisted Self-Interviewing (ACASI) procedures. Only baseline data were utilized for the analyses in this study. The Institutional Review Board approved all study procedures.
IN FOCUS
Baker, J. L., Rodgers, C. R. R., Davis, Z. M., Gracely, E., and Bowleg, L.
Table 1: Summary of Participant Demographic Variables (N = 157) Variable Age
M(SD) N
%
Did not graduate from high school (HS)
73
46.4
HS Graduate/Graduate Equivalency Diploma
53
33.8
Some college/associate’s degree
24
15.3
39.85(8.72)
Level of Education
College graduate (Bachelor of Arts/Bachelor of Science)
5
3.2
Some graduate school
2
1.3
Yes
33
21.2
No
124
78.8
Heterosexual
130
82.8
Gay/Homosexual
9
5.7
Bisexual
14
8.9
Other
2
1.3
Not Sure
2
1.3
87
55.4
20
12.7
Employed
Sexual Identification (n = 156)
Relationship Status Single/Never Married Committed Relationship Domestic Partnership
10
7.0
Married
11
7.0
Separated
6
3.8
Divorced
11
7.0
Widowed
9
5.7
Other
3
1.4
$10,000 or less
115
73.2
$10,001 to $20,000
22
14.0
$20,001 to $40,000
13
8.2
$40,001 to $60,000
4
2.3
$60,001 to $80,000
2
1.2
Over $80,000
1
1.1
Annual Personal Income
Mode of HIV Transmission Sex with HIV+ man
129
82.2
Sex with HIV+ woman
2
1.3
Sharing needles
23
14.6
Needle Stick
3
1.9
JOGNN 2014; Vol. 43, Issue 5
667
IN FOCUS
Satisfaction with Health Care among African American Women Living with HIV/AIDS
that includes the questions and concepts that were available for statistical analyses from the baseline data. This model was tested in the final analysis. Theoretical Model. The theoretical model (see Figure 1) encompasses findings in the literature that demonstrated relationships between participants’ characteristics and satisfaction with health care. It is also an expansion of Gerbert, Love, Caspers, Linkins, and Burack’s (1999) theoretical model of the relationship between the characteristics of PLWH/A, the characteristics of physicians, patient involvement in health care, and improved health outcomes. Because satisfaction with health care has been shown to be significantly associated with involvement in health care among PLWH/A (Apollo et al., 2006; Roberts, 2002; Sullivan, Stein, Savetsky, & Samet, 2000), it preceded involvement in health care in the adaptation of the Gerbert model (1999). This model would have allowed an examination of whether participant and HCP characteristics influenced levels of satisfaction with health care and whether higher levels of satisfaction lead to enhanced patient involvement in health care. Testable Model. The testable model (see Figure 2) includes the questions that were available for statistical analysis from the original study’s questionnaire. Due to the small number of HCPs in the original study (N = 10), there was not enough variability in the available sample for a meaningful analysis of the effects of HCP characteristics such as gender and race/ethnicity on satisfaction with health care levels. Thus, HCP characteristics were not tested in the analysis. Further, there were no questions in the available data that encompassed patient involvement in health care thereby limiting the ability to test whether higher levels of satisfaction health care would be associated to enhanced involvement in health care. Thus, the testable model was used in this study to explore the relationships between patients’ characteristics and satisfaction with health care.
Cluster of differentiation 4 (CD4) count was selfreported in response to the following question: “What was the count of your most recent CD4 T-cell?” Participants provided a numerical response. Viral load was self-reported in response to the following question: “Was the result of your most recent viral load test detectable?” (response = yes/no). With regards to the reliability and validity of self -reported CD4 count and viral load, Kinsler and colleagues (2008) found significant correlations between health status markers provided by participants and those abstracted from paper medical records for CD4 cell counts (Kappa coefficient = .85, p < .001) and viral load (Kappa coefficient = .56, p < .001).
Predictor variables included HIV health status, injection drug use, and communication with HCPs about risk transmission. HIV health status consisted of four separate items that were used to assess the HIV health status of participants. Participants self-reported health status using a 5-point Likert-type scale in response to the following state-
Injection drug use (IDU) was the self-report variable of participant drug use in the past 30 days. The item that was included in the analysis was the following: “In the last 30 days, have you injected any drugs?” (response = yes/no). With regard to the reliability and validity of self- reported IDU measures, Johnson and colleagues (2000) examined the reliability and validity of selfreported IDU among a sample of 259 drug users not in treatment. Self-report of injection behavior in the last 30 days was relatively reliable (Pearson r(127) = .88). Validity of self-reported drug use was determined by comparing responses to questions about drug use in the last 48 hours to urinalysis results. Total agreement between self-report and urinalysis results was over 84% with comparable Kappa coefficients for cocaine (Kappa coefficient = .671) and opiates (Kappa coefficient = .58). Thus, pursuant to Johnson et al. (2000), the self-reported injection drug use items were considered reliable and valid and utilized in the analysis. In a recent study among a sample of out-of-treatment drug users the authors also found high validity of self-report of injection drug use with a 95% accuracy rate (Kappa coefficient = .54) and high reliability for self-report of amphetamine use in the past 30 days (Kappa coefficient = .75; Napper, Fisher, Johnson, & Wood, 2010).
JOGNN, 43, 664-676; 2014. DOI: 10.1111/1552-6909.12491
http://jognn.awhonn.org
Measures Predictor Variables
668
ment: “In general, would you say your health is . . . ” (1 = poor to 5 = excellent). The item current medication use was self-reported in response to the following question: “Are you currently using HIV drug therapy?” (response = yes/no). In a recent metaanalysis, Simoni and colleagues (2006) found that self-reported adherence was significantly associated to viral load, CD4 count, and other indirect measures such as pill count in 79% of studies comparing measurement approaches (Simoni et al., 2006).
IN FOCUS
Baker, J. L., Rodgers, C. R. R., Davis, Z. M., Gracely, E., and Bowleg, L.
Patient characteristics Medication use (ART) Self-rated health Substance use
Increased satisfaction with health care
Undetectable viral load
Active involvement in health care: Complying with HCP recommenddations
CD4 count
Less perceived discrimination
Trust of HCP
Enhanced Communication with HCP
HCP & patient gender concordance
HCP & patient race concordance
HCP
Patient race
HCP
Patient gender
Figure 1. Theoretical model. ART = antiviral therapy; HCP = health care provider.
In the baseline questionnaire, all items relating to communication with HCPs focused specifically on communication about HIV risk transmission. Thus, communication with HCPs was assessed with three separate items: (a) I discuss sex with my primary health care provider; (b) During my office visits, my primary health care provider and I usually talk about the risk of my transmitting HIV to others; and (c) The subject of using condoms comes up during my regular HIV care visits. Participants rated their agreement with these statements using a 7-point Likert-type scale (1 = strongly disagree – 7 = strongly agree). The communication items had low internal consistency (Cronbach’s α = .44). As a result, the three communication items were treated as distinct variables for purpose of analysis as there were no other items from the
JOGNN 2014; Vol. 43, Issue 5
questionnaire that measured communication with HCPs.
Outcome Variables Satisfaction with health care was assessed with three items: (a) I am satisfied with the medical services I received at this clinic, (b) I am satisfied with the social support services I receive at this clinic, and (c) I am satisfied with the HIV prevention services that I receive from the staff at this clinic. Participants rated their agreement with these items using a 7-point Likert-type scale (1 = strongly disagree – 7 = strongly agree). The satisfaction items had low internal consistency (Cronbach’s α = .33), indicating that they are not measuring the same underlying trait. Therefore, the satisfaction items were treated as separate items and not as a scale.
669
IN FOCUS
Satisfaction with Health Care among African American Women Living with HIV/AIDS
Participant characteristics
• • • • •
Medication use (ART) Self-rated health Substance use Undetectable viral load CD4 count
Satisfaction with health care
• • •
Communication with health care provider
• • •
Medical services Social support services HIV prevention services
Condom use HIV transmission Sex
Figure 2. Testable model. ART = antiviral therapy.
Data Analysis Descriptive statistics and frequency counts were used to characterize the study sample, and hypotheses were tested using methods appropriate to the level of the variables. Chi-squared tests were utilized for the bivariate analyses to compare proportions and t tests (or Mann-Whitney U tests) were used to compare means. Pearson’s (or Spearman’s depending on the data distribution) correlations were used to determine the relationship between the study variables. Regression analyses were used to determine the relationships between the predictor variables and the outcome variables. All statistical analyses were conducted using SPSS statistical software version 16. Hypothesis 1
To address and test Hypothesis #1, bivariate analyses between the predictor variables and the outcome variables were conducted using MannWhitney U tests for dichotomous predictor variables (e.g., detectable viral load and current drug therapy use) and correlation analysis for continuous variables (e.g., CD4 count).
670
count, current medication use, injection drug use, self-reported health status, viral load) and the outcome variables (i.e., satisfaction with medical services, social support services, and HIV prevention services). Hypothesis 2
To address and test Hypothesis #2, bivariate relationships were examined between the communication with HCP scores and the satisfaction with health care scores using correlation analysis. Linear regression procedures were utilized to determine the relationship between the communication with HCP measures (i.e., communication about condom use, HIV transmission and sex) and the outcome variables (i.e., satisfaction with medical services, social support services, and HIV prevention services).
Results Sample Description
Following these bivariate comparisons, a regression model (De Vincenzi, 1994) was constructed using the predictor variables and the satisfaction with health care items as the outcome variables. Linear regression was used to determine the relationship between the predictor variables (i.e., CD4
Participants were 157 African American women from the Protect and Respect study. Characteristics of study participants are displayed in Table 1. The mean age of the participants was 39.85 years (SD = 8.72). Approximately one half of the sample reported being single. More than 73% of participants reported an annual income of less than $10,000 and 78% were unemployed. Thirty-five percent had obtained a high school diploma. The
JOGNN, 43, 664-676; 2014. DOI: 10.1111/1552-6909.12491
http://jognn.awhonn.org
IN FOCUS
Baker, J. L., Rodgers, C. R. R., Davis, Z. M., Gracely, E., and Bowleg, L.
majority of participants (83%) identified as heterosexual, and 82% reported heterosexual transmission as their primary exposure category.
Among racial/ethnic minority women living with HIV, health care disparities lend to dissatisfaction with health care, which can result in treatment underutilization and discontinuity of care.
Description of Study Variables Participant Health Characteristics Several psychosocial and health variables were considered potential predictors of satisfaction with health care. These variables included participant self-reported health status, CD4 count, viral load, injection drug use, and ART use. Descriptive statistics for these variables are reported in Table 2. As shown in Table 2, injection drug use among the sample was relatively low, with only 26% of participants reporting drug use in the past 30 days. With regards to HIV health status, 65% of participants reported having a detectable viral load. The mean CD4 count was 459 (SD = 333). Considering that a CD4 count of less than 200 signifies an AIDS diagnosis (Wu et al., 2000), the mean CD4 count of the participants indicated that a majority of the women were in good health; more specifically, the AIDS virus was not progressing and their immune systems were functioning well. Lastly, almost one half of the participants (49%) reported current medication therapy use at baseline.
Satisfaction with Health Care Overall, the women reported satisfaction with their health care based on the three satisfaction variables. Eighty-eight percent of the participants (n = 140) reported being satisfied with medical services. A majority of participants (65%, n = 101) also reported satisfaction with HIV prevention services; 62% (n = 95) reported satisfaction with social support services.
Bivariate Correlations Significant bivariate correlations among the study variables are presented in Tables 3 and 4. Selfreported health status and a detectable viral load were significantly correlated with satisfaction with HIV prevention services (p < .05) and satisfaction with medical services (p < .05). Also, communication about sex and communication about condom use were significantly correlated with satisfaction with social support services (p < .05).
Regression Analyses Communication with Health Care Providers about Risk Transmission Overall, participants reported some degree of communication with their HCPs based on the three communication variables. The majority of the women (63%, n = 96) reported discussing sex with their HCPs and discussing condom use with their HCPs (63%, n = 95). A majority of the sample (66%, n = 101) also reported communicating with their HCPs about transmitting HIV to others.
To test Hypothesis 1, regression models were used to assess whether participant characteristics had an effect on satisfaction with health care. A separate regression model was used to assess the effect of each participant characteristic on each of the three satisfaction variables (see columns 1–3 in Table 5). Table 5 shows the effect of the dependent variables (satisfaction with health care items) regressed on the independent variables (participant characteristics). None of the participant characteristics significantly predicted participant
Table 2: Summary of Baseline Means of Participant Characteristics (N = 157) Variable
n (%)
Mean
SD
Median
Minimum
Maximum
Viral load detectable – yes
82(65.1) 459.36
332.63
400.0
0
2,000
3.12
1.14
3.0
1
5
How long at this clinic (years)
12.00
82.56
4
0
998
First learned of HIV + status (months)
5.74
3.33
6
0.3
240
Most recent CD4 count Current medication use- yes
77(63.1)
Self-reported health status Illegal drug use in past 30 days – yes
JOGNN 2014; Vol. 43, Issue 5
38(25.9)
671
IN FOCUS
Satisfaction with Health Care among African American Women Living with HIV/AIDS
Table 3: Summary of Spearman Correlations between Numeric Predictor Variables and Satisfaction Scales Satisfaction with
Medical Services
HIV Prevention Services
−.15
.06
Self-reported health status
.16∗
−.03
.17∗
Communication with HCP about sex
.05
−.21 ∗∗
−.06
−.07
−.15
Most recent CD4 count
Communication with HCP about HIV transmission Communication with HCP about condom use
−.02
Social Support Services
−.17 −.06
∗∗
−.24
−.03
Note. CD4 count = CD4 + T cells; HCP = health care provider. ∗ Correlation significant at p < .05 (two-tailed). ∗∗ Correlation significant at p < .01 (two-tailed)
Contrary to existing literature, African American women living with HIV can be satisfied with their health care.
satisfaction with medical services (p > .05, see column 1 in Table 5), satisfaction with social support services (p > .05, see column 2 in Table 5), or satisfaction with HIV prevention services (p > .05, see column 3 in Table 5). Thus, Hypothesis 1 was not supported. To test Hypothesis 2, regression models were generated to examine the effect of the communication variables on each satisfaction variable. A separate regression model was used to assess the effect of each communication variable on each of the three satisfaction variables (see columns 1–3 in Table 6). Table 6 shows the effect of the dependent variables (satisfaction with health care) regressed on the independent variables (communication with HCPs). None of the communication variables significantly predicted participant satisfaction with medical services (p > .05, see column 1 in Table 6), satisfaction with social support services (p > .05, see column 2 in Table 6), or satisfaction with HIV prevention services (p > .05, see column 3 in Table 6). Therefore, this hypothesis is not supported.
Discussion Although AAWLWH/A account for a sizable majority of people in need of health care services, and research documents that PLWH/A who are satisfied with their HIV care experience more optimal health outcomes, there is a significant gap in the literature on their health care experiences including satisfaction with health care. Thus, a key goal of this study was to report on rates of satisfaction with health care among AAWLWH/A and provide new information about what factors influence satisfaction with health care. We found that the majority of participants reported being satisfied with health care, specifically their medical, social support, and HIV prevention services. However, we did not observe any evidence that satisfaction with these aspects of health care were significantly correlated with reported health status, medication therapy use, illegal drug use, CD4 count or viral load.
Results from this study demonstrated that the participants were satisfied with health care in all three aspects of care analyzed: medical services, social support services, and HIV prevention services. The women also reported high ratings of communication with their HCPs about condom use,
The nonsignificant findings in the relationship between satisfaction with health care and psychosocial and health characteristics are not consistent with previous literature, which suggests a relationship between psychosocial and health characteristics and satisfaction with health care (Burke et al., 2003; Burke-Miller et al., 2006; Sullivan et al., 2000). It is possible that the analyzed variables (e.g., CD4 count, viral load, medication therapy use) may not be the only factors that influence
JOGNN, 43, 664-676; 2014. DOI: 10.1111/1552-6909.12491
http://jognn.awhonn.org
Summary of Results
672
sex, and transmission of HIV. However, we did not find any significant relationships between psychosocial/health characteristics and satisfaction with health care.
IN FOCUS
Baker, J. L., Rodgers, C. R. R., Davis, Z. M., Gracely, E., and Bowleg, L.
Table 4: Summary of Means ± SD for Satisfaction Scales as a Function of Categorical Variables Satisfaction with
n
Medical Services
Social Support Services
HIV Prevention Services
73
6.49 ± 1.2∗
4.36 ± 2.3∗
6.27 ± 1.5
HS/Graduate Equivalency Diploma
53
∗
4.36 ± 2.4
4.91 ± 2.1
6.16 ± 1.1
More than HS
31
6.55 ± 0.9
5.78 ± 1.6
6.59 ± 0.63
$10,000 Or Less
115
6.39 ± 1.2
4.84 ± 2.3
6.3 ± 1.2
$10,001 to $20,000
22
6.57 ± 0.7
5.14 ± 1.8
6.48 ± 0.9
$20,001 to $40,000
13
5.77 ± 1.8
4.84 ± 1.7
5.77 ± 1.7
Education Less than high school (HS)
Income
∗
1.0 ± 0
6.25 ± 1.5
$40,001 to $60,000
4
5.55 ± 2.4
$ 60,001 to $80,000
2
7.0 ± 0
6.5 ± 0.7
6.5 ± 0.7
Over $80,000
1
7.0 ± 0
1.0 ± 0∗
7.0 ± 0
Current Medication -yes
77
6.39 ± 0.9
4.56 ± 2.3
6.42 ± 1.1
Drug use in last 30 days -yes
38
6.31 ± 1.34
4.84 ± 2.2
6.36 ± 1.2
82
6.37 ± 1.3
4.89 ± 2.2
6.32 ± 1.1
130
6.3 ± 1.3
4.63 ± 2.2a
6.3 ± 1.2
Health status
Viral Load Detectable – yes
a
Sexual orientation Heterosexual – yes
Note. ∗ Correlation significant at p < .05 (two-tailed).
patient satisfaction with health care among this sample. Although it was hypothesized that communication with HCPs would be significantly correlated with satisfaction with health care, results indicate there was no statistically significant relationship. These findings are also not consistent with previous studies that suggest relationships between communication with HCPs and satisfaction with health care among PLWH/A (Apollo et al., 2006; Marelich & Murphy, 2003; Wilson & Kaplan, 2000). However, ratings of communication with HCPs were higher than expected based on the current literature base. Discrimination continues to be a reality for PLWH/A (Bogart et al., 2010) in ways that may prevent some, particularly AAWLWH/A who encounter judgment and stigma (Black & Miles, 2002), from discussing sensitive topics with their HCPs. Thus, this finding is encouraging as it demonstrates that AAWLWH/A can feel comfortable discussing very personal and stigmatized
JOGNN 2014; Vol. 43, Issue 5
issues including sex and condom use with their HCPs.
Limitations Although this study provides meaningful information related to satisfaction with health care among AALWH/A, there are limitations. One limitation of this study was the reliability of the measures. A major focus of this study was to demonstrate the relationships between participant health characteristics and satisfaction with health care. Because this was a secondary data analysis, a valid measure of patient satisfaction could not be used for analyses. Further, the satisfaction items were not reliable as a scale. Instead, three satisfaction items were used separately to assess different aspects of satisfaction with health care. None of the psychosocial and health characteristics was significantly associated with satisfaction with health care. Thus, reliability of self-reports of the measures should be considered.
673
IN FOCUS
Satisfaction with Health Care among African American Women Living with HIV/AIDS
Table 5: Regression Analysis for Participant Characteristics as Predictors of Satisfaction with Healthcare (N = 157) Satisfaction with
Satisfaction with Social
Satisfaction with HIV
Variable
Medical Services
Support Services
Prevention Services
Self-Report
β SE F p
β SE F p
β SE F p
Health Status
.09 0.1 1.0.31
−.06 0.2 .15.69
.13 .09 2.1 .15
Current Medication Use
.54 .22 .06 .81
−.25 0.4 .34 .56
−.15 .22 .45 .50
Detectable Viral Load
0.2 0.2 0.4 .52
−0.2 0.4 15 0.7
−.12 .24.22 .63
CD4 count
.06 .21 .39.53
−.001 .001 1.9.17
.09 .02.84.35
Drug Use
−.07 .24 .09 .77
−.14 .44 .09 .76
−.12 23 .32 .57
R2
0.004
0.020
0.009
Note. Each column provides results from the set of regressions using one of the three satisfaction variables as the dependent variable; SE = standard error; F = f distribution; CD4 count = CD4 + T cells.
This study was a secondary data analysis, which limited the extent to which other variables that may be associated with satisfaction with health care, such as time spent with HCP, discrimination from HCPs, and trust (Benkert et al., 2006; O’Malley et al., 2004; Stone, 2004; Warde, 2001), could be analyzed. The quantitative design of this study is another possible limitation. A mixed methods study that involved the use of qualitative methods may have been a more appropriate design for investigating new concepts particularly associations between health characteristics and satisfaction with health care. Lastly, study results could be confounded by the relationships between the HCPs and the study participants, as most of the women were patients of the HIV care clinic where the Protect and Respect program was located for an average of 13 years (see Table 1). Patient–provider relationships have been found to be more important than other com-
ponents of care in its association with overall satisfaction (Dang et al., 2012). Thus, it is possible that long-term relationships between the women and their HCPs may have inflated their ratings of satisfaction. Given that a majority of women did indicate satisfaction with the health care services provided, it would appear that the HCPs are providing a very high level standard of care to their patients. However, HCP characteristics, provider engagement, or specific clinic attributes were not available to be analyzed to assess their relationship to satisfaction with health care among the sample.
Conclusion Given the disparities of HIV infection among women, health care providers will continue to be sought after to provide HIV care to African American women. Because satisfaction with health care can influence the quality of care received, health
Table 6: Summary of Regression Analysis for Communication with Health Care Provider as Predictor of Satisfaction with Health Care (N = 157) Satisfaction with
Satisfaction with Social
Satisfaction with HIV
Variable
Medical Services
Support Services
Prevention Services
Discuss sex with provider
β SE F p
β SE F p
β SE F p
−.06 .047 1.6 .20
−.13 .09 2.3 .13
−.006 .05 .02 .90
Discuss risk of HIV transmission to others
−.07 .049 1.9 .17
−.06 .09 .39 .54
−.06 .05 1.6 .21
Subject of condoms use comes up in HIV care visits
−.08 .047 2.8 .09
−.15 .09 3.1 .08
−.03 .05 .47 .50
R2
.018
0.021
0.003
Note. Each column provides results from the set of regressions using one of the three satisfaction variables as the dependent variable; SE = standard error; F = f distribution.
674
JOGNN, 43, 664-676; 2014. DOI: 10.1111/1552-6909.12491
http://jognn.awhonn.org
IN FOCUS
Baker, J. L., Rodgers, C. R. R., Davis, Z. M., Gracely, E., and Bowleg, L.
outcomes, and adherence to provider recommendations among people living with HIV/AIDS, health care providers’ ability to elicit satisfaction from their patients is just as important as the health care services they provide. We contribute to the patient satisfaction literature by specifically focusing on satisfaction with health care among AAWLWH/A. However, the results may be reflective of care provided by the health care providers of the specific clinic they attended. To enhance the understanding of satisfaction among AAWLWH/A, additional research should utilize qualitative methods to explore how the context of the patient– provider relationship influences satisfaction with health care. Further analyses with a larger sample size, more reliable and validated measures, and a mixed methods approach may also improve assessments of satisfaction with health care among AAWLWH/A.
Burke-Miller, J., Cook, J., Cohen, M., Hessol, N., Wilson, T., Richardson, J., & Gange, S. (2006). Longitudinal relationships between use of highly active antiretroviral therapy and satisfaction with care among women living with HIV/AIDS. American Journal of Public Health, 96(6), 1044–1051. doi:10.2105/ajph.2005.061929 Buseh, A., Kelber, S., Stevens, T., Poedel, R., & Park, C. G. (2009). Interpersonal and attitudinal factors in healthcare utilization among HIV-infected urban African American men. Challenge, 5(2), 1– 18. Campbell, J., Ramsay, J., & Greene, J. (2001). Age, gender, socioeconomic, and ethnic differences in patients’ assessments of primary health care. Quality in Health Care, 10, 90–95. Cargill, V. A., & Stone, V. E. (2005). HIV/AIDS: A minority health issue. Medical Clinics of North America, 89(4), 895–912. Centers for Disease Control and Prevention. (2011). HIV surveillance among women. Atlanta, GA: Author. Retrieved from http://www. cdc.gov/hiv/topics/surveillance/resources/slides/women/index. htm Centers for Disease Control and Prevention. (2012a). CDC fact sheet: New HIV infections in the United States. Atlanta, GA: Author. Retrieved from http://www.cdc.gov/nchhstp/newsroom/ docs/2012/hiv-infections-2007--2010.pdf Centers for Disease Control and Prevention. (2012b). HIV in the United
Acknowledgement The authors thank Jennifer A. Faerber, PhD, Division of Rheumatology, Children’s Hospital of Philadelphia and Thomas R. Tenhave, PhD (deceased), former Director of the Center for Clinical Epidemiology and Biostatistics at the University of Pennsylvania.
States: The stages of care. Atlanta, GA: Author. Retrieved from http://www.cdc.gov/hiv/pdf/research_mmp_stagesofcare.pdf Centers for Disease Control and Prevention. (2013). Fact sheet: HIV among women. Atlanta, GA: Author. Retrieved from http://www.cdc.gov/hiv/topics/women/pdf/women.pdf Cohen, J. (2008). Challenges and successes in reducing health disparities: Workshop summary. Washington, DC: National Academies Press. Copeland, V. C., Scholle, S. H., & Binko, J. A. (2003). Patient satisfaction. Journal of Health & Social Policy, 17(2), 35–48.
REFERENCES Abel, E., & Painter, L. (2003). Factors that influence adherence to HIV
Dang, B., Westbrook, R., Rodriguez-Barradas, M., & Giordano, T. (2012). Identifying drivers of overall satisfaction in patients re-
medications: Perceptions of women and health care providers.
ceiving HIV primary care: A cross-sectional study. PLOS One,
Journal of the Association of Nurses in AIDS Care, 14(4), 61–69.
7(8), 1–7.
Apollo, A., Golub, S. A., Wainberg, M. L., & Indyk, D. (2006). Patient-
De Vincenzi, I. (1994). A longitudinal study of human immunodeficiency
provider relationships, HIV, and adherence: Requisites for a part-
virus transmission by heterosexual partners. New England Jour-
nership. Social Work in Health Care, 42(3/4), 209–224.
nal of Medicine, 331(6), 341–346.
Benkert, R., Peters, R., Clark, R., & Keeves-Foster, K. (2006). Effects of
El-Bassel, N., Caldeira, N., Ruglass, L., & Gilbert, L. (2009). Addressing
perceived racism, cultural mistrust, and trust in providers on sat-
the unique needs of African American women in HIV prevention.
isfaction with care. Journal of the National Medical Association, 98(9), 1532–1540. Black, B., & Miles, M. (2002). Calculating the risks and benefits of disclosure in African American women who have HIV. Journal of Obstetric, Gynecologic, & Neonatal Nursing, 31(6), 688–697.
American Journal of Public Health, 99, 996–1001. Emlet, C. (2007). Experiences of stigma in older adults living with HIV/AIDS: A mixed methods analysis. AIDS Patient Care & STDs, 21(10), 740–752. Gerbert, B., Love, C., Caspers, N., Linkins, K., & Burack, J. H.
Bodenlos, J., Grothe, K., Whitehead, D., Konkle-Parker, D., Jones, G.,
(1999). “Making all the difference in the world”: How physi-
& Brantley, P. (2007). Attitudes toward health care providers and
cians can help HIV-seropositive patients become more involved
appointment attendance in HIV/AIDS Patients. Journal of the
in their healthcare. AIDS Patient Care & STDs, 13(1), 29–
Association of Nurses in AIDS Care, 18(3), 65–73.
39.
Bogart, L., Bird, S., Walt, L., Delahanty, D., & Figler, J. (2004). Associa-
Hall, J. A., & Dornan, M. C. (1988). Meta-analysis of satisfaction with
tion of stereotypes about physicians to health care satisfaction,
medical care: Description of research domain and analysis of
help-seeking behavior, and adherence to treatment. Social Sci-
overall satisfaction levels. Social Science & Medicine, 27, 637–
ence & Medicine, 58, 1049–1058.
644.
Bogart, L., Wagner, G., Galavan, F., & Klein, D. (2010). Longitudinal re-
Heckman, B. D., Catz, S. L., Heckman, T. G., Miller, J. G., & Kalichman,
lationships between antiretroviral treatment adherence and dis-
S. C. (2004). Adherence to antiretroviral therapy in rural persons
crimination due to HIV-serostatus, race, and sexual orientation
living with HIV disease in the United States. AIDS Care, 16(2),
among African-American men with HIV. Annals of Behavioral Medicine, 40(2), 184–190. Burke, J., Cook, J., Cohen, H., Wilson, T., Anastos, K., Young, M., & Gange, S. (2003). Dissatisfaction with medical care among women with HIV: Dimensions and associated factors. AIDS Care, 15(4), 451–462. doi:10.1080/0954012031000134692
JOGNN 2014; Vol. 43, Issue 5
219–230. doi:10.1080/09540120410001641066 Institute of Medicine. (2003). Unequal treatment: Confronting racial and ethnic disparities in health care. Washington, DC: National Academies Press. Johnson, M., Fisher, D., Montoya, I., Booth, R., Rhodes, F., Andersen, M., & Williams, M. (2000). Reliability and validity of
675
IN FOCUS
Satisfaction with Health Care among African American Women Living with HIV/AIDS
not-in-treatment drug users’ follow-up self-reports. AIDS and Behavior, 4, 373–380.
STDs, 16(1), 43–50.
Kaiser Family Foundation. (2013). Women and HIV fact sheet. Retrieved
Schneider, J., Kaplan, S. H., Greenfield, S., Li, W., & Wilson, I. B. (2004).
from http://kff.org/hivaids/fact-sheet/women-and-hivaids-in-the-
Better physician-patient relationships are associated with higher
united-states/
reported adherence to antiretroviral therapy in patients with HIV
Kinsler, J., Cunningham, C., Mohanty, S., & Wong, M. (2008). Are medical records a more reliable and valid source of CD4 count, viral
infection. Journal of General Internal Medicine, 19(11), 1096– 1103. doi:JGI30418
load, and outpatient visit data than self-reports? A comparison
Simoni, J., Kurth, A., Pearson, C., Pantalone, D., Merrill, J., & Frick, P.
with electronic medical records as the gold standard. Health
(2006). Self-report measures of antiretroviral therapy adherence:
Services and Outcomes Research Methodology, 8, 187–192.
A review with recommendations for HIV research and clinical
Kleeberger, C., Buechner, J., & Palella, F. (2004). Changes in adherence to highly active antiretroviral therapy medications in the Multi-center AIDS Cohort Study. AIDS, 18, 683–688. LaVeist, T., & Carroll, T. (2002). Race of physician and satisfaction with care among African American patients. Journal of the National Medical Association, 94, 937–943.
management. AIDS Behavior, 10, 227–245. Sohler, N., Li, X., & Cunningham, C. (2007). Perceived discrimination among severely disadvantaged people with HIV infection. Public Health Reports, 122(3), 347–355. Stone, V. E. (2004). Optimizing the care of minority patients with HIV/AIDS. Clinical Infectious Diseases, 38(3), 400–404.
LaVeist, T., & Nuru-Jeter, A. (2002). Is doctor-patient race concordance
Sullivan, L. M., Stein, M. D., Savetsky, J. B., & Samet, J. H. (2000).
associated with greater satisfaction with care? Journal of Health
The doctor-patient relationship and HIV-infected patients’ satis-
and Social Behavior, 43, 296–306.
faction with primary care physicians. Journal of General Internal
Marelich, W. D., & Murphy, D. A. (2003). Effects of empowerment among HIV-positive women on the patient-provider relationship. AIDS Care, 15(4), 475–481. Napper, L., Fisher, D., Johnson, M., & Wood, M. (2010). The reliability and validity of drug users’ self reports of amphetamine use among primarily heroin and cocaine users. Addictive Behaviors, 35, 350–354.
Medicine, 15(7), 462–469. U.S. Department of Health and Human Services. (2010). U.S. Department of Health and Human Services: National HIV/AIDS strategy.
Retrieved
from
http://aids.gov/federal-resources/
national-hiv-aids-strategy/nhas.pdf Warde, C. (2001). Time is of the essence. Journal of General Internal Medicine, 16, 712–713.
O’Malley, A., & Forrest, C. (2002). Beyond the examination room: Pri-
Wilson, I. B., & Kaplan, S. (2000). Physician-patient communication in
mary care performance and the patient-physician relationship
HIV disease: The importance of patient, physician, and visit char-
for low-income women. Journal of General Internal Medicine,
acteristics. Journal of Acquired Immune Deficiency Syndrome,
17(1), 66–74.
25(5), 417–425.
O’Malley, A., Sheppard, V., Schwartz, M., & Mandelblatt, J. (2004). The
Wong, M. D., Shapiro, M. F., Boscardin, W. J., & Ettner, S. L. (2002).
role of trust in use of preventive services among low-income
Contribution of major diseases to disparities in mortality. New
African-American women. Preventive Medicine, 38(6), 77– 85. Roberts, K. J. (2002). Physician-patient relationships, patient satisfaction, and antiretroviral medication adherence among HIV in-
676
fected adults attending a public health clinic. AIDS Patient Care
JOGNN, 43, 664-676; 2014. DOI: 10.1111/1552-6909.12491
England Journal of Medicine, 347(20), 1585–1592. Wu, A. W., Gifford, A., Asch, S., Cohn, S. E., Bozzette, S. A., & Yurk, R. (2000). Quality-of-care indicators for HIV/AIDS. Disease Management & Health Outcomes, 7(6), 315–330.
http://jognn.awhonn.org